|
I recall seeing some (I think German?) amateur's website with a design for a DIY analyser to measure phase noise ... I don't suppose anyone else recalls where it might have been? -- Best regards,
Robin Szemeti
Redpoint Consulting Limited
E: robin@...
T: +44 (0) 1299 405028
M: +44 (0) 7971 883371
CONFIDENTIALITY NOTICE
The information contained in this e-mail is intended only for the
confidential use of the above named recipient. If you are not the
intended recipient or person responsible for delivering it to the
intended recipient, you have received this communication in error
and must not distribute or copy it.
Please accept the sender's apologies, notify the sender immediately
by return e-mail and delete this communication.
Thank you.
-- Robin Szemeti - G1YFG
|
|
|
|
|
|
Robin
My DIY method is as per section 3.1 “Down converter and IF filter with a spectrum analyzer” in https://www.npl.co.uk/special-pages/guides/gpg68_noise.aspx?ext
It is still work in progress but I can measure a 432MHz carrier down to -140dBc/Hz at 50kHz offset using a Signal Hound SA44B spectrum analyser, the 10.7MHz filter improving the measurement range by about 30dB. As you would expect the improvement is a lot less close in and not quantitatively useful below 5kHz, in the current form.
Also see http://www.sm5bsz.com/index.htm
73s
Richard
GD8EXI
On 30/11/2022, 00:37, "Robin Szemeti - G1YFG via groups.io" <robin@...> wrote:
toggle quoted message
Show quoted text
I recall seeing some (I think German?) amateur's website with a design for a DIY analyser to measure phase noise ... I don't suppose anyone else recalls where it might have been?
|
|
|
Much appreciate, I shall peruse the information. That Xilinx Spartan based project looked neat, but I suspect the cost of the Spartan dev board is more than most radios!
On Wed, 30 Nov 2022 at 12:08, Richard GD8EXI < perwick@...> wrote:
Robin
My DIY method is as per section 3.1 “Down converter and IF filter with a spectrum analyzer” in https://www.npl.co.uk/special-pages/guides/gpg68_noise.aspx?ext
It is still work in progress but I can measure a 432MHz carrier down to -140dBc/Hz at 50kHz offset using a Signal Hound SA44B spectrum analyser, the 10.7MHz filter improving the measurement range by about 30dB. As you would expect the improvement is a lot less close in and not quantitatively useful below 5kHz, in the current form.
Also see http://www.sm5bsz.com/index.htm
73s
Richard
GD8EXI
On 30/11/2022, 00:37, "Robin Szemeti - G1YFG via groups.io" <robin@...> wrote:
I recall seeing some (I think German?) amateur's website with a design for a DIY analyser to measure phase noise ... I don't suppose anyone else recalls where it might have been?
--
Robin Szemeti - G1YFG
|
|
|
Richard EXI,
How wide a IF filter do you use? Do you find a really wide one like 200KHz and take a reading in one go ... or do you use a fairly narrow one like 25KHz and then sweep it across with the LO signal?
On Wed, 30 Nov 2022 at 12:08, Richard GD8EXI < perwick@...> wrote:
Robin
My DIY method is as per section 3.1 “Down converter and IF filter with a spectrum analyzer” in https://www.npl.co.uk/special-pages/guides/gpg68_noise.aspx?ext
It is still work in progress but I can measure a 432MHz carrier down to -140dBc/Hz at 50kHz offset using a Signal Hound SA44B spectrum analyser, the 10.7MHz filter improving the measurement range by about 30dB. As you would expect the improvement is a lot less close in and not quantitatively useful below 5kHz, in the current form.
Also see http://www.sm5bsz.com/index.htm
73s
Richard
GD8EXI
On 30/11/2022, 00:37, "Robin Szemeti - G1YFG via groups.io" <robin@...> wrote:
I recall seeing some (I think German?) amateur's website with a design for a DIY analyser to measure phase noise ... I don't suppose anyone else recalls where it might have been?
--
Robin Szemeti - G1YFG
|
|
|
Robin a good question, as I said earlier this is still work in progress.
Initially I tried using ceramic 10.7MHz filters designed for use in FM radios with pass-bands of 280 and 12.5kHz from the junk box. However they had significant ripple in both the pass and stop-bands with poor shape factors.
I am now trying the opposite extreme using a 9.0MHz, 2.4KHz BW 6 element crystal ladder filter, photo attached. Note the ferrite transformers at both the input and output to try to get 50ohm. This has a better shape factor and stopband, with less ripple, traces attached. However a lot of effort in piecing together a wideband trace if needed.
After answering your note I was going to ask the group a similar question. What do others find works best re LO and filter on a tight budget? A notch filter with a very flat passband should be better?
The current system has the following parts
A HB crystal TXO as LO.
(Also tried using a Leo Bodner GPS source but not quite good enough for this application, however much better than many of the commercial rigs under test!)
A Mini-Circuits ZFM-11 mixer.
Spectrum Communications 9MHz SSB filter with extra matching transformers.
Numerous attenuators and PGA-103+ LNA to set the correct levels.
A Signal Hound SA44B spectrum analyzer.
73s
Richard
GD8EXI
On 04/12/2022, 01:30, "Robin Szemeti - G1YFG via groups.io" <robin@...> wrote:
toggle quoted message
Show quoted text
Richard EXI,
How wide a IF filter do you use? Do you find a really wide one like 200KHz and take a reading in one go ... or do you use a fairly narrow one like 25KHz and then sweep it across with the LO signal?
On Wed, 30 Nov 2022 at 12:08, Richard GD8EXI <perwick@...> wrote:
Robin
My DIY method is as per section 3.1 “Down converter and IF filter with a spectrum analyzer” in https://www.npl.co.uk/special-pages/guides/gpg68_noise.aspx?ext
It is still work in progress but I can measure a 432MHz carrier down to -140dBc/Hz at 50kHz offset using a Signal Hound SA44B spectrum analyser, the 10.7MHz filter improving the measurement range by about 30dB. As you would expect the improvement is a lot less close in and not quantitatively useful below 5kHz, in the current form.
Also see http://www.sm5bsz.com/index.htm
73s
Richard
GD8EXI
On 30/11/2022, 00:37, "Robin Szemeti - G1YFG via groups.io <http://groups.io> " <robin@... <http://robin@...> > wrote:
I recall seeing some (I think German?) amateur's website with a design for a DIY analyser to measure phase noise ... I don't suppose anyone else recalls where it might have been?
|
|
|
The key I think is building a good diplexer on the mixer RF out, and possibly the RF in to present a good 50R match not just at the desired frequency but at the image and LO too ... a "bridged T" is a good choice. I was planning to use my Marconi 2041 as the LO, as it has (supposedly) good phase noise performance.
I shall think more on this,I suspect it is going to lead me down the rabbithole of connecting up my SA and sig gen with HPIB, then I could automate taking measurements and sweeping the SA and SigGen across in sync ... fortunately, it appears USB to HPIB interfaces are suitably expensive, so that should put me off for a while.
On Sun, 4 Dec 2022 at 15:16, Richard GD8EXI < perwick@...> wrote:
Robin a good question, as I said earlier this is still work in progress.
Initially I tried using ceramic 10.7MHz filters designed for use in FM radios with pass-bands of 280 and 12.5kHz from the junk box. However they had significant ripple in both the pass and stop-bands with poor shape factors.
I am now trying the opposite extreme using a 9.0MHz, 2.4KHz BW 6 element crystal ladder filter, photo attached. Note the ferrite transformers at both the input and output to try to get 50ohm. This has a better shape factor and stopband, with less ripple, traces attached. However a lot of effort in piecing together a wideband trace if needed.
After answering your note I was going to ask the group a similar question. What do others find works best re LO and filter on a tight budget? A notch filter with a very flat passband should be better?
The current system has the following parts
A HB crystal TXO as LO.
(Also tried using a Leo Bodner GPS source but not quite good enough for this application, however much better than many of the commercial rigs under test!)
A Mini-Circuits ZFM-11 mixer.
Spectrum Communications 9MHz SSB filter with extra matching transformers.
Numerous attenuators and PGA-103+ LNA to set the correct levels.
A Signal Hound SA44B spectrum analyzer.
73s
Richard
GD8EXI
On 04/12/2022, 01:30, "Robin Szemeti - G1YFG via groups.io" <robin@...> wrote:
Richard EXI,
How wide a IF filter do you use? Do you find a really wide one like 200KHz and take a reading in one go ... or do you use a fairly narrow one like 25KHz and then sweep it across with the LO signal?
On Wed, 30 Nov 2022 at 12:08, Richard GD8EXI <perwick@...> wrote:
Robin
My DIY method is as per section 3.1 “Down converter and IF filter with a spectrum analyzer” in https://www.npl.co.uk/special-pages/guides/gpg68_noise.aspx?ext
It is still work in progress but I can measure a 432MHz carrier down to -140dBc/Hz at 50kHz offset using a Signal Hound SA44B spectrum analyser, the 10.7MHz filter improving the measurement range by about 30dB. As you would expect the improvement is a lot less close in and not quantitatively useful below 5kHz, in the current form.
Also see http://www.sm5bsz.com/index.htm
73s
Richard
GD8EXI
On 30/11/2022, 00:37, "Robin Szemeti - G1YFG via groups.io <http://groups.io> " <robin@... <http://robin@...> > wrote:
I recall seeing some (I think German?) amateur's website with a design for a DIY analyser to measure phase noise ... I don't suppose anyone else recalls where it might have been?
--
Robin Szemeti - G1YFG
|
|
|
May I recommend the arduino based AR488 interfaces from USB to HPIB (google will find the code and there are a few designs around including one by Adrian Godwin that works on the back of a centronics plug). They cost next to nothing to build (the centronics connector seems the most expensive part) and they behave like the prologix. In many ways I'd rather have Ethernet to GBIP but that does still seem expensive, maybe there is a project looming for me....
I've played with one to control my HP8903A audio analyser and SA because it would break free from doing sweeps with an assumed pen recorder.
Alan
The key I think is building a good diplexer on the mixer RF out, and possibly the RF in to present a good 50R match not just at the desired frequency but at the image and LO too ... a "bridged T" is a good choice. I was planning to use my Marconi 2041 as the LO, as it has (supposedly) good phase noise performance.
I shall think more on this,I suspect it is going to lead me down the rabbithole of connecting up my SA and sig gen with HPIB, then I could automate taking measurements and sweeping the SA and SigGen across in sync ... fortunately, it appears USB to HPIB interfaces are suitably expensive, so that should put me off for a while.
On Sun, 4 Dec 2022 at 15:16, Richard GD8EXI < perwick@...> wrote:
Robin a good question, as I said earlier this is still work in progress.
Initially I tried using ceramic 10.7MHz filters designed for use in FM radios with pass-bands of 280 and 12.5kHz from the junk box. However they had significant ripple in both the pass and stop-bands with poor shape factors.
I am now trying the opposite extreme using a 9.0MHz, 2.4KHz BW 6 element crystal ladder filter, photo attached. Note the ferrite transformers at both the input and output to try to get 50ohm. This has a better shape factor and stopband, with less ripple, traces attached. However a lot of effort in piecing together a wideband trace if needed.
After answering your note I was going to ask the group a similar question. What do others find works best re LO and filter on a tight budget? A notch filter with a very flat passband should be better?
The current system has the following parts
A HB crystal TXO as LO.
(Also tried using a Leo Bodner GPS source but not quite good enough for this application, however much better than many of the commercial rigs under test!)
A Mini-Circuits ZFM-11 mixer.
Spectrum Communications 9MHz SSB filter with extra matching transformers.
Numerous attenuators and PGA-103+ LNA to set the correct levels.
A Signal Hound SA44B spectrum analyzer.
73s
Richard
GD8EXI
On 04/12/2022, 01:30, "Robin Szemeti - G1YFG via groups.io" <robin@...> wrote:
Richard EXI,
How wide a IF filter do you use? Do you find a really wide one like 200KHz and take a reading in one go ... or do you use a fairly narrow one like 25KHz and then sweep it across with the LO signal?
On Wed, 30 Nov 2022 at 12:08, Richard GD8EXI <perwick@...> wrote:
Robin
My DIY method is as per section 3.1 “Down converter and IF filter with a spectrum analyzer” in https://www.npl.co.uk/special-pages/guides/gpg68_noise.aspx?ext
It is still work in progress but I can measure a 432MHz carrier down to -140dBc/Hz at 50kHz offset using a Signal Hound SA44B spectrum analyser, the 10.7MHz filter improving the measurement range by about 30dB. As you would expect the improvement is a lot less close in and not quantitatively useful below 5kHz, in the current form.
Also see http://www.sm5bsz.com/index.htm
73s
Richard
GD8EXI
On 30/11/2022, 00:37, "Robin Szemeti - G1YFG via groups.io <http://groups.io> " <robin@... <http://robin@...> > wrote:
I recall seeing some (I think German?) amateur's website with a design for a DIY analyser to measure phase noise ... I don't suppose anyone else recalls where it might have been?
--
Robin Szemeti - G1YFG
|
|
|
That sounds perfect, I love an Arduino project!
Now I just have to remember where I recently saw some cheap GPIB leads!
May I recommend the arduino based AR488 interfaces from USB to HPIB (google will find the code and there are a few designs around including one by Adrian Godwin that works on the back of a centronics plug). They cost next to nothing to build (the centronics connector seems the most expensive part) and they behave like the prologix. In many ways I'd rather have Ethernet to GBIP but that does still seem expensive, maybe there is a project looming for me....
I've played with one to control my HP8903A audio analyser and SA because it would break free from doing sweeps with an assumed pen recorder.
Alan The key I think is building a good diplexer on the mixer RF out, and possibly the RF in to present a good 50R match not just at the desired frequency but at the image and LO too ... a "bridged T" is a good choice. I was planning to use my Marconi 2041 as the LO, as it has (supposedly) good phase noise performance.
I shall think more on this,I suspect it is going to lead me down the rabbithole of connecting up my SA and sig gen with HPIB, then I could automate taking measurements and sweeping the SA and SigGen across in sync ... fortunately, it appears USB to HPIB interfaces are suitably expensive, so that should put me off for a while.
On Sun, 4 Dec 2022 at 15:16, Richard GD8EXI < perwick@...> wrote:
Robin a good question, as I said earlier this is still work in progress.
Initially I tried using ceramic 10.7MHz filters designed for use in FM radios with pass-bands of 280 and 12.5kHz from the junk box. However they had significant ripple in both the pass and stop-bands with poor shape factors.
I am now trying the opposite extreme using a 9.0MHz, 2.4KHz BW 6 element crystal ladder filter, photo attached. Note the ferrite transformers at both the input and output to try to get 50ohm. This has a better shape factor and stopband, with less ripple, traces attached. However a lot of effort in piecing together a wideband trace if needed.
After answering your note I was going to ask the group a similar question. What do others find works best re LO and filter on a tight budget? A notch filter with a very flat passband should be better?
The current system has the following parts
A HB crystal TXO as LO.
(Also tried using a Leo Bodner GPS source but not quite good enough for this application, however much better than many of the commercial rigs under test!)
A Mini-Circuits ZFM-11 mixer.
Spectrum Communications 9MHz SSB filter with extra matching transformers.
Numerous attenuators and PGA-103+ LNA to set the correct levels.
A Signal Hound SA44B spectrum analyzer.
73s
Richard
GD8EXI
On 04/12/2022, 01:30, "Robin Szemeti - G1YFG via groups.io" <robin@...> wrote:
Richard EXI,
How wide a IF filter do you use? Do you find a really wide one like 200KHz and take a reading in one go ... or do you use a fairly narrow one like 25KHz and then sweep it across with the LO signal?
On Wed, 30 Nov 2022 at 12:08, Richard GD8EXI <perwick@...> wrote:
Robin
My DIY method is as per section 3.1 “Down converter and IF filter with a spectrum analyzer” in https://www.npl.co.uk/special-pages/guides/gpg68_noise.aspx?ext
It is still work in progress but I can measure a 432MHz carrier down to -140dBc/Hz at 50kHz offset using a Signal Hound SA44B spectrum analyser, the 10.7MHz filter improving the measurement range by about 30dB. As you would expect the improvement is a lot less close in and not quantitatively useful below 5kHz, in the current form.
Also see http://www.sm5bsz.com/index.htm
73s
Richard
GD8EXI
On 30/11/2022, 00:37, "Robin Szemeti - G1YFG via groups.io <http://groups.io> " <robin@... <http://robin@...> > wrote:
I recall seeing some (I think German?) amateur's website with a design for a DIY analyser to measure phase noise ... I don't suppose anyone else recalls where it might have been?
--
Robin Szemeti - G1YFG
--
Robin Szemeti - G1YFG
|
|
|
The nice thing about building the interface on the back of the connector is you don't need a GPIB lead, just plug the board into the back of the instrument (the board and plug are about the size of an HP HPIB plug). Then you need to hunt out cheap micro USB lead (which is not so hard!). Back in the day I had a rack of disks and an amazingly slow quarter inch tape drive running over HPIB - sadly all gone along with a load of HPIB leads which would have been useful now....
Alan
That sounds perfect, I love an Arduino project!
Now I just have to remember where I recently saw some cheap GPIB leads!
May I recommend the arduino based AR488 interfaces from USB to HPIB (google will find the code and there are a few designs around including one by Adrian Godwin that works on the back of a centronics plug). They cost next to nothing to build (the centronics connector seems the most expensive part) and they behave like the prologix. In many ways I'd rather have Ethernet to GBIP but that does still seem expensive, maybe there is a project looming for me....
I've played with one to control my HP8903A audio analyser and SA because it would break free from doing sweeps with an assumed pen recorder.
Alan The key I think is building a good diplexer on the mixer RF out, and possibly the RF in to present a good 50R match not just at the desired frequency but at the image and LO too ... a "bridged T" is a good choice. I was planning to use my Marconi 2041 as the LO, as it has (supposedly) good phase noise performance.
I shall think more on this,I suspect it is going to lead me down the rabbithole of connecting up my SA and sig gen with HPIB, then I could automate taking measurements and sweeping the SA and SigGen across in sync ... fortunately, it appears USB to HPIB interfaces are suitably expensive, so that should put me off for a while.
On Sun, 4 Dec 2022 at 15:16, Richard GD8EXI < perwick@...> wrote:
Robin a good question, as I said earlier this is still work in progress.
Initially I tried using ceramic 10.7MHz filters designed for use in FM radios with pass-bands of 280 and 12.5kHz from the junk box. However they had significant ripple in both the pass and stop-bands with poor shape factors.
I am now trying the opposite extreme using a 9.0MHz, 2.4KHz BW 6 element crystal ladder filter, photo attached. Note the ferrite transformers at both the input and output to try to get 50ohm. This has a better shape factor and stopband, with less ripple, traces attached. However a lot of effort in piecing together a wideband trace if needed.
After answering your note I was going to ask the group a similar question. What do others find works best re LO and filter on a tight budget? A notch filter with a very flat passband should be better?
The current system has the following parts
A HB crystal TXO as LO.
(Also tried using a Leo Bodner GPS source but not quite good enough for this application, however much better than many of the commercial rigs under test!)
A Mini-Circuits ZFM-11 mixer.
Spectrum Communications 9MHz SSB filter with extra matching transformers.
Numerous attenuators and PGA-103+ LNA to set the correct levels.
A Signal Hound SA44B spectrum analyzer.
73s
Richard
GD8EXI
On 04/12/2022, 01:30, "Robin Szemeti - G1YFG via groups.io" <robin@...> wrote:
Richard EXI,
How wide a IF filter do you use? Do you find a really wide one like 200KHz and take a reading in one go ... or do you use a fairly narrow one like 25KHz and then sweep it across with the LO signal?
On Wed, 30 Nov 2022 at 12:08, Richard GD8EXI <perwick@...> wrote:
Robin
My DIY method is as per section 3.1 “Down converter and IF filter with a spectrum analyzer” in https://www.npl.co.uk/special-pages/guides/gpg68_noise.aspx?ext
It is still work in progress but I can measure a 432MHz carrier down to -140dBc/Hz at 50kHz offset using a Signal Hound SA44B spectrum analyser, the 10.7MHz filter improving the measurement range by about 30dB. As you would expect the improvement is a lot less close in and not quantitatively useful below 5kHz, in the current form.
Also see http://www.sm5bsz.com/index.htm
73s
Richard
GD8EXI
On 30/11/2022, 00:37, "Robin Szemeti - G1YFG via groups.io <http://groups.io> " <robin@... <http://robin@...> > wrote:
I recall seeing some (I think German?) amateur's website with a design for a DIY analyser to measure phase noise ... I don't suppose anyone else recalls where it might have been?
--
Robin Szemeti - G1YFG
--
Robin Szemeti - G1YFG
|
|
|
Andy G4JNT
Any interest in one or two of these? Note that the two on teh top right have a DB25 D-type at one end (don't know what they were) The one with all the loose pairs has a pair of connectors so is some sort of breakout / monitoring thingy
All proceeds, over and above postage, to go to SCRBG / Bell Hill beacons
The nice thing about building the interface on the back of the connector is you don't need a GPIB lead, just plug the board into the back of the instrument (the board and plug are about the size of an HP HPIB plug). Then you need to hunt out cheap micro USB lead (which is not so hard!). Back in the day I had a rack of disks and an amazingly slow quarter inch tape drive running over HPIB - sadly all gone along with a load of HPIB leads which would have been useful now....
Alan
That sounds perfect, I love an Arduino project!
Now I just have to remember where I recently saw some cheap GPIB leads!
May I recommend the arduino based AR488 interfaces from USB to HPIB (google will find the code and there are a few designs around including one by Adrian Godwin that works on the back of a centronics plug). They cost next to nothing to build (the centronics connector seems the most expensive part) and they behave like the prologix. In many ways I'd rather have Ethernet to GBIP but that does still seem expensive, maybe there is a project looming for me....
I've played with one to control my HP8903A audio analyser and SA because it would break free from doing sweeps with an assumed pen recorder.
Alan The key I think is building a good diplexer on the mixer RF out, and possibly the RF in to present a good 50R match not just at the desired frequency but at the image and LO too ... a "bridged T" is a good choice. I was planning to use my Marconi 2041 as the LO, as it has (supposedly) good phase noise performance.
I shall think more on this,I suspect it is going to lead me down the rabbithole of connecting up my SA and sig gen with HPIB, then I could automate taking measurements and sweeping the SA and SigGen across in sync ... fortunately, it appears USB to HPIB interfaces are suitably expensive, so that should put me off for a while.
On Sun, 4 Dec 2022 at 15:16, Richard GD8EXI < perwick@...> wrote:
Robin a good question, as I said earlier this is still work in progress.
Initially I tried using ceramic 10.7MHz filters designed for use in FM radios with pass-bands of 280 and 12.5kHz from the junk box. However they had significant ripple in both the pass and stop-bands with poor shape factors.
I am now trying the opposite extreme using a 9.0MHz, 2.4KHz BW 6 element crystal ladder filter, photo attached. Note the ferrite transformers at both the input and output to try to get 50ohm. This has a better shape factor and stopband, with less ripple, traces attached. However a lot of effort in piecing together a wideband trace if needed.
After answering your note I was going to ask the group a similar question. What do others find works best re LO and filter on a tight budget? A notch filter with a very flat passband should be better?
The current system has the following parts
A HB crystal TXO as LO.
(Also tried using a Leo Bodner GPS source but not quite good enough for this application, however much better than many of the commercial rigs under test!)
A Mini-Circuits ZFM-11 mixer.
Spectrum Communications 9MHz SSB filter with extra matching transformers.
Numerous attenuators and PGA-103+ LNA to set the correct levels.
A Signal Hound SA44B spectrum analyzer.
73s
Richard
GD8EXI
On 04/12/2022, 01:30, "Robin Szemeti - G1YFG via groups.io" <robin@...> wrote:
Richard EXI,
How wide a IF filter do you use? Do you find a really wide one like 200KHz and take a reading in one go ... or do you use a fairly narrow one like 25KHz and then sweep it across with the LO signal?
On Wed, 30 Nov 2022 at 12:08, Richard GD8EXI <perwick@...> wrote:
Robin
My DIY method is as per section 3.1 “Down converter and IF filter with a spectrum analyzer” in https://www.npl.co.uk/special-pages/guides/gpg68_noise.aspx?ext
It is still work in progress but I can measure a 432MHz carrier down to -140dBc/Hz at 50kHz offset using a Signal Hound SA44B spectrum analyser, the 10.7MHz filter improving the measurement range by about 30dB. As you would expect the improvement is a lot less close in and not quantitatively useful below 5kHz, in the current form.
Also see http://www.sm5bsz.com/index.htm
73s
Richard
GD8EXI
On 30/11/2022, 00:37, "Robin Szemeti - G1YFG via groups.io <http://groups.io> " <robin@... <http://robin@...> > wrote:
I recall seeing some (I think German?) amateur's website with a design for a DIY analyser to measure phase noise ... I don't suppose anyone else recalls where it might have been?
--
Robin Szemeti - G1YFG
--
Robin Szemeti - G1YFG
|
|
|
Andy G4JNT
I've just realised, on looking again, there are some centronics connectors in that lot as well as HPIB. That explains the DB25 connectors at one end. SO there are only a couple of HPIB leads there
Any interest in one or two of these? Note that the two on teh top right have a DB25 D-type at one end (don't know what they were) The one with all the loose pairs has a pair of connectors so is some sort of breakout / monitoring thingy
All proceeds, over and above postage, to go to SCRBG / Bell Hill beacons
The nice thing about building the interface on the back of the connector is you don't need a GPIB lead, just plug the board into the back of the instrument (the board and plug are about the size of an HP HPIB plug). Then you need to hunt out cheap micro USB lead (which is not so hard!). Back in the day I had a rack of disks and an amazingly slow quarter inch tape drive running over HPIB - sadly all gone along with a load of HPIB leads which would have been useful now....
Alan
That sounds perfect, I love an Arduino project!
Now I just have to remember where I recently saw some cheap GPIB leads!
May I recommend the arduino based AR488 interfaces from USB to HPIB (google will find the code and there are a few designs around including one by Adrian Godwin that works on the back of a centronics plug). They cost next to nothing to build (the centronics connector seems the most expensive part) and they behave like the prologix. In many ways I'd rather have Ethernet to GBIP but that does still seem expensive, maybe there is a project looming for me....
I've played with one to control my HP8903A audio analyser and SA because it would break free from doing sweeps with an assumed pen recorder.
Alan The key I think is building a good diplexer on the mixer RF out, and possibly the RF in to present a good 50R match not just at the desired frequency but at the image and LO too ... a "bridged T" is a good choice. I was planning to use my Marconi 2041 as the LO, as it has (supposedly) good phase noise performance.
I shall think more on this,I suspect it is going to lead me down the rabbithole of connecting up my SA and sig gen with HPIB, then I could automate taking measurements and sweeping the SA and SigGen across in sync ... fortunately, it appears USB to HPIB interfaces are suitably expensive, so that should put me off for a while.
On Sun, 4 Dec 2022 at 15:16, Richard GD8EXI < perwick@...> wrote:
Robin a good question, as I said earlier this is still work in progress.
Initially I tried using ceramic 10.7MHz filters designed for use in FM radios with pass-bands of 280 and 12.5kHz from the junk box. However they had significant ripple in both the pass and stop-bands with poor shape factors.
I am now trying the opposite extreme using a 9.0MHz, 2.4KHz BW 6 element crystal ladder filter, photo attached. Note the ferrite transformers at both the input and output to try to get 50ohm. This has a better shape factor and stopband, with less ripple, traces attached. However a lot of effort in piecing together a wideband trace if needed.
After answering your note I was going to ask the group a similar question. What do others find works best re LO and filter on a tight budget? A notch filter with a very flat passband should be better?
The current system has the following parts
A HB crystal TXO as LO.
(Also tried using a Leo Bodner GPS source but not quite good enough for this application, however much better than many of the commercial rigs under test!)
A Mini-Circuits ZFM-11 mixer.
Spectrum Communications 9MHz SSB filter with extra matching transformers.
Numerous attenuators and PGA-103+ LNA to set the correct levels.
A Signal Hound SA44B spectrum analyzer.
73s
Richard
GD8EXI
On 04/12/2022, 01:30, "Robin Szemeti - G1YFG via groups.io" <robin@...> wrote:
Richard EXI,
How wide a IF filter do you use? Do you find a really wide one like 200KHz and take a reading in one go ... or do you use a fairly narrow one like 25KHz and then sweep it across with the LO signal?
On Wed, 30 Nov 2022 at 12:08, Richard GD8EXI <perwick@...> wrote:
Robin
My DIY method is as per section 3.1 “Down converter and IF filter with a spectrum analyzer” in https://www.npl.co.uk/special-pages/guides/gpg68_noise.aspx?ext
It is still work in progress but I can measure a 432MHz carrier down to -140dBc/Hz at 50kHz offset using a Signal Hound SA44B spectrum analyser, the 10.7MHz filter improving the measurement range by about 30dB. As you would expect the improvement is a lot less close in and not quantitatively useful below 5kHz, in the current form.
Also see http://www.sm5bsz.com/index.htm
73s
Richard
GD8EXI
On 30/11/2022, 00:37, "Robin Szemeti - G1YFG via groups.io <http://groups.io> " <robin@... <http://robin@...> > wrote:
I recall seeing some (I think German?) amateur's website with a design for a DIY analyser to measure phase noise ... I don't suppose anyone else recalls where it might have been?
--
Robin Szemeti - G1YFG
--
Robin Szemeti - G1YFG
|
|
|
I have numerous bits of kit to connect though ... Alan Melia has some leads cheap. The problem with USB is you plug it in and it appears on a random TTY port each time, leading to an electronic version of "where's waldo" ... I can only imagine the amount of fun that leads to if you have 2 or 3 bits of kit connected at the same time each one randomly appearing on a different TTY ...
The nice thing about building the interface on the back of the connector is you don't need a GPIB lead, just plug the board into the back of the instrument (the board and plug are about the size of an HP HPIB plug). Then you need to hunt out cheap micro USB lead (which is not so hard!). Back in the day I had a rack of disks and an amazingly slow quarter inch tape drive running over HPIB - sadly all gone along with a load of HPIB leads which would have been useful now....
Alan
That sounds perfect, I love an Arduino project!
Now I just have to remember where I recently saw some cheap GPIB leads!
May I recommend the arduino based AR488 interfaces from USB to HPIB (google will find the code and there are a few designs around including one by Adrian Godwin that works on the back of a centronics plug). They cost next to nothing to build (the centronics connector seems the most expensive part) and they behave like the prologix. In many ways I'd rather have Ethernet to GBIP but that does still seem expensive, maybe there is a project looming for me....
I've played with one to control my HP8903A audio analyser and SA because it would break free from doing sweeps with an assumed pen recorder.
Alan The key I think is building a good diplexer on the mixer RF out, and possibly the RF in to present a good 50R match not just at the desired frequency but at the image and LO too ... a "bridged T" is a good choice. I was planning to use my Marconi 2041 as the LO, as it has (supposedly) good phase noise performance.
I shall think more on this,I suspect it is going to lead me down the rabbithole of connecting up my SA and sig gen with HPIB, then I could automate taking measurements and sweeping the SA and SigGen across in sync ... fortunately, it appears USB to HPIB interfaces are suitably expensive, so that should put me off for a while.
On Sun, 4 Dec 2022 at 15:16, Richard GD8EXI < perwick@...> wrote:
Robin a good question, as I said earlier this is still work in progress.
Initially I tried using ceramic 10.7MHz filters designed for use in FM radios with pass-bands of 280 and 12.5kHz from the junk box. However they had significant ripple in both the pass and stop-bands with poor shape factors.
I am now trying the opposite extreme using a 9.0MHz, 2.4KHz BW 6 element crystal ladder filter, photo attached. Note the ferrite transformers at both the input and output to try to get 50ohm. This has a better shape factor and stopband, with less ripple, traces attached. However a lot of effort in piecing together a wideband trace if needed.
After answering your note I was going to ask the group a similar question. What do others find works best re LO and filter on a tight budget? A notch filter with a very flat passband should be better?
The current system has the following parts
A HB crystal TXO as LO.
(Also tried using a Leo Bodner GPS source but not quite good enough for this application, however much better than many of the commercial rigs under test!)
A Mini-Circuits ZFM-11 mixer.
Spectrum Communications 9MHz SSB filter with extra matching transformers.
Numerous attenuators and PGA-103+ LNA to set the correct levels.
A Signal Hound SA44B spectrum analyzer.
73s
Richard
GD8EXI
On 04/12/2022, 01:30, "Robin Szemeti - G1YFG via groups.io" <robin@...> wrote:
Richard EXI,
How wide a IF filter do you use? Do you find a really wide one like 200KHz and take a reading in one go ... or do you use a fairly narrow one like 25KHz and then sweep it across with the LO signal?
On Wed, 30 Nov 2022 at 12:08, Richard GD8EXI <perwick@...> wrote:
Robin
My DIY method is as per section 3.1 “Down converter and IF filter with a spectrum analyzer” in https://www.npl.co.uk/special-pages/guides/gpg68_noise.aspx?ext
It is still work in progress but I can measure a 432MHz carrier down to -140dBc/Hz at 50kHz offset using a Signal Hound SA44B spectrum analyser, the 10.7MHz filter improving the measurement range by about 30dB. As you would expect the improvement is a lot less close in and not quantitatively useful below 5kHz, in the current form.
Also see http://www.sm5bsz.com/index.htm
73s
Richard
GD8EXI
On 30/11/2022, 00:37, "Robin Szemeti - G1YFG via groups.io <http://groups.io> " <robin@... <http://robin@...> > wrote:
I recall seeing some (I think German?) amateur's website with a design for a DIY analyser to measure phase noise ... I don't suppose anyone else recalls where it might have been?
--
Robin Szemeti - G1YFG
--
Robin Szemeti - G1YFG
--
Robin Szemeti - G1YFG
|
|
|
The 2 with DB25 are probably Centronics printer cables for old PCs, they used a 36 way ‘micro-ribbon’ connector. The HPIB is a 24 way type. We used a lot of Commodore PETs at work C/W GPIB and had loads of HPIB cables, all probably long since skipped. 73, Mike, Mike Stevens, G8CUL/M0CUL/F4VRB.
From: UKMicrowaves@groups.io <UKMicrowaves@groups.io> On Behalf Of Andy G4JNT
Sent: 04 December 2022 18:16
To: UKMicrowaves@groups.io
Subject: Re: [UKMicrowaves] DIY Phase noise measurement? Any interest in one or two of these? Note that the two on teh top right have a DB25 D-type at one end (don't know what they were) The one with all the loose pairs has a pair of connectors so is some sort of breakout / monitoring thingy All proceeds, over and above postage, to go to SCRBG / Bell Hill beacons The nice thing about building the interface on the back of the connector is you don't need a GPIB lead, just plug the board into the back of the instrument (the board and plug are about the size of an HP HPIB plug). Then you need to hunt out cheap micro USB lead (which is not so hard!). Back in the day I had a rack of disks and an amazingly slow quarter inch tape drive running over HPIB - sadly all gone along with a load of HPIB leads which would have been useful now.... That sounds perfect, I love an Arduino project! Now I just have to remember where I recently saw some cheap GPIB leads! May I recommend the arduino based AR488 interfaces from USB to HPIB (google will find the code and there are a few designs around including one by Adrian Godwin that works on the back of a centronics plug). They cost next to nothing to build (the centronics connector seems the most expensive part) and they behave like the prologix. In many ways I'd rather have Ethernet to GBIP but that does still seem expensive, maybe there is a project looming for me.... I've played with one to control my HP8903A audio analyser and SA because it would break free from doing sweeps with an assumed pen recorder. The key I think is building a good diplexer on the mixer RF out, and possibly the RF in to present a good 50R match not just at the desired frequency but at the image and LO too ... a "bridged T" is a good choice. I was planning to use my Marconi 2041 as the LO, as it has (supposedly) good phase noise performance. I shall think more on this,I suspect it is going to lead me down the rabbithole of connecting up my SA and sig gen with HPIB, then I could automate taking measurements and sweeping the SA and SigGen across in sync ... fortunately, it appears USB to HPIB interfaces are suitably expensive, so that should put me off for a while. On Sun, 4 Dec 2022 at 15:16, Richard GD8EXI <perwick@...> wrote: Robin a good question, as I said earlier this is still work in progress.
Initially I tried using ceramic 10.7MHz filters designed for use in FM radios with pass-bands of 280 and 12.5kHz from the junk box. However they had significant ripple in both the pass and stop-bands with poor shape factors.
I am now trying the opposite extreme using a 9.0MHz, 2.4KHz BW 6 element crystal ladder filter, photo attached. Note the ferrite transformers at both the input and output to try to get 50ohm. This has a better shape factor and stopband, with less ripple, traces attached. However a lot of effort in piecing together a wideband trace if needed.
After answering your note I was going to ask the group a similar question. What do others find works best re LO and filter on a tight budget? A notch filter with a very flat passband should be better?
The current system has the following parts
A HB crystal TXO as LO.
(Also tried using a Leo Bodner GPS source but not quite good enough for this application, however much better than many of the commercial rigs under test!)
A Mini-Circuits ZFM-11 mixer.
Spectrum Communications 9MHz SSB filter with extra matching transformers.
Numerous attenuators and PGA-103+ LNA to set the correct levels.
A Signal Hound SA44B spectrum analyzer.
73s
Richard
GD8EXI
On 04/12/2022, 01:30, "Robin Szemeti - G1YFG via groups.io" <robin@...> wrote: Richard EXI,
How wide a IF filter do you use? Do you find a really wide one like 200KHz and take a reading in one go ... or do you use a fairly narrow one like 25KHz and then sweep it across with the LO signal?
On Wed, 30 Nov 2022 at 12:08, Richard GD8EXI <perwick@...> wrote: Robin
My DIY method is as per section 3.1 “Down converter and IF filter with a spectrum analyzer” in https://www.npl.co.uk/special-pages/guides/gpg68_noise.aspx?ext
It is still work in progress but I can measure a 432MHz carrier down to -140dBc/Hz at 50kHz offset using a Signal Hound SA44B spectrum analyser, the 10.7MHz filter improving the measurement range by about 30dB. As you would expect the improvement is a lot less close in and not quantitatively useful below 5kHz, in the current form.
Also see http://www.sm5bsz.com/index.htm
73s
Richard
GD8EXI
On 30/11/2022, 00:37, "Robin Szemeti - G1YFG via groups.io <http://groups.io> " <robin@... <http://robin@...> > wrote: I recall seeing some (I think German?) amateur's website with a design for a DIY analyser to measure phase noise ... I don't suppose anyone else recalls where it might have been?
--
Robin Szemeti - G1YFG
--
Robin Szemeti - G1YFG
|
|
|
Andy G4JNT
The prologix HPIB / USB interface goes via an FTDI chip COM port interface. The nice thing about FTDI is that they programme a different unique identifier into every chip, so the PC recognises individual COM ports and never reassigns them different numbers. I've only ever used FTDI devices and never had ambiguity, find Wally etc. problems with serial ports
If only they did the same with USB soundcard chips :-(
I have numerous bits of kit to connect though ... Alan Melia has some leads cheap. The problem with USB is you plug it in and it appears on a random TTY port each time, leading to an electronic version of "where's waldo" ... I can only imagine the amount of fun that leads to if you have 2 or 3 bits of kit connected at the same time each one randomly appearing on a different TTY ...
The nice thing about building the interface on the back of the connector is you don't need a GPIB lead, just plug the board into the back of the instrument (the board and plug are about the size of an HP HPIB plug). Then you need to hunt out cheap micro USB lead (which is not so hard!). Back in the day I had a rack of disks and an amazingly slow quarter inch tape drive running over HPIB - sadly all gone along with a load of HPIB leads which would have been useful now....
Alan
That sounds perfect, I love an Arduino project!
Now I just have to remember where I recently saw some cheap GPIB leads!
May I recommend the arduino based AR488 interfaces from USB to HPIB (google will find the code and there are a few designs around including one by Adrian Godwin that works on the back of a centronics plug). They cost next to nothing to build (the centronics connector seems the most expensive part) and they behave like the prologix. In many ways I'd rather have Ethernet to GBIP but that does still seem expensive, maybe there is a project looming for me....
I've played with one to control my HP8903A audio analyser and SA because it would break free from doing sweeps with an assumed pen recorder.
Alan The key I think is building a good diplexer on the mixer RF out, and possibly the RF in to present a good 50R match not just at the desired frequency but at the image and LO too ... a "bridged T" is a good choice. I was planning to use my Marconi 2041 as the LO, as it has (supposedly) good phase noise performance.
I shall think more on this,I suspect it is going to lead me down the rabbithole of connecting up my SA and sig gen with HPIB, then I could automate taking measurements and sweeping the SA and SigGen across in sync ... fortunately, it appears USB to HPIB interfaces are suitably expensive, so that should put me off for a while.
On Sun, 4 Dec 2022 at 15:16, Richard GD8EXI < perwick@...> wrote:
Robin a good question, as I said earlier this is still work in progress.
Initially I tried using ceramic 10.7MHz filters designed for use in FM radios with pass-bands of 280 and 12.5kHz from the junk box. However they had significant ripple in both the pass and stop-bands with poor shape factors.
I am now trying the opposite extreme using a 9.0MHz, 2.4KHz BW 6 element crystal ladder filter, photo attached. Note the ferrite transformers at both the input and output to try to get 50ohm. This has a better shape factor and stopband, with less ripple, traces attached. However a lot of effort in piecing together a wideband trace if needed.
After answering your note I was going to ask the group a similar question. What do others find works best re LO and filter on a tight budget? A notch filter with a very flat passband should be better?
The current system has the following parts
A HB crystal TXO as LO.
(Also tried using a Leo Bodner GPS source but not quite good enough for this application, however much better than many of the commercial rigs under test!)
A Mini-Circuits ZFM-11 mixer.
Spectrum Communications 9MHz SSB filter with extra matching transformers.
Numerous attenuators and PGA-103+ LNA to set the correct levels.
A Signal Hound SA44B spectrum analyzer.
73s
Richard
GD8EXI
On 04/12/2022, 01:30, "Robin Szemeti - G1YFG via groups.io" <robin@...> wrote:
Richard EXI,
How wide a IF filter do you use? Do you find a really wide one like 200KHz and take a reading in one go ... or do you use a fairly narrow one like 25KHz and then sweep it across with the LO signal?
On Wed, 30 Nov 2022 at 12:08, Richard GD8EXI <perwick@...> wrote:
Robin
My DIY method is as per section 3.1 “Down converter and IF filter with a spectrum analyzer” in https://www.npl.co.uk/special-pages/guides/gpg68_noise.aspx?ext
It is still work in progress but I can measure a 432MHz carrier down to -140dBc/Hz at 50kHz offset using a Signal Hound SA44B spectrum analyser, the 10.7MHz filter improving the measurement range by about 30dB. As you would expect the improvement is a lot less close in and not quantitatively useful below 5kHz, in the current form.
Also see http://www.sm5bsz.com/index.htm
73s
Richard
GD8EXI
On 30/11/2022, 00:37, "Robin Szemeti - G1YFG via groups.io <http://groups.io> " <robin@... <http://robin@...> > wrote:
I recall seeing some (I think German?) amateur's website with a design for a DIY analyser to measure phase noise ... I don't suppose anyone else recalls where it might have been?
--
Robin Szemeti - G1YFG
--
Robin Szemeti - G1YFG
--
Robin Szemeti - G1YFG
|
|
|
Hmm ... I guess multiple Arduinos is an option ... I can always search the address bus to find out what is on each one. I guess its a toss up between more Nanos or more cables ...
The prologix HPIB / USB interface goes via an FTDI chip COM port interface. The nice thing about FTDI is that they programme a different unique identifier into every chip, so the PC recognises individual COM ports and never reassigns them different numbers. I've only ever used FTDI devices and never had ambiguity, find Wally etc. problems with serial ports
If only they did the same with USB soundcard chips :-(
I have numerous bits of kit to connect though ... Alan Melia has some leads cheap. The problem with USB is you plug it in and it appears on a random TTY port each time, leading to an electronic version of "where's waldo" ... I can only imagine the amount of fun that leads to if you have 2 or 3 bits of kit connected at the same time each one randomly appearing on a different TTY ...
The nice thing about building the interface on the back of the connector is you don't need a GPIB lead, just plug the board into the back of the instrument (the board and plug are about the size of an HP HPIB plug). Then you need to hunt out cheap micro USB lead (which is not so hard!). Back in the day I had a rack of disks and an amazingly slow quarter inch tape drive running over HPIB - sadly all gone along with a load of HPIB leads which would have been useful now....
Alan
That sounds perfect, I love an Arduino project!
Now I just have to remember where I recently saw some cheap GPIB leads!
May I recommend the arduino based AR488 interfaces from USB to HPIB (google will find the code and there are a few designs around including one by Adrian Godwin that works on the back of a centronics plug). They cost next to nothing to build (the centronics connector seems the most expensive part) and they behave like the prologix. In many ways I'd rather have Ethernet to GBIP but that does still seem expensive, maybe there is a project looming for me....
I've played with one to control my HP8903A audio analyser and SA because it would break free from doing sweeps with an assumed pen recorder.
Alan The key I think is building a good diplexer on the mixer RF out, and possibly the RF in to present a good 50R match not just at the desired frequency but at the image and LO too ... a "bridged T" is a good choice. I was planning to use my Marconi 2041 as the LO, as it has (supposedly) good phase noise performance.
I shall think more on this,I suspect it is going to lead me down the rabbithole of connecting up my SA and sig gen with HPIB, then I could automate taking measurements and sweeping the SA and SigGen across in sync ... fortunately, it appears USB to HPIB interfaces are suitably expensive, so that should put me off for a while.
On Sun, 4 Dec 2022 at 15:16, Richard GD8EXI < perwick@...> wrote:
Robin a good question, as I said earlier this is still work in progress.
Initially I tried using ceramic 10.7MHz filters designed for use in FM radios with pass-bands of 280 and 12.5kHz from the junk box. However they had significant ripple in both the pass and stop-bands with poor shape factors.
I am now trying the opposite extreme using a 9.0MHz, 2.4KHz BW 6 element crystal ladder filter, photo attached. Note the ferrite transformers at both the input and output to try to get 50ohm. This has a better shape factor and stopband, with less ripple, traces attached. However a lot of effort in piecing together a wideband trace if needed.
After answering your note I was going to ask the group a similar question. What do others find works best re LO and filter on a tight budget? A notch filter with a very flat passband should be better?
The current system has the following parts
A HB crystal TXO as LO.
(Also tried using a Leo Bodner GPS source but not quite good enough for this application, however much better than many of the commercial rigs under test!)
A Mini-Circuits ZFM-11 mixer.
Spectrum Communications 9MHz SSB filter with extra matching transformers.
Numerous attenuators and PGA-103+ LNA to set the correct levels.
A Signal Hound SA44B spectrum analyzer.
73s
Richard
GD8EXI
On 04/12/2022, 01:30, "Robin Szemeti - G1YFG via groups.io" <robin@...> wrote:
Richard EXI,
How wide a IF filter do you use? Do you find a really wide one like 200KHz and take a reading in one go ... or do you use a fairly narrow one like 25KHz and then sweep it across with the LO signal?
On Wed, 30 Nov 2022 at 12:08, Richard GD8EXI <perwick@...> wrote:
Robin
My DIY method is as per section 3.1 “Down converter and IF filter with a spectrum analyzer” in https://www.npl.co.uk/special-pages/guides/gpg68_noise.aspx?ext
It is still work in progress but I can measure a 432MHz carrier down to -140dBc/Hz at 50kHz offset using a Signal Hound SA44B spectrum analyser, the 10.7MHz filter improving the measurement range by about 30dB. As you would expect the improvement is a lot less close in and not quantitatively useful below 5kHz, in the current form.
Also see http://www.sm5bsz.com/index.htm
73s
Richard
GD8EXI
On 30/11/2022, 00:37, "Robin Szemeti - G1YFG via groups.io <http://groups.io> " <robin@... <http://robin@...> > wrote:
I recall seeing some (I think German?) amateur's website with a design for a DIY analyser to measure phase noise ... I don't suppose anyone else recalls where it might have been?
--
Robin Szemeti - G1YFG
--
Robin Szemeti - G1YFG
--
Robin Szemeti - G1YFG
--
Robin Szemeti - G1YFG
|
|
|
Hi. Re:- The problem with USB is you plug it in and it appears on a random TTY port each time, leading to an electronic version of "where's waldo" Indeed, they often "go walkabout" too, if for whatever reason the system has to re-enumerate the blessed things, as can happen when another "thing" is connected on the fly! (Or RFI "nukes" the USB side so they disappear then re-appear to the OS when the RF has gone..) That is fixable (the variable allocation issue that is) in both Windows and Linux, possibly BSD, I have no knowledge of Apple MAC's. Windows: Use Device Mangler to drill down to the active settings a com-port uses, then assign it the port you want, that is physically "not in use". (Even what Windows thinks is "in use" if you know does not exist, are free to choose.) Then... So long as you connect your gadget to the exact same USB port by the same method (hubs/splitters etc) you will get the same COMx port. https://sourceforge.net/p/fldigi/wiki/windows_com_howto/ For example. Linux: Use "UDEV rules" to tame the thing. Not trivial, but easy when you know how. Too long to go into here, ask if needed. Or take a look at:- https://linuxconfig.org/tutorial-on-how-to-write-basic-udev-rules-in-linuxIn both cases, it helps enormously, if you use FTDI chipset based USB to Serial devices (as Andy 'JNT suggests) or if Arduino based, make sure they present a "Unique" identity to the host if they emulate such things themselves. Then UDEV Rules under Linux (and the Windows dodge tweak above) work very well.. (Or you can use the "Path" to the device, but then you have to ensure it is connected to the PC the exact same way each time!) The problem is that the vast majority of "cheap" USB to whatever devices, have no unique attribute (such as a serial number.) Where as, that is what FTDI provide, making life easy. ("Genuine" FTDI devices, can also have many of those attributes customised using a software utility (Windows only though. There is a Linux tool to do similar things, but it's command line only and somewhat confusing) to change things. (Useful for example to change the "sense" of handshake or data lines, when using opo-isolators etc, making PCB layouts a bit easier, and potentially reducing the component count.) Note that some "Fake" FTDI chips appear to allow that, but when you power them down/up any changes you might have made, are lost. 73. Dave G8KBV PS: You can also graft on "proper" GPIB driver IC's to the AR488 USB/GPIB projects, so you can have one interface controlling more then one instrument, using the correct levels etc, much like the "Professional" devices... IIRC, there are sections of code to uncomment so they can use such drivers correctly. Not looked for a while though. -- Created on and sent from a Unix like PC running and using free and open source software:
|
|
|
Take care with "Cheap" GPIB / IEEE488 / IEC625* leads.
( * Often found on R&S equipment, electrically the same, but using what look like regular 'D' type connectors.)
If I had a £ for every time a bad lead was the eventual found cause for weird things happening on customer sites in the past job, I could have retired years ago!
If just one ground wire (the data and handshake lines are twisted pairs) comes adrift at one end, it can make a serious mess of that bus, but not under all conditions, or for all devices on the same bus! "Buzzing" out a cable will not detect it, as there are multiple "ground" wires...
If you find some surplus cables, avoid the moulded on connector types, as they are nigh on impossible to repair, short of cutting off the ends, and hand wiring new connectors on. (And you won't know they are bad, until chaos ensues at some point in time.
Where as the older types where the ends have two shells that are held together with screws, they are very repairable indeed! (If fiddly.)
Avoid unscreened ribbon cables (even short ones) especially in a "RF rich" environment!
The above learnt from nearly 30 years supporting customer EMC test systems, large and small... Oddly, the larger systems are easier to debug!
73
Dave G8KBV
--
Created on and sent from a Unix like PC running and using free and open source software:
|
|
|
