Topics

uBitx v5 issue


Bill Robbins
 

I am just completing my V5.  No power applied yet.   Everything is remove from the board and it lays naked in front of me.  The following connectors are installed; Power, Raduino, Audio, as well as the IFR510's.

My problem is that the both leads of the antenna connector, presumably ground and lead, show continuity to ground.  They are shorted. I find no solder bridges anywhere.

Ideas??

Bill


Jerry Gaffke
 

Antenna connector looks shorted to ground because there is a DC path through K3-13,11, K1-12,14, L1,L2,L3,L4, T2-1,6
Jerry, KE7ER


On Mon, Nov 23, 2020 at 12:33 PM, Bill Robbins wrote:
I am just completing my V5.  No power applied yet.   Everything is remove from the board and it lays naked in front of me.  The following connectors are installed; Power, Raduino, Audio, as well as the IFR510's.

My problem is that the both leads of the antenna connector, presumably ground and lead, show continuity to ground.  They are shorted. I find no solder bridges anywhere.

Ideas??

Bill


Curt
 

Bill

great you are checking these things as you assemble.  do examine the schematic and you will see what Jerry is citing - funny thing I was once worried as you!  oh yes - when you power up and touch the regulator on the raduino - don't be alarmed when you discover it is quite warm.  and that your BFO calibration may be way off. 

welcome to the global bitx community.  73

curt


Bill Robbins
 

Thanks Curt. 


On Nov 23, 2020, at 5:03 PM, Curt via groups.io <wb8yyy@...> wrote:

Bill

great you are checking these things as you assemble.  do examine the schematic and you will see what Jerry is citing - funny thing I was once worried as you!  oh yes - when you power up and touch the regulator on the raduino - don't be alarmed when you discover it is quite warm.  and that your BFO calibration may be way off. 

welcome to the global bitx community.  73

curt


iz oos
 

I am not totally sure, but that DC Path at the antenna input might provider some degree of protection in case of statics.


Il lun 23 nov 2020 11:07 PM Bill Robbins <wa8cdu@...> ha scritto:
Thanks Curt. 


On Nov 23, 2020, at 5:03 PM, Curt via groups.io <wb8yyy=yahoo.com@groups.io> wrote:

Bill

great you are checking these things as you assemble.  do examine the schematic and you will see what Jerry is citing - funny thing I was once worried as you!  oh yes - when you power up and touch the regulator on the raduino - don't be alarmed when you discover it is quite warm.  and that your BFO calibration may be way off. 

welcome to the global bitx community.  73

curt


Evan Hand
 

iz,
The dc path does not really help in protecting the rig in my opinion.  Q90 is the first driver stage in the transmitter, it is connected before the receiver low pass filter, and has been known to fail when the antenna is connected and there are strong nearby RF signals.  There is a preventative measure that I and some others have taken to protect that device.  Back to Back 1n4148 diodes across pins 12 and 16 of K1 would be some protection from strong RF.  Lightning would be beyond the simple diode protection.  A lightning arrestor on the feed line is a better solution to protect against damage from large static discharges, or better yet ground all antennas when storms are near

My opinions so there could be better solutions.  Feedback always welcome.
73
Evan
AC9TU. 


Jerry Gaffke
 

I think Iz is right, that DC path does help with static on the antenna, 
which can get quite substantial with nothing more than the wind whistling past.
Evan is also right, the DC path does not protect Q90 from something like
a nearby FM or TV broadcast station due to the 30mhz LPF not allowing it through.

HF will get through the 30mhz LPF, and likely get at least somewhat snubbed out by the diodes in the first mixer.

Gordon GIbby has pointed out that the base-emitter diode of Q90 will conduct
only on positive RF peaks from a nearby (VHF?) transmission, causing C80 to charge up.
The resulting negative voltage on the base during negative peaks can exceed the Vebo spec
of a 2n3904, 6 volts max, and blow out Q90.  He suggests a 1n4148 or similar, anode at Q90-emitter,
cathode at Q90-base, to prevent Q90-base from going more than a diode drop below the emitter, and thus
avoid a violation of Vebo.

Farhan has suggested that there might be more birdies in the receiver if the back-to-back diodes are
added, due to the wideband nature of everything between antenna and first mixer.  The diodes will
create lots of harmonics, creating lots of new mixer products that could sneak through the IF filters.

Most uBitx owners haven't bothered worrying about protecting Q90,
but a few have been bit.

Jerry, KE7ER


On Mon, Nov 23, 2020 at 04:39 PM, Evan Hand wrote:
iz,
The dc path does not really help in protecting the rig in my opinion.  Q90 is the first driver stage in the transmitter, it is connected before the receiver low pass filter, and has been known to fail when the antenna is connected and there are strong nearby RF signals.  There is a preventative measure that I and some others have taken to protect that device.  Back to Back 1n4148 diodes across pins 12 and 16 of K1 would be some protection from strong RF.  Lightning would be beyond the simple diode protection.  A lightning arrestor on the feed line is a better solution to protect against damage from large static discharges, or better yet ground all antennas when storms are near

My opinions so there could be better solutions.  Feedback always welcome.
73
Evan
AC9TU. 


Bob Lunsford
 

Don't recall if my V6 had one but I remember seeing a picture of the regulator and it had a about a 3/4-in square heat regulator attached to the regulator. If one is in your parts drawer, I think it would be wise to attach it to the regulator. Or, you could bend up a small strip of aluminum and make one. No real danger of shorting and no need to insulate it from the tab. It would provide peace of mind when the regulator heats up and most of the heat is drawn away and dissipated.

Bob — KK5R

On Monday, November 23, 2020, 5:03:34 PM EST, Curt via groups.io <wb8yyy@...> wrote:


Bill

great you are checking these things as you assemble.  do examine the schematic and you will see what Jerry is citing - funny thing I was once worried as you!  oh yes - when you power up and touch the regulator on the raduino - don't be alarmed when you discover it is quite warm.  and that your BFO calibration may be way off. 

welcome to the global bitx community.  73

curt


Bob Lunsford
 

Correction: ...a 3/4-in square heat radiator attached to the regulator....

On Monday, November 23, 2020, 10:38:02 PM EST, Bob Lunsford via groups.io <nocrud222@...> wrote:


Don't recall if my V6 had one but I remember seeing a picture of the regulator and it had a about a 3/4-in square heat regulator attached to the regulator. If one is in your parts drawer, I think it would be wise to attach it to the regulator. Or, you could bend up a small strip of aluminum and make one. No real danger of shorting and no need to insulate it from the tab. It would provide peace of mind when the regulator heats up and most of the heat is drawn away and dissipated.

Bob — KK5R

On Monday, November 23, 2020, 5:03:34 PM EST, Curt via groups.io <wb8yyy@...> wrote:


Bill

great you are checking these things as you assemble.  do examine the schematic and you will see what Jerry is citing - funny thing I was once worried as you!  oh yes - when you power up and touch the regulator on the raduino - don't be alarmed when you discover it is quite warm.  and that your BFO calibration may be way off. 

welcome to the global bitx community.  73

curt


Evan Hand
 

Jerry,
Thank you for the feedback.  It is a good summary of the data that has been presented in this forum as far as I know.  

No one has yet published any data on either the susceptibility to high RF (what is the actual threshold before failure), on the effect of the diode on Q90 to transmission, or back to back diodes on reception. 

There was concern about the reverse diode on the base of Q90 having a negative impact on the SSB transmission.  Has anyone tried that?  If so, it would be good to know if it works with no adverse effects.  

I have done the back to back diodes on the input to my v4 board and have not noticed any negative effects on the receive capability.  I have NOT done any significant testing to verify the before and after results of the diodes, so this is not conclusive of this being a good solution other than one that works at my QTH where the noise floor is very high.  The high noise could mask any of the potential problems that Farhan points out.  NOTE: the noise issue is on all of my rigs that include an Icom 7300, Hermes HL2, QCX+, and an RSP1a as well as 2 v4 and one v5 uBITX.  I have only modified one of my v4 boards to test the effect of the diodes.

I have not had the Q90 failure problem, mainly because I very rigorously only connect one HF rig to an antenna at a time (I do have three HF antennas up).  I do have a 50 watt Kenwood TM-D710G 2m/70cm that I do use at the same time, and so far it has not destroyed any of my uBITX Q90s.  I have also tried to measure the signal strength of the VHF/UHF antenna to the closest HF antenna, and could not get an effective measurement.  I tried that with my oscilloscope.  The sensitivity is only 2mv/division (1x probe setting).  That should have been low enough to measure any signal large enough to overcome the 0.6volt diode threshold.  The measurements would be specific to my QTH and antenna placement.

To sum up my opinion at this point; without more data, I would do one of the diode protection schemes only if you will be using the rig near other high RF transmitters when connected to an antenna, or if you do experience a Q90 failure.  I am still leaning towards the back to back diodes as that is switched out when the uBITX goes into transmit.  This could be modified to a single diode connected as to protect the base to emitter junction, just installed at the switched point like the back to back diodes.

As a side note, the back to back diodes was cited on this reflector (sorry, I do not have the original post), and lead me to further internet research on my part.  I found that the scheme (with variations) have been used to protect receivers quite often for the exact same reason: protecting sensitive components connected to the antenna from strong RF fields.  Here is one of the articles that I found:
http://www.ad5x.com/images/Articles/FrontEndProt.pdf
and another
http://www.na0tc.org/lib/exe/fetch.php?media=technical:rx_protect.pdf
Both use variations on the back to back diodes.  These are only two of the references that I found.


Jerry, again thank you!  Also thanks to Iz and Farhan for sharing their thoughts.  This is how I learn.
73
Evan
AC9TU


Ted
 

Would the common 1n914 serve the same purpose?   Being as common as it is, etc.


Ted
k3rta


Evan Hand
 

Ted,
Yes, the 1n914 or equal would work as far as the information that I have found and what I have tried.  I believe that the 1n4148 is a direct replacement for the now obsolete 1n914.

73
Evan
AC9TU


Bill Cromwell
 

Hi Ted,

I am preparing to replace Q90 in my V3 and I will be installing the diodes. I have used them in other solid state rigs in the past and a couple of time I have had to replace them (two different radios). The usual failure mode is for them to fail-short. It becomes obvious then that they have failed because incoming signals pretty much vanish. In both cases the rest of the radio was preserved. The sacrificial diodes did their job.

I have never had any problems with radios that have those diodes installed. Most of the caveats that have been posted here are *speculation* about what might happen. It hasn't happened. None of the posts have asserted that any problem has actually happened. I have two experiences where the front ends were saved and no experiences where the diodes caused any problems. And that is experience, not speculation.

73,

Bill KU8H

bark less - wag more

On 11/24/20 11:40 AM, Ted via groups.io wrote:
Would the common 1n914 serve the same purpose?   Being as common as it is, etc.
Ted
k3rta


Jerry Gaffke
 

Evan,

Good to know that your back-to-back diodes present no issues with reception.
Seems a fine solution.  If incoming RF is kept below 4 volts peak-to-peak,
there should be no trouble with Q90 blowing due to Veb-max of 6v being violated.

I would not call the 1n914 obsolete, they still sell them on Mouser.
1n914's are an older part, were not spec'd as tight as the newer 1n4148, but otherwise identical.
I suspect that any parts sold as 1n914's in the last few decades come off the same line
as the 1n4148's, just slapped with different labeling.

>  There was concern about the reverse diode on the base of Q90 having a negative impact on the SSB transmission. 

I'd be surprised if this was an issue, but worth looking at if you have a way to measure IMD.
Transmit RF into Q90 should be down around 100mv rms, so currents through
a silicon diode would be measured in picoamps.
Also, we already have the base-emitter diode in there to cause distortion of large signals, 
adding a 1n4148 going the other way across that junction diode would make it symetrical, 
and may well reduce distortion rather than add to it.

Jerry, KE7ER




On Tue, Nov 24, 2020 at 03:38 AM, Evan Hand wrote:

I have done the back to back diodes on the input to my v4 board and have not noticed any negative effects on the receive capability. 


Evan Hand
 

Jerry,

Again, Thank You.  What you state makes sense.  When I figure out how to measure IMD with the tools that I have I will try it on my other v4 uBITX board.

73
Evan
AC9TU


Bob Lunsford
 

Would be nice to install a couple of solder terminals so the two diodes could be easily replaced. I bought 100 switching diodes from Radio Shack when they were business and have only used less than five of the batch. Switching the radio to ground by way of a switch that causes the radio input to go to ground when not being used is another form of cheap insurance.

Replacing the diodes would be much better than replacing Q90 in my book. Less trouble and faster.

Bob — KK5R

On Tuesday, November 24, 2020, 12:44:31 PM EST, Bill Cromwell <wrcromwell@...> wrote:


Hi Ted,

I am preparing to replace Q90 in my V3 and I will be installing the
diodes. I have used them in other solid state rigs in the past and a
couple of time I have had to replace them (two different radios). The
usual failure mode is for them to fail-short. It becomes obvious then
that they have failed because incoming signals pretty much vanish. In
both cases the rest of the radio was preserved. The sacrificial diodes
did their job.

I have never had any problems with radios that have those diodes
installed. Most of the caveats that have been posted here are
*speculation* about what might happen. It hasn't happened. None of the
posts have asserted that any problem has actually happened. I have two
experiences where the front ends were saved and no experiences where the
diodes caused any problems. And that is experience, not speculation.

73,

Bill  KU8H

bark less - wag more

On 11/24/20 11:40 AM, Ted via groups.io wrote:
> Would the common 1n914 serve the same purpose?   Being as common as it
> is, etc.
>
>
> Ted
> k3rta
>






Bill Cromwell
 

Hi,

The diodes are not very likey to need replacing. Or do you live in Florida - the lightning capital of the world? An impulse from a nearby lightning strike might take the diodes (and might not). Of course, a direct hit by lightning and they won't help. But you will have bigger problems than the diodes or Q90.

73,

Bill KU8

bark less - wag more

On 11/24/20 6:22 PM, Bob Lunsford via groups.io wrote:
Would be nice to install a couple of solder terminals so the two diodes could be easily replaced. I bought 100 switching diodes from Radio Shack when they were business and have only used less than five of the batch. Switching the radio to ground by way of a switch that causes the radio input to go to ground when not being used is another form of cheap insurance.
Replacing the diodes would be much better than replacing Q90 in my book. Less trouble and faster.
Bob — KK5R
On Tuesday, November 24, 2020, 12:44:31 PM EST, Bill Cromwell <wrcromwell@gmail.com> wrote:
Hi Ted,
I am preparing to replace Q90 in my V3 and I will be installing the
diodes. I have used them in other solid state rigs in the past and a
couple of time I have had to replace them (two different radios). The
usual failure mode is for them to fail-short. It becomes obvious then
that they have failed because incoming signals pretty much vanish. In
both cases the rest of the radio was preserved. The sacrificial diodes
did their job.
I have never had any problems with radios that have those diodes
installed. Most of the caveats that have been posted here are
*speculation* about what might happen. It hasn't happened. None of the
posts have asserted that any problem has actually happened. I have two
experiences where the front ends were saved and no experiences where the
diodes caused any problems. And that is experience, not speculation.
73,
Bill  KU8H
bark less - wag more
On 11/24/20 11:40 AM, Ted via groups.io wrote:
> Would the common 1n914 serve the same purpose?   Being as common as it
> is, etc.
>
>
> Ted
> k3rta
>