Topics

Equipment reviews in RadCom.

Peter Chadwick
 

I was a bit disappointed that the FTDX101 review in October's RadCom only listed Tx 3rd and 5th order IPs, as it is the higher orders that generally cause trouble to other users. The ARRL reviews go up to the 9th and on occasion even higher, and since the transceivers with valved PAs disappeared off the market, the high order Tx IMD performance of transceivers has degraded appreciably over the years. There will be an article containing an analysis of this in the ARRL's National Contest Journal in the New Year, based on an update of my 2013 RSGB Convention presentation.


Does anyone else feel it would be helpful to include higher order IP results?


vy 73


Peter G3RZP

David J Taylor GM8ARV
 

From: Peter Chadwick via Groups.Io

I was a bit disappointed that the FTDX101 review in October's RadCom only listed Tx 3rd and 5th order IPs, as it is the higher orders that generally cause trouble to other users. The ARRL reviews go up to the 9th and on occasion even higher, and since the transceivers with valved PAs disappeared off the market, the high order Tx IMD performance of transceivers has degraded appreciably over the years. There will be an article containing an analysis of this in the ARRL's National Contest Journal in the New Year, based on an update of my 2013 RSGB Convention presentation.

Does anyone else feel it would be helpful to include higher order IP results?

vy 73
Peter G3RZP
========================================

Peter,

For me personally, no.

I would have liked to see audio response plots, TX and RX, particularly with reference to other receivers taking an identical signal. Which RX was better pulling signals out of the noise? Comparisons not allowed, I suppose! I missed any comments on frequency accuracy, drift, lack of 10 MHz reference input. Does the rig even have Fusion? Not mentioned in the sellers adverts. Was that mode tested?

Now I do accept that a detailed review would take much longer, and probably fill more pages than the editor would allow! I did appreciate the comments on the front panel not having the functions the reviewer wanted. Are there no user-programmable buttons & controls? For me I would like to see far fewer buttons, knob & dials!

73,
David GM8ARV
--
SatSignal Software - Quality software for you
Web: http://www.satsignal.eu
Email: david-taylor@...
Twitter: @gm8arv

Neil Smith G4DBN
 

I wonder if a better measure might be to inject something standardised which looks like real voice, then measure the amount of energy transmitted in adjacent frequencies, perhaps in specified bandwidths so there are some numbers rather than graphs, and doing it with a dynamically-varying level so it picks up any of the other weirdness casue by poor ALC design or terrible compression algorithms.  What I care about is whether the S9+60dB signal 3kHz away is going to cover up the weak signal I'm trying to hear.  It may have good IMD3 and 5, but if the higher-order IMDs are crap, or there is phase noise or spurs, the radio should be exposed as antisocial.

I guess the standards used for commercial ACPR measurements are based on digital signals.  I suggest a test sample should be recorded by one of the stentorian voices we hear at the weekends bellowing "CQ CONTEST".  Proper real-world test.

Neil G4DBN


Glossary: ACPR = Adjacent Channel Power Ratio



On 14/09/2019 15:06, Peter Chadwick via Groups.Io wrote:

I was a bit disappointed that the FTDX101 review in October's RadCom only listed Tx 3rd and 5th order IPs, as it is the higher orders that generally cause trouble to other users. The ARRL reviews go up to the 9th and on occasion even higher, and since the transceivers with valved PAs disappeared off the market, the high order Tx IMD performance of transceivers has degraded appreciably over the years. There will be an article containing an analysis of this in the ARRL's National Contest Journal in the New Year, based on an update of my 2013 RSGB Convention presentation.


Does anyone else feel it would be helpful to include higher order IP results?


vy 73


Peter G3RZP

Peter Chadwick
 

An alternative is to measure the bandwidth at various points when the tx is modulated with noise which has a narrow notch in the pass band, as was done with analogue FDM systems. The 'Marconi Instrumentation' magazine back in the early 1970s had an article on using that method for measuring ISB transmitters. You would probably want pseudo white noise with a lower peak to average ratio than true 'white' noise. I suspect that 2 tone measurements would still be useful, especially with two tones spaced a few tens of Hertz apart to exercise power supply dynamic regulation and ALC.


vy 73


Peter G3RZP



------ Original Message ------
From: "Neil Smith G4DBN" <neil@...>
To: RSGB-Workshop@groups.io
Sent: Saturday, 14 Sep, 19 At 15:56
Subject: Re: [RSGB-Workshop] Equipment reviews in RadCom.

I wonder if a better measure might be to inject something standardised which looks like real voice, then measure the amount of energy transmitted in adjacent frequencies, perhaps in specified bandwidths so there are some numbers rather than graphs, and doing it with a dynamically-varying level so it picks up any of the other weirdness casue by poor ALC design or terrible compression algorithms. What I care about is whether the S9+60dB signal 3kHz away is going to cover up the weak signal I'm trying to hear. It may have good IMD3 and 5, but if the higher-order IMDs are crap, or there is phase noise or spurs, the radio should be exposed as antisocial.

I guess the standards used for commercial ACPR measurements are based on digital signals. I suggest a test sample should be recorded by one of the stentorian voices we hear at the weekends bellowing "CQ CONTEST". Proper real-world test.

Neil G4DBN


Glossary: ACPR = Adjacent Channel Power Ratio



On 14/09/2019 15:06, Peter Chadwick via Groups.Io wrote:

I was a bit disappointed that the FTDX101 review in October's RadCom only listed Tx 3rd and 5th order IPs, as it is the higher orders that generally cause trouble to other users. The ARRL reviews go up to the 9th and on occasion even higher, and since the transceivers with valved PAs disappeared off the market, the high order Tx IMD performance of transceivers has degraded appreciably over the years. There will be an article containing an analysis of this in the ARRL's National Contest Journal in the New Year, based on an update of my 2013 RSGB Convention presentation.


Does anyone else feel it would be helpful to include higher order IP results?


vy 73


Peter G3RZP

geoffrey pike
 

Hi Peter,
Yes i noticed this, but more importantly for those who care there was no comment that some of the 5th orders where worse than the 3rds. 
I suspect the PA has a pair of RD100s, but they have earth straps on them so you cant see.
What is needed is an extended version of this for those who want to see it, lets face it you usually record up to 11th so why not
have it published somewhere.
cheers
Geoff
GI0GDP

On Saturday, 14 September 2019, 15:06:56 BST, Peter Chadwick via Groups.Io <g8on@...> wrote:


I was a bit disappointed that the FTDX101 review in October's RadCom only listed Tx 3rd and 5th order IPs, as it is the higher orders that generally cause trouble to other users. The ARRL reviews go up to the 9th and on occasion even higher, and since the transceivers with valved PAs disappeared off the market, the high order Tx IMD performance of transceivers has degraded appreciably over the years. There will be an article containing an analysis of this in the ARRL's National Contest Journal in the New Year, based on an update of my 2013 RSGB Convention presentation.


Does anyone else feel it would be helpful to include higher order IP results?


vy 73


Peter G3RZP

Robert G8RPI
 

The comment in the text that "two tone distortion products were generally quite reasonable." is not very helpful. What were the ones that weren't reasonable? With my cynical side I would suspect editorial pressure with an eye on advertising revenue.

Robert G8RPI.

Richard Lamont
 

On 14/09/2019 15:06, Peter Chadwick via Groups.Io wrote:
I was a bit disappointed that the FTDX101 review in October's RadCom
only listed Tx 3rd and 5th order IPs, as it is the higher orders that
generally cause trouble to other users. The ARRL reviews go up to the
9th and on occasion even higher
[snip]
Does anyone else feel it would be helpful to include higher order IP
results?
Possibly. I feel it would be equally useful to show what happens if the
audio input is massively overdriven with, say, a 1 kHz sine wave into
the mic socket at the maximum level that could be applied without
risking damage. That might sort the sheep from the goats.

73,
Richard G4DYA

Dave GW4GTE
 

re FTdx101 review.
My thoughts
1. Calling IMD products 'generally acceptable' without qualification is not technical comment, it's personal opinion. It's a moot point though, in the sense that personal opinion in other contexts is - I would say - ok. Such as "I found the radio great to work with". However.....
2. Picking out another comment " as with all modern radios the performance cannot be faulted" is open to challenge. 
I was astounded to see a 3rd order IMD figure of -22dB on top band!!! This (unless I missed it) was not commented on.-22 dB is appallingly bad, and this sort of figure has no place in a modern radio. I also saw no comment either on the wildly differing IMD performance between bands.
3. "9 MHz IF decimated to 24 kHz". Does anyone have an opinion on whether the process gain is enough to avoid this being a weak spot in the architecture.
4. I don't see much value in measuring higher order IMD levels. Third order is the most important if you're on the adjacent channel. No amount of over-hyped receive overload/selectivity performance is going to remove the crud from -22dB IMD !
5.I would like to see the IMD measurements made at the manufacturer's specified power level - 100W if it's 100W, not 104W or 107W. How is the rig driven to get the higher power? I would also like to see the IMD at a more realistic output level of -3dB as a comparison. 
6. The review makes no mention of the absence of a 10 MHz ref clock input. Surely a user has a right to expect that feature on a high-end radio.
7. Receive IP3 : 50 kHz spacing, with 3 kHz roofing filter in line? (thus de-stressing most of the set). I'd like to see comparisons with the roofing filter out. People may like to know if it's worth using it.
8. Most reviews (from any source by any author - I'm generalising) shine a kindly light on the DUT, and one has to self-interpret the detail for the bad bits or have other knowledge (ref clock). I'm not suggesting that this review is biased, just that for me there are weak points that weren't highlighted. And it's often the weak points that make the difference to someone making a big buying decision. 
9. As ever 'caveat emptor'.  Yaesu hype: "The conclusive choice - transmit signal purity" The reality: IMD3, 160m, -22 dB. Of course other than Apache, no manufacturer ever, but ever states tx IMD figures so the rest are as bad.
Dave GW4GTE

Dr. David Kirkby from Kirkby Microwave Ltd
 

On Sun, 15 Sep 2019 at 08:58, Robert G8RPI via Groups.Io <robert8rpi=yahoo.co.uk@groups.io> wrote:
The comment in the text that "two tone distortion products were generally quite reasonable." is not very helpful.

I think my teachers in the state school I went to would have picked me up for such a comment. 

Robert G8RPI.

--
Dr. David Kirkby,
Kirkby Microwave Ltd,
drkirkby@...
https://www.kirkbymicrowave.co.uk/
Telephone 01621-680100./ +44 1621 680100

Registered in England & Wales.
Company number 08914892.
Registered office:
Stokes Hall Lodge,
Burnham Rd,
Althorne,
Chelmsford,
Essex,
CM3 6DT,
United Kingdom