No transmit modulation (I think)

Post Reply
kk7bco
 

Posts: 8
Joined: 05 Dec 2022, 03:32

No transmit modulation (I think)

Post by kk7bco »

Hello,
I built a kit from the "official" link this summer, and found it to work very well on receive, however I have failed to make a single contact. After carrying the radio, a mast and antenna up a (hard hike) hill and failing to make it out, I started doing tests.

Yesterday I bought an oscilloscope and dummy load at the local Hamfest.

The attachment with decent sine modulation is from my Icom 746Pro as a test, with Volts/div set at 1V, and low low power. The second image is the (tr)uSDX connected in exactly the same way, but with the volts/div set to 50mV for better visibility. This is the only change in settings or setup. Even the same coax cable.

I have tried re-flowing solder joints that looked questionable. I tried wiring up an external mic to see if the capsule was the issue. The VFO and TX do seem to work to some degree, as they generate a change in the display on the scope, and the Icom (which is 60cm away) picks up a strong interference signal when I hit transmit into the dummy load. I have tried SSB, AM, FM and CW, the scope sees them all as CW.

So my very ignorant deduction is that the finals are working, but TX modulation is not.

Any input is welcome.

73.
Attachments
IMG_20221204_194755868a.jpg
IMG_20221204_194755868a.jpg (44.66 KiB) Viewed 1893 times
IMG_20221204_194556273a.jpg
IMG_20221204_194556273a.jpg (43.42 KiB) Viewed 1893 times
dl6sez
 

Posts: 399
Joined: 30 Dec 2021, 22:54

Re: No transmit modulation (I think)

Post by dl6sez »

Hello OM,

if you input a sine wave in an SSB Modulator you will get a sine wave of your tuned tx qrg -fsine or +fsine depending on choosen side band.
So what you explain for your (tr)uSDX signals always looking like CW eg. a sine wave.... that's what to expect!
I don't understand what the ugly waves on the second oscillogram are? Is this your Icom trx? Maybe at very low low power, as you have written, there is no real good SSB Modulation on your Icom.
It also has to show a super nice sine of trx qrg + or - fmod when modulated with a sine!

73 de Chris
Chris DL6SEZ, JN48XL near Ulm, Southern Germany
kk7bco
 

Posts: 8
Joined: 05 Dec 2022, 03:32

Re: No transmit modulation (I think)

Post by kk7bco »

Hello Chris,
Both pictures of the scopes are of me inputting a tone generated with my voice into the mic, not a direct tone injection to the SSB modulator. Sorry if I did not explain this well enough.

The setup is: vocal "tone" fed into mic attached to transmitter. Scope connected to dummy load on the back of the trasmitter.

The Icom shows what I would expect, as different tones make different sines on the scope. With the (tr)uSDX I just get Null or the solid band in the upper picture.

My current level of experience does not know what "trx qrg" means, and google is unable to help me. This makes me believe that I am also not fully understanding your answer.

NOTE:
All I am trying to do is make sure that my (tr)uSDX is working on transmit. Since I have been unable to make a contact and have failed to hear myself on a webSDR I assume that I have a problem, but I may be wrong, or course.

73
Constrainted
 

Posts: 56
Joined: 20 Dec 2022, 20:51

Re: No transmit modulation (I think)

Post by Constrainted »

Did you find a solution? I think my radio developed the same problem after two weeks of successful use on SSB and CW ( for tuning, SWR, PWR meter.) From the main board, C13 seems OK, not shorted. I could not find C14, but there is an unmarked surface mount component near Q1.

The power out corresponds to, but not exactly, the power indicated on the radio.

73 de Constrainted
Constrainted
 

Posts: 56
Joined: 20 Dec 2022, 20:51

Re: No transmit modulation (I think)

Post by Constrainted »

Constrainted wrote: 02 Jan 2023, 12:54 Did you find a solution? I think my radio developed the same problem after two weeks of successful use on SSB and CW ( for tuning, SWR, PWR meter.) From the main board, C13 seems OK, not shorted. I could not find C14, but there is an unmarked surface mount component near Q1.

The power out corresponds to, but not exactly, the power indicated on the radio.

73 de Constrainted
OK. On the 40m section of the RF board, C39 is empty and C38 is shorted as measured with an analog multimeter. (The schematic shows C39 1n and C38 empty.)

73 de Constrainted
dl6sez
 

Posts: 399
Joined: 30 Dec 2021, 22:54

Re: No transmit modulation (I think)

Post by dl6sez »

Hello Constrained,

sorry but i have to comment that you are right, but your skills of reading and interpreting the (tr)uSDX circuit are not really good enough to help somebody here....
Otherwise you would have seen in the circuit that everything is okay and you measured it!
Paralleling Cs and a big DC short in the antenna path is what there should be...

73 de Chris
Chris DL6SEZ, JN48XL near Ulm, Southern Germany
Constrainted
 

Posts: 56
Joined: 20 Dec 2022, 20:51

Re: No transmit modulation (I think)

Post by Constrainted »

dl6sez wrote: 03 Jan 2023, 21:59 Hello Constrained,

sorry but i have to comment that you are right, but your skills of reading and interpreting the (tr)uSDX circuit are not really good enough to help somebody here....
Otherwise you would have seen in the circuit that everything is okay and you measured it!
Paralleling Cs and a big DC short in the antenna path is what there should be...

73 de Chris
dl6sez

Thanks for saving me from mistakenly replacing this capacitor.

I see now the path LPF_OUT3 takes through IM43 to GND. Why am I not getting modulation on 40m? Any ideas?

73 de Constrainted
dl6sez
 

Posts: 399
Joined: 30 Dec 2021, 22:54

Re: No transmit modulation (I think)

Post by dl6sez »

Hello OM,

i can only conclude when all is right with your measurements that maybe the input of the ADC in the Atmega328 is killed. No further diagnosis is possible here without measuring by myself.
Maybe your seller killed the input in advance.... CW works without ADC. You will everytime see a sine output on this rig also on SSB, because it is doing it different than other standard SSB modulators...

73 de Chris

P.S.: I would reprogram the firmware than reset it and if it does not help change Atmega328, but only with a real, so not cheap today non chinese Atmega328PB.
Chris DL6SEZ, JN48XL near Ulm, Southern Germany
kk7bco
 

Posts: 8
Joined: 05 Dec 2022, 03:32

Re: No transmit modulation (I think)

Post by kk7bco »

Hello,
sorry for my late update.

I find the schematics difficult to read, and ultimately kept falling back on the PA being a potential issue. I found a batch of BS170 made by Motorola (USA) on eBay. Removing the FETs I found it impossible to clear the Source hole on the board - I assume it is multi-layer on that hole. I also tested that one of the three BS170 was reading differently, so I fear I either damaged it with a bad antenna connection, or by overheating it with the iron.

I decided to futureproof the problem by socketising the FET PA stage, using some low profile header sockets. Notice that I also re-wound T2 with the 3/21 ratio that DL2MAN recommends.

I am now getting good SSB modulation on the scope. Looks like the maximum power output I am getting with 11.1v is around 2.8w, which is not enough for me to see on the local web-SDR using the dipole on my roof.

The new FETs stick up a little more, since I added the sockets, which means they just touch the base. My housing is made from PLA, and in my brief CQ calls watching the scope, they generated enough heat to deform the base. So my next step will be making a new case which has a heatsink in the base (though just making an aluminium base plate would work, and adding some heat compound to the tops of the FETs). I have not put a thermometer on them, but they are very hot to the touch.

73
Attachments
IMG_20230108_130758040a.jpg
IMG_20230108_130758040a.jpg (60.11 KiB) Viewed 1646 times
s57l
 

Posts: 56
Joined: 30 Dec 2021, 21:30

Re: No transmit modulation (I think)

Post by s57l »

If the output filters are properly tuned, fets will generate almost no heat.
You do not need a heatsink or other case, you need to tune the colis.
You will get close to 5w and sockets should not cause any problem.
And as long as your soldering iron is strong/hot enough, removing fets is quite easy.

73 gl
dl6sez
 

Posts: 399
Joined: 30 Dec 2021, 22:54

Re: No transmit modulation (I think)

Post by dl6sez »

Hello OM,

good to hear you found the real error.
But what i can't believe is that this FETs are from Motorola...
If they do no problem, but i would only buy semiconductors at serious suppliers like mouser or farnell or other another big distributor....
As the OM S57l before told you, FETs don't get hot if tuned properly, but i hope you don't have an additional problem because i never had hot FETs, also not priar to tuning the lpfs... bad copy of FETs maybe...

Please have a look at your PA bias max it can make Fets hot it has to be at 128 for BS170 and PA bias min at 0..... These bias values are not comparable with classical PA drive.

73 de Chris
Chris DL6SEZ, JN48XL near Ulm, Southern Germany
kk7bco
 

Posts: 8
Joined: 05 Dec 2022, 03:32

Re: No transmit modulation (I think)

Post by kk7bco »

I will look at the filters, and get back to you. Making an alluminium back plate was easy for me, and I actually like it.

More soon,
73
kk7bco
 

Posts: 8
Joined: 05 Dec 2022, 03:32

Re: No transmit modulation (I think)

Post by kk7bco »

I need to close this thread.

My conclusion is that one of the original BS170s got damaged. The replacement Motorola units behaved strange and overheated. I got some Fairchild FETs and they ran cool again.

I was still getting some strange behaviour when trying to run FT8, but SSB looks normal, so I consider the board to be (almost) fixed.

IF YOUR (TR)USDX SEEMS TO NOT HAVE SSB MODULATION but is still transmitting something, and looks fine on CW, there is a chance that you have a single FET that is not working correctly. When I would listen on a nearby radio with just a chunk of coax for an antenna, or a dummy load, transmitted audio just sounded distorted and compressed at peak level.

I hope this can be of some use to someone else.

73
dl6sez
 

Posts: 399
Joined: 30 Dec 2021, 22:54

Re: No transmit modulation (I think)

Post by dl6sez »

Hello OM,

your conclusion is totally wrong.
Not good is the additionally extra capacitance and not good contact resistance ( in rf other than in dc ) due to your sockets used for the BS170s. An aluminium backplate as near as the normal 3d printed plastic to the toroids will make an effect due to eddy currents and change the inductance and losses....
Distortion in nearby rx has nothing to do with broken FETs.
Motorola does not produce semiconductors since many years ;) so the BS170 have been chinese ones...

73 de Chris
Chris DL6SEZ, JN48XL near Ulm, Southern Germany
DL2MAN
 

Posts: 706
Joined: 30 Dec 2021, 19:18
Contact:

Re: No transmit modulation (I think)

Post by DL2MAN »

dl6sez wrote: 12 Feb 2023, 19:21 Hello OM,

your conclusion is totally wrong.
Not good is the additionally extra capacitance and not good contact resistance ( in rf other than in dc ) due to your sockets used for the BS170s. An aluminium backplate as near as the normal 3d printed plastic to the toroids will make an effect due to eddy currents and change the inductance and losses....
Distortion in nearby rx has nothing to do with broken FETs.
Motorola does not produce semiconductors since many years ;) so the BS170 have been chinese ones...

73 de Chris
Sharp as a razors edge. Thanks Chris !

73 Manuel; DL2MAN
Post Reply

Users browsing this forum: Bing [Bot] and 2 guests