Strange PCM3060 Failure Mode

General discussion and support for the Peaberry SDR V2.

Strange PCM3060 Failure Mode

Postby k1qn » Mon Jun 22, 2015 2:40 pm

This is a continuation of the "Fluctuating power" post. The problem is pretty much isolated to the PCM3060.

What I observe at this point (after much resoldering) is the TX power will be fine for maybe 30-60 minutes of operation in JT65. Then it will drop slowly to zero over about 2-3 minutes. In another 30 seconds or so, it'll return to nominal (1W).

All the DC and RF voltages are nominal (I have an RF voltage probe, but no 'scope.) The audio out of the PCM3060 falls in step with the RF output. I'm looking at R17-R20, (VOUTL-, ...VOUTR+). I think VCOM was following in step, but don't recall.

The question is, what's causing it. Could it be one of the 5volt lines to the chip? One of the grounds to the chip? Keep in mind, the chip has been resoldered several times, with the same or similar performance. Visually, it's a thing of beauty - but this doesn't always tell the tale.

I'm beginning to believe it's a bad chip. A connection in the chip may be heating, resistance increasing till it's open. Then it can cool down again. Either that, or a static buildup in the chip. Then it discharges. I'm an old EE from the vacuum tube days, so I'm not really savvy on the physics of the chips. These are just my guesses.

Has anyone seen a similar fault in the PCM3060?

My next diagnostic step will be to freeze the chip with spray solution when it gets to about 50% power and watch the RF power.
Al Gerheim, K1QN
Celebrating 50 years in ham radio!
k1qn
 
Posts: 90
Joined: Sun Oct 20, 2013 9:42 pm

Re: Strange PCM3060 Failure Mode

Postby kz1x » Mon Jun 22, 2015 3:55 pm

Al

The scenario you posit is rather unlikely. Not impossible, but, it's not where I'd look first. I didn't read anything you posted that would steer me to believe there's something wrong with that IC.

RF power comes from the working of the following ICs: U7, U8, U9, U10, and U14. All run from the output of U2.

The problem you are describing sounds a lot like a thermally related artifact on the output of U2.
Verify that C4 is correctly installed. Provide a scope-trace snapshot of both +5V lines while the problem appears.

Steve KZ1X
kz1x
 
Posts: 232
Joined: Wed Aug 07, 2013 1:28 am

Re: Strange PCM3060 Failure Mode

Postby k1qn » Mon Jun 22, 2015 4:59 pm

I'll check C4 when I get home.

All the DC voltages remain nominal as the power drops. The RF voltage to U8 remain nominal as the power drops. The only voltages to follow the drop in power are the audio voltage measurements at the PCM3060 ends of R17-R20. This leads to drops to U8, T3, and on out the amplifier.

I agree "a bad PCM3060" is unlikely, but we're on the dark side here... :-) One of the circumstantial datum is that when I resolder the PCM3060, the nature of the power drop changes (faster, slower, etc...)

BTW, I have a working Peaberry V2 (160, 80M), and I can compare as a last resort.
Al Gerheim, K1QN
Celebrating 50 years in ham radio!
k1qn
 
Posts: 90
Joined: Sun Oct 20, 2013 9:42 pm

Re: Strange PCM3060 Failure Mode

Postby kz1x » Mon Jun 22, 2015 7:43 pm

OK. Well, stranger things have happened but I think the soldering issue is a red herring.

I will be VERY surprised to find this chip is the culprit. I'd sooner believe the control program, a Windows utility grabbing the sound control and muting it, etc.
kz1x
 
Posts: 232
Joined: Wed Aug 07, 2013 1:28 am

Re: Strange PCM3060 Failure Mode

Postby k1qn » Mon Jun 22, 2015 7:53 pm

The other odd thing, which I forgot to mention, is that if I cut the 12 VDC input and turn it back on, the PB will recover, and go back to 1W RF out.
Al Gerheim, K1QN
Celebrating 50 years in ham radio!
k1qn
 
Posts: 90
Joined: Sun Oct 20, 2013 9:42 pm

Re: Strange PCM3060 Failure Mode

Postby k1qn » Mon Jun 22, 2015 8:16 pm

C4 checks out. U2 output voltage checks out while RF power is falling. I checked it from the "north end" of R30, far, far away from the chip itself. I try to never check right at the chip.

Checking the "hot ends" of R17-R20 is pretty convincing evidence that the output of the PCM3060 is diminishing with the RF output power.

Very strange indeed. (Why do _I_ get all the impossible debugging cases?) :-)
Al Gerheim, K1QN
Celebrating 50 years in ham radio!
k1qn
 
Posts: 90
Joined: Sun Oct 20, 2013 9:42 pm

Re: Strange PCM3060 Failure Mode

Postby k1qn » Mon Jun 22, 2015 9:33 pm

Software faults would probably be sudden and not related to operating time. The power diminishes slowly, taking 2-5 minutes to go from 1W to 0. I'm betting on the chip, and I'm getting ready to replace it. The freezer spray will be the final diagnostic item.
Al Gerheim, K1QN
Celebrating 50 years in ham radio!
k1qn
 
Posts: 90
Joined: Sun Oct 20, 2013 9:42 pm

Re: Strange PCM3060 Failure Mode

Postby k1qn » Mon Jun 22, 2015 9:46 pm

Is there any way the audio input could go silent between the USB port and the input of the PCM3060? The key here is that it diminishes slowly from nominal to nothing. It tales 2-5 minutes.
Al Gerheim, K1QN
Celebrating 50 years in ham radio!
k1qn
 
Posts: 90
Joined: Sun Oct 20, 2013 9:42 pm

Re: Strange PCM3060 Failure Mode

Postby k1qn » Mon Jun 22, 2015 11:22 pm

It may be cured. I'm running JT65, and I've transmitted on something like 15-20 consecutive "ODDS", and it's still up at 1W.

The last fixes were reflowing R48 and R49, which deliver set SCL and SDA from the 3.3volt supply. Also, all the pins on the USB connector, and the usual eye strain in the region of PCM3060.

If you hear no more, it's fixed. :-)
Al Gerheim, K1QN
Celebrating 50 years in ham radio!
k1qn
 
Posts: 90
Joined: Sun Oct 20, 2013 9:42 pm

Re: Strange PCM3060 Failure Mode

Postby k1qn » Tue Jun 23, 2015 8:52 pm

Nope!

After running just fine for a day, it went back to its old behavior.

Next question: Is is possible to remove the PCM3060 without a heat gun?
Al Gerheim, K1QN
Celebrating 50 years in ham radio!
k1qn
 
Posts: 90
Joined: Sun Oct 20, 2013 9:42 pm

Next

Return to Peaberry SDR V2

Who is online

Users browsing this forum: No registered users and 1 guest

cron