[glyphmon]

Okay, so I finally ate enough ice cream and Trader Joe's milk chocolate peanut butter cups to get into a headspace to tackle this.

We still have the under-voltage situation at the Screen Grid (pin 4) on each of the 6A8 and 6K7.  Expected is 100V.  Tested has ranged from 61V to 65V.  R4 has been tested good and substituted with new stock also testing good, resulting in no difference.  Voltages testing as expected from the voltage dividing resistor and on most other pins (small voltages discrepancies to be discussed later, if necessary).

In summary from my previous post:
Testing without an antenna connected at all yielded approximately the same results as before: 175V coming from the R14/R15 junction up to R4; 63V between R4 and the Screen Grid pins (pin 4)
Testing with antenna connected and tuned to my AM transmitter: 187V and 91V respectively

Tonight, with one leg of C15 disconnected, receiving no signal:
Both 6A8 and 6K7 in: 175V and 65V respectively
With 6A8 only removed: 203V and 138V respectively
With 6K7 only removed: 177V and 66V respectively
With both removed: 220V and 219V respectively

So, my amateur brain says this final result with almost no difference across R4 is because there's no load?
And looking at the the result with only 6A8 removed, seems like there's clearly more voltage than just a single tube requires?  Makes sense since there are supposed to be two, but only one is installed.
And looking at the result with only 6K7 removed, it is almost the same as any other test with both tubes in, and C15 in or out.
Now, my amateur brain's conclusion would be that there might me something wrong with the 6A8, since whenever that tube is removed there seems to be an abundance of voltage, but I've also swapped the 6A8 with another non-tested 6A8, but it is from at least a semi-functional radio with no substantial difference.

Anyone care to chime in on what these results imply?
Would it be worth ordering a replacement 6A8 in case both of the ones I have access to may have both failed in the same way?
Does the fact that there's almost no difference in the voltages measured with C15 connected versus disconnected suggest I have a bad new capacitor?
Further diagnostic options?

Best wishes,
Roy

[morzh]

Roy

A suggestion.

One thing that is makes it a bit less easy to deal with is your referring to the tubes' pins by their number alone.
And the sch does not show them necessarily.

Were you referring to them by their function (P, K, Screen Grid, Suppressor, and such) it would be way easier. And then you could add the numbers, if you want.


Mike

[glyphmon]

Mike,

I have edited the post to say Screen Grid (pin 4).

Best wishes,
Roy

[morzh]

What seems strange to me in the first place is, why the voltage is dependent on your having antenna connected or not. This is a bias-former, the R14/15/16 part and it should not really depend much on your input signal.

[RodB]

Roy,

I'm a little confused on your voltage readings. In the first line above where both tubes are in, are you saying that the screen grid voltage on one tube is different from the screen grid voltage of the other?

[glyphmon]

Mike,

There really isn't much difference between with antenna connected and not.  Same approximately 35V - 40V discrepancy from expected occurs either way.

I mention specifically about the antenna being connected in case it is relevant, since the procedure for testing voltages specifies tuning the radio to BC 1000 or where there is no signal, and with the volume all the way down.  I have observed using that guidance that there is little practical difference between with the antenna connected and without, and so again, I mention it simply for transparency in the information I'm providing.

Given the meager difference in readings, I suspect it is not particularly relevant.  No signal should be no signal, right?  It is very good at not receiving a signal with the antenna disconnected.

Best wishes,
Roy

[glyphmon]

Rod,

The two voltages listed on each line are on either side of the R4 resistor as requested be tested previously.

The first (larger) number being on the side coming from the R14/R15 junction, the second (smaller) being between the R4 and the screen grid pins of the two tubes.  Since they are all connected, and the values at each screen grid pin are within 0.1V of each other and the number coming off of the near side of R4, I am reporting only a single value in each case for the screen grid pins and the side of R4 connected to them.

Best wishes,
Roy

[morzh]

I meant these:

-- Testing without an antenna connected at all yielded approximately the same results as before: 175V coming from the R14/R15 junction up to R4; 63V between R4 and the Screen Grid pins (pin 4)
--  Testing with antenna connected and tuned to my AM transmitter: 187V and 91V respectively

I would not consider the 12V difference from the bias-setting Candohm and 28V at the Screen Grids insignificant.
The difference is one thing but the voltage drops a lot.

[morzh]

One thing I've been meaning to ask, as I don't want to go back and re-read the whole thing:

1) Have you done the full recap as we discussed?
2) Have you tested all your tubes, and, if yes (and they tested good on a tester that tests a bit more than the emission), have you tried replacing them?


I probably should've stated that before, but the tube is the same as the capacitor, that is a component and should be confirmed good BEFORE you do any troubleshooting.
As well as most of teh components, especially those that are easily removable and tested.

[RodB]

I see it now. You are showing us the voltages on either side of R4. In that case the 6A8 screen grid is drawing considerably more current than it should. Usually indicates a short. Without the 6A8 in, there doesn't seem to be a short. Can you disconnect the screen grid of the 6A8 and get near normal reception? There will probably be lower volume.

[glyphmon]

Mike,

The difference in that case was not the antenna.  With the antenna disconnected, I have not been able to receive any signal, so far at least.

With the antenna connected, and not tuned to any station or signal, the readings are as they have been, 175V and 61V-65V.  With the antenna still connected and tuned to my AM transmitter, then we get the 187V and 91V.

It is the difference between tuning to a station or tuning to NO station (as the testing guidelines specify).

This data specifically was in response to MrFixr55's question from 5/14: "Before desoldering, try measuring the Screen (Pin 4) voltages while receiving a signal from your AM transmitter.  This should cause a high negative AVC voltage which should lower tube current.  This will be an interesting piece of info."

Best wishes,
Roy

[glyphmon]

Mike,

Full recap (excluding Mica) is done.
I do not have access to a tube tester.  Two tubes are new from Bob Dobush but neither of the 6A8 nor 6K7 were among the new ones.  I have attempted swapping from my 37-620 for its 6A8 and 6K7 and the test results are the same but admittedly have only tested those with both tubes in at the same time.  Though similarly "operable" I have no guarantee that the Philco's tubes are good either.

Rod,

So, completely desolder the connection for the screen grid on the 6A8?  I will try to do this test tonight.
Reception so far hasn't really been problematic, but I also don't have a great frame of reference.  There is only one nearby AM radio station which does come in a little and is understandable.  My AM transmitter comes in great.  And as before, I was able to tune into a Romanian station on shortwave.

In lieu of a tube tester, I'm not opposed to ordering a replacement 6A8, if that makes sense as a next diagnostic step.

Best wishes,
Roy

[morzh]

OK, still maybe a bit of too much difference.

The question Rod asked (and to which I also alluded in my post preceding his about the tubes' testing) still remains.

[glyphmon]

Mike,

Without a tester, the best I can do is to swap the tubes for a known good one which I am willing to order.  I will try to disconnect the screen grid on the 6A8 tonight to see if that makes a difference on reception as Rod has suggested.  I am tempted to order two since swapping in the 6A8 from my 37-620 did not make the problem go away.

Is a short on the screen grid of 6A8 a common failure?  It would seem odd otherwise that I would just so happen to have two that exhibit this problem.  I will double-check also that swapping just the 6A8 or just the 6K7 does not impact the results.

Best wishes,
Roy

[MrFixr55]

Hi Gang,

I have seen shorted screens but only in Pentode Output tubes like 41, 6K6, 42, 6F6, etc. this usually occurs when the plate circuit is interrupted. The screen then draws all the current, which it was very much not designed for. I have seen screens glow incandescent when the plate circuit of an output tube is open. I suppose that the same thing could happen to a6K7 or 6A8 but since the curent through a 6K7 is so low, I don't think that this failure is common. There is a bit more plate dissipation in a 6A8 but Idont think that the issue is common there either.

Rod, I went back and reviewed the thread. What is interesting is that with a strong signal (your AM transmitter which is giving a much stronger signal 10 feet away than the 50 KW blowtorch that is 50 miles away) is causing the screen voltage to go higher (but still not within spec). This is likely due to the AVC action. The stronger the signal the stronger (more negative) the grid bias. The more negative the bias, the less the current. (This may not be true in a class A amplifier or in a convertor .)The less current, the higher the screen voltage.

I wonder if one of those tubes is gassy.