Atwater Kent Model 35 Troubles

[Winky Dink]

A few days ago I finished the restoration of an adorable little Atwater Kent Model 35.


https://flic.kr/p/2iqxwk1

I chose this model because (1) It fits between my Model 30 and Model 37, (2) I have spare parts to fix almost any electronic deficiency, (3) The shape and the upside-down chassis are somewhat unique, and (4) Most of all, because the TRF is simple enough even for me. Or so I thought.

When I finished the chassis restoration, it looked great:


https://flic.kr/p/2iqAaCr

And it performed well--the best reception and the best audio of any of my radios from that era. And it was loud--had to shorten the antenna to less than 12 inches to have a satisfactory range of volume. Here's a sample of the audio:


https://flic.kr/p/2iqxx2b

But, when I reinstalled it in the cabinet, it didn't work right. Took it out, and it still didn't work right. Here's a before/after audio sample:


https://flic.kr/p/2iqBfHL

I've checked and rechecked all of the components, continuities, capacitance, resistance; retested and juggled the tubes (six '01A's), and in desperation I switched power supplies and speakers. All to no avail.

Here's a couple of schematics which I used. The one labeled "Model 35 Only" is my modification of the Rider/Beitman schematics to show the Model 35 with only one filament rheostat. The Gernsback schematic is original, showing two rheostats.


https://flic.kr/p/2iqQCja

In addition, Leigh Bassett's modern drawing is at http://www.atwaterkent.info/TechData...wingIndex.html

So, experienced people, do you have any advice for me?

Thanks,
Henry

[Gregb]

Double check the C- voltage on the output tube.

Gregb

[Winky Dink]

This is not a double check, but a first check. Here's the ARBE III setup:



Results (I hope one of these is the right way to measure grid bias):

From grid of the output tube to F- (A-/C+) +20V.
From grid of the output tube to F+ (A+/B-) +27V.
From grid of the output tube to chassis ground +24V

Thanks, Greg.

[jr_tech]

Possibly the 22v and -4.5 v supply connections are swapped?

jr

[Winky Dink]

Couldn't have swapped connections inside the ARBE, and the battery cable is not connected to C-22V or B+22V. Of course, I may have reversed the meter leads. With the battery cable totally disconnected from the ARBE, the C+ voltages (with the positive meter lead on C+) are C-4.5 -20v, C-9 -21v, and C-22 -34v. The A and B voltages are OK. Could I have done something to the power supply after the radio performed so well on its first test? I dunno. I changed power supplies, and it still didn't work.

[jr_tech]

Are the C voltages on the second supply also way out of specification?
I guess that I would try 3 flashlight batteries in series for the C supply and see how the radio works.

jr

[Winky Dink]

Good idea on the flashlight batteries. I'll try that, but first I'm going to check out my two K-101A battery eliminators along with Models 20 and 30 which are basically the same as the 35. If I can get a for-sure good power supply, then I try the 35 again. Thanks.

[Winky Dink]

Thank you Gregb, thank you jr tech.
This morning I set up the radio with a K101A Battery Eliminator adjusted for appropriate voltages. The radio worked normally. Not very good, but normal. Then I went back to the ARBE III supplemented with three D-cells for the C- voltage. It worked well.



Not an elegant solution, but it worked and it's only temporary. Now I can finish the project.

- Henry

[Gregb]

Good to hear you got it going again.

Gregb

[jr_tech]

It might be interesting to measure the current that the radio draws from the C supply.... it should be very close to zero. If it is high, that might explain the power supply failure(s).

jr