The PHILCO Phorum
Atwater Kent 35 - Printable Version

+- The PHILCO Phorum (https://philcoradio.com/phorum)
+-- Forum: Various and Sundry Categories (https://philcoradio.com/phorum/forumdisplay.php?fid=37)
+--- Forum: Other Radios (Non-Philco) (https://philcoradio.com/phorum/forumdisplay.php?fid=12)
+--- Thread: Atwater Kent 35 (/showthread.php?tid=24045)

Pages: 1 2 3 4

462ron - 462ron - 02-06-2023

Hi Chas, no metal at all in my bench. These aren’t binocular coils like in certain other models. I have loosened the screws on the two outer tuning cap pulleys so I could rotate each independently of each other, I’ve tried a multitude of positions on these caps after first doing the recommended AK syncing procedure. I will recheck the two detector bias resistors, 270 and 180 ohms. I replaced them with carbon resistors because the originals were open. I’m going to work on replacing the computer ps so I can get a good 6V or better. I might try the rejuvenation process of running the filaments a little high, although they tested good!

Ron

RE: Atwater Kent 35 - Chas - 02-06-2023

Ron,

Thanks for the pix, overall it looks good!. Yes, solenoid orthogonal placed RF coils, O.K.

I do see that it is the single cone bearing tuning condensers.

There IS an obscure issue with certain tuning condensers, not just A-K. Although I have not directly experienced it, I believe Jeff of ARF has. The mixed metals and individual pieces that make up many tuning condensers that are nut/bolt assembled corrode in between the bits, the plates, the spacer washers and any connections to the wired circuits. So effectively, the tuning condenser at RF is not presenting the required capacity to tune the BC band.
This would be an easy check with a grid dip meter into any of the coils to find resonance or other means.

None the less the fix is to totally disassemble the tuning condenser(s) buff the joints, all of them, including the spacer washers. Buffing the entire surface of the plates is not required. Then, carefully re-assemble.

I would consider this only after eliminating any other possible cause of poor reception as it is tedious. The rust on the stator mounts seems to indicate this corrosion maybe present. A check with an ohmmeter is inconclusive as it is a DC measurement. Overall, deteriorated solder joints of 95 years old can create a similar poor RF joint despite DC measurements with an ohmmeter.

Since the radio will work with just the detector, directly feed the antenna into the 2rd RF plate circuit after disabling B+. The detector tuner could be cleaned up first, then tested for efficiency, expect it to tune broadly but it should be more effective.

The full procedure for rejuvenation can be found at many sources however, I use the method found in documents of the 20's which is repeated here:
http://www.nostalgiaair.org/References/Articles/TheFlash/Flash08.htm

No matter who's procedure, do NOT connect the plate or the grid to anything during the process of flashing and ageing and do not attempt to measure emission at elevated filament voltage.

The reason: The thorium will be immediately stripped from the surface of the filament at the elevated temperatures of flashing and aging by potentials on the plate or grid. The Radio Fans restoration method the same applies that is why the B and C batteries are disconnected when in home use.

If your confident the emission measured by your tube tester is good then rejuvenation should not be attempted.
Initially, I used my tube tester to flash and age thoriated tubes. I used a tube base adapter (home made) in conjunction with an interposer (home made), of which connects only the filament to avoid any accidental connection to potential within the tester. It was tedious and required careful steps. The Jefferson tube rejuvenator solved all of that and has simply two steps flashing and aging, I do have to time the process...

The tube tester must test at 5 volts filament NOT 6...

It is also possible the RF stages are oscillating, since the OEM grid resistors should be 600 ohms, the current low value could be causing oscillation that would block any desired RF signal. One way to tell is bring an operating transistor pocket portable radio near any of the RF coils. turn on the A-K and then tune the transistor radio though the band listening for an abrupt "dead" air indicating oscillation on the transistor radio. The transistor radio could also whistle if it is tuned near a broadcast signal from hetrodyne. This oscillation should move when the A-K is tuned. Also, the A-K could pickup the oscillator of the transistor radio which could result in whistle on any received station or "dead" air in the A-K.

Further, when the A-K RF stages are working well, bringing a transistor radio near the detector coil and tune the transistor to the same station as the A-K the amplification of the A-K will bring the station in on the transistor radio quite loudly. There will be a slight drop in the A-K volume as the transistor radios ferrite loop absorbs some of the RF signal and slightly de-tunes the A-K.

A useful tool for a "sanity check".

Chas

462ron - 462ron - 02-06-2023

Thanks Chas! Yes this radio sat in a friends garage for years before he gave it to me. He was an oil furnace tech and it was given to him by some customer so who knows how long or in what conditions it was kept! The amount of rust on the frames makes your suggestion of a complete tear down and burnishing connections reasonable, certainly easy enough to do. The 350 ohm grid resistors do jive with the values given on the Atwater Kent info site. It’s amazing to me how crude, devoid of components and of simple design we are use to seeing in later superhet sets, and yet illusive as it is to get operational! I have already dug out a 13.8V power supply I had for repairing car radios and 8 track players when I was doing tv/radio repair back in the 70s and early 80s. I plugged it into my IsoTap adjustable isolation transformer and adjusted it until I got it down to around 6.5V output. It works great, tubes glow a little brighter, and eliminates all hash and noise I was getting from the computer ps I was using. I will try dismantling the tuning caps and clean them up, also try those steps you mentioned with a pocket transistor radio. Thanks for the suggestions, it’ll be a little while before I can get back with my progress!

Ron

RE: Atwater Kent 35 - Chas - 02-06-2023

You will be fine Icon_smile

If you need to get my attention you can either PM or use my email.

I jump into the Phorum almost everyday. Though I have a dozen Philcos on the shelf, silent, I am a fan of 20's battery radio.

Best!

Chas

RE: Atwater Kent 35 - Paul Philco322 - 02-06-2023

Oil burner guys, Edison guys, plumbers, all got to go in the "cellar". One of my prized sets is an RCA 56X5 my dad took from a basement, sat in our cellar for about 30 years, I found it and restored it, one of my faves.

Good luck with yours.

Paul

RE: Atwater Kent 35 - MrFixr55 - 02-06-2023

For years, Grandma promised me her Radiola 17. I loved playing with it in her basement. She had the Radiola 100A speaker, but removed from its enclosure, as the whole thing used to be in a high boy cabinet (aftermarket) Unfortunately, Dad didn't want me to take it when I was still living with my family. When I got married and bought my own house, and Grandma had to move out of hers, it was gone. She thought the oil burner guy took it. I went and bought one off eBay because I loved the thing.

RE: Atwater Kent 35 - MrFixr55 - 02-06-2023

Chas, that is great info, not just on tube rejuvenation, but using the transistor radio to check for oscillation in the RF Section (I have not had that issue but I have 1K resistors in the ones that burned out) and using the transistor to check for effective RF gain. (I have a Grundig digital all wave that I have sometimes used to pick up IF to check for a working converter.)

So far, I have not had a TRF give me fits, I guess I am lucky.

RE: Atwater Kent 35 - MrFixr55 - 02-06-2023

Ron great idea about using the 12V PS and Variac to get the 6.0 - 6.5V that Chas recommended.

462ron - 462ron - 02-16-2023

So I have now taken apart all the tuning condensers as Chas has suggested. The frames are a bit rusted and even some of the fins and spacers did appear to have some corrosion so I buffed all the fins where they would contact the spacers and buffed the spacers too. It’s all constructed with aluminum plates and spacers so no dissimilar metals. I hear no difference. I’ve cleaned all the pins on all the tubes and cleaned all contacts where the tubes plug in. I’ve scanned for spurious oscillations with a transistor radio, no improvement. The station I had on the transistor radio was the same one I can get on the AK and the transistor radio did get a bump in volume when held close to the AK and then goes back to normal after turning the AK off or turning the tuning knob. I even scoped looking for oscillations. Tried higher then  lower grid resistor values, same with grid leak resistor. Fiddled with synchronizing of the 3 tuning condensers and got the one station I receive a tad better. So now back to thinking a set of new tubes to try would be next but I’m not spending any more $$ on this set! I don’t know if it’s worth experimenting with changing from a grid leak to a diode detector with getting only one station that’s on the weak side.

Ron

462ron - 462ron - 02-19-2023

So some good news and not so good! First, after going through everything as mentioned in my last post, I got to looking it over with a more discerning eye and what I found was there was a black spot on the 1st RF transformer. I remeasured it and compared to the same transformer on my parts set and even though they checked out close, I decided to just change it. After firing it up and resynchronizing it, I now get a few more stations and on the best one I even had to turn down the volume some! So I went ahead with my original plan to change over to a diode detector in place of the grid leak. Well that threw off the tuning on the last RF stage to the point I only barely get one station after I synchronized again and it’s distorted more than before. I’m thinking the 3rd RF transformer isn’t a good match for this mod  Icon_thumbdown. Looks like I’m done. I have attached the original schematic along with my modifications!

Ron    

RE: Atwater Kent 35 - MrFixr55 - 02-19-2023

Hi Ron,

Totally love W3LNB's redraw. They are much clearer than A-Ks drawings.

One possible source of the distortion is that you are using a UX200 or 200A as an amplifier. They make horrid amplifiers, as the tube has gas. It is meant solely for use as a grid leak detector. Its claim to fame is that it is more sensitive as a detector than a 201A. However, 201s and 112s can be subbed as a detector without mod.

Per the RCA Receiving tube Manual RC13 (1937?), both the diode and grid leak triode detectors load down the previous RF Amp, thereby changing its sensitivity. I dunno if there is a difference on loading with the diode vs grid leak circuits and if that difference changes the resonant frequency of the tubed circuit. That has to be taken into consideration. Otherwise, the circuit looks great. Guess maybe it is better to try this on a 3 dial radio. Maybe restore the wiring and finish restoring your 35 (U r almost there) and wire up the 37 chassis as your test bead, using a 201 in the detector, biased as an amp, using 67V on the plate as well as using 67V on the RF Amp plates.

One thing that drives my interest in this experiment is that there are articles on the internet extolling the crystal or diode detector as a high fidelity detector. There are experimenters that discuss the biased detector (not so easy to do in a battery radio) as a higher fideity detector because of its supposed infinite input impedance.

462ron - 462ron - 02-20-2023

Hi John, I got to thinking that the reason the alignment went way off on the 3rd RF tank circuit could be that in the original factory version if you look both the tuning cap and the cold end of the transformer went to chassis ground, basically the two were in parallel. In my modification, tuning cap still had a chassis ground but now the transformer cold end only has an RF ground through that 100 pf cap, so now the tuning cap is in series with the 100 pf, throwing resonance out of whack. I redid my mod by putting back the ground to the cold end of the transformer thus restoring the original configuration. Then I ungrounded the cathode of the 1N34, and attached it to the 27K/390K divider. That brought the alignment much closer to where it was before. Still, the audio is weaker than the factory design and no real improvement in distortion. I did experiment with changing the original 25V supply to the 200A and found it actually sounded better by going lower, even works down to 9V. Next time I get to working on it I’m going to use your suggestion of swapping out the gas tube for a 01A. If that doesn’t cut it, I may just return it to factory design and call it a day! Thanks for your interest and suggestions.

Ron

RE: Atwater Kent 35 - MrFixr55 - 02-20-2023

Congrats on the progress so far. I read somewhere that the less voltage on a 200A the better. I read also that a 200 is more sensitive but less loud. That is strange, as the amplification factor for a 200 is higher than a 201. The spec for plate voltage on a 200is for 25V as a detector. For a 201 as a grid leak detector, the plate voltage should be 45V. Despite what might be put on the box, I can't find any specs for use of a '00 or 00A as an amp or for any other use than a grid leak detector.

You may also want to use a headphone coupled by an AA5 transformer, as a 201 puts out about 10mW of power. make sure that you are using the correct negative bias on the Output tube depending on the tube used and the B+ voltage used.

The first radios marketed by Westinghouse (Aeriola Sr.) and GE for commercial use, as well as the IP501 marketed to the US Navy used a single tube as a superregenerative grid leak detector. In the case of the Radiola II and III, the Westinghouse RC, the GE 1300 and 1400 pair (Radiola V) and others, there was a stage or 2of audio after the detector. What negated the need for RF stages was the regeneration. Regeneration was covered by the 1914 Armstrong patent.

There was very little use of diode detection till about 1933. An interesting departure is the Philco Model 3 auto radio, which used a 71A (of all things) as a detector, with the plate tied to the cathode as one element and the grid as the other, making it a diode. I wonder if no one used diodes until the Fleming Valve patent ran out. The history of RCA is the history of lawsuits, as they were primarily a marketing and patent holding company until they actually started manugfacturing radios.

462ron - 462ron - 02-21-2023

Interesting history John on these old sets and RCA, history I don’t know nor have ever read. I guess I really should bone up on this and the peculiarities about these early tubes that Chas mentioned as it is interesting and is relevant to my AK. Anyway, I don’t have any headphones but I am using an output transformer and speaker from an AA5. I also have an old output transformer lying around that I’m not sure what it came from as I think it was part of the stuff my dad had when he did radio/tv repair in the late 50s early 60s. I tried it and it seemed to be a tad louder so this leads me to wonder how much the distortion has to do with a mismatch of impedance here! As a last ditch effort, I did swap the 200A for a 01A but I didn’t notice much difference in distortion. Maybe this experiment might be better done on a superhet set with more amplification and known performance. I’m leaning towards reverting back to original factory design where I was getting a few more stations and let it be!

Ron

462ron - 462ron - 02-22-2023

Also, I forgot to mention, I don’t have any spare tubes for this set so when I swapped out the 200A detector and put in a 01A, I had to use that 200A in the socket that I stole the 01A from. I’ve already undid my mod and returned this set to factory original but one last thing I will try is to put the 45V to the detector that you mentioned when I have the 01A in there! I will see if that changes the volume or distortion.

Ron