[davemc]

I am working on restoring a 37-604. The schematic is not in Nostalgia Air, but the 604 is:

http://www.nostalgiaair.org/PagesByModel...013796.pdf

The filter choke, part 59 in the 604, has been replaced. The original was likely 50-60hy, 2250 ohms DC. The replacement is probably about 4-5hy and 235 ohms. The choke only feeds the 6A8. A close replacement is available from Hammond/AES etc.: 155C, 60hy, 8ma, 2750 ohms DC.

Since the original is gone, and a good replacement is installed, would it not be possible to simply add a series resistor to total about 2250 ohms and add a LARGER filter capacitor? The original was only 8uf. Since I have to recap anyway, why not just use a 22 or 47uf 160 volt?

I'm not sure what the original function of this choke was, but not may radios seem to use one in this position. Perhaps it has something to do with reduction of oscillator flutter or modulation hum?

Opinions welcome!

[davemc]

Since I could not get an answer either here or on ARF, and since this was driving me crazy (plus I wanted to finish the restoration), I decided to do some digging. Sure enough I found the answer in my Radiotron Designers Handbook (3rd edition).

It turns out that the 6A8G tube, which set uses, is VERY susceptable to frequency changes when the supply voltage changes. There are other tubes which are not as susceptable, such as the 6K8. The result is "flutter" or motorboating. To reduce this tendency, the book says that it is recommended that the anode-grid supply be taken directly from the rectifier cathode through a dropping resistor and high value capacitor to B-. This reduces "flutter" and reduces hum (modulation).

Since this is an AC/DC set, with low B+, I suppose Philco decided to use a choke instead of a high-value resistor.

I left the replacement low-resistance choke in place and increased the filter capacitor to 22uf. The radio works, but I cannot tell if "flutter" has been reduced.

[davemc]

UPDATE: I just started on a Philco 55. It uses the same type choke to feed the 6A7 oscillator "plate". So apparently this is not that unusual. I had not seen it before, and now I find it in a second set - also a relatively inexpensive 5-tuber.

[davemc]

Correction: Philco 66!

[BDM]

Got me, but interesting find non the less.

[Texasrocker]

Sometimes, "experimentation" brings very-fruitful results! As Im sure your already aware, a old-timer vintage radio repairman taught me many yrs ago, it is always OK to increase your 8 uF type caps to higher uF ratings if necessary, just as long as you use the proper voltage ratings required and/or increase them also. The more "uF's" ,... the lower the hum-level will be! That can be a very-good thing in quieting-down a vintage set for improvement, than actually following orig schemats and parts lists! Remember, back in those days, the mfgrs line workers many-times "chunked-in" whatever parts would work that was avail on the assembly lines ( in stock at the time of orig assembly),..that was even "close" to recommended specs in orig schemats "uF's",.. just as long as the voltage ratings were correct! I have also purchased chokes from AES in the past and they always worked great when needed. Perhaps Ron or Chuck will help more with exact details specifics of your restoration. Good luck on your restoration!!

[exray]

Part of the logic here is since the B+ is so low the total current draw thru a single-choke, like a 1-2K speaker field coil, would cause an unacceptable amount of voltage drop and swing. I suspect thats the flutter being referred to.

This 37-604 is a good example. Its a small radio with a small speaker and that precludes having a big honking field coil. Note that the speaker FC is 2400 ohms. That says small wire. While the inductance *could* be good for filtering the voltage drop would be too much so they used it as the speaker field and nothing else. In essence it provides no substantial function in filtering as is typically expected. Moreover it doesn't drop voltage to the rest of the set.

The 2250 ohm choke feeds the 6A7 and the 78 screen. They don't draw any substantial current nor do they fluctuate in current draw so the voltage drop isn't significant.

The 400 ohm choke handles the output stage and the 78 plate. While the 78 IF amp doesn't play much of a role the 43 output certainly does. The 2250 ohm drop wouldn't be suitable there.

So its a creative juggling act to balance the current draw/voltage drop accordingly. The tubes are being run at their bare minimum voltages in this scheme so that was their way of distribution what little voltage they had to work with. Just another way to skin the cat.

[Ron Ramirez]

Many 1937 Philco models used a similar circuit to distribute B+ to the 6A8 oscillator "plate" as well as screen grids of the RF, IF and 6A8 tubes. Only instead of a choke, they used a resistor!

The 37-620 is a good example. A 9000 ohm resistor is used , being fed raw B+ from the junction of the 5Y4G filament, input filter (62), 12 uF, and the input of the field coil. The other end of this 9000 ohm resistor is connected to the positive end of a 16 uF electrolytic ( 18 ) and to a voltage divider circuit, feeding the points mentioned above.

Older Philcos, such as Model 60, did not use this separate feed for the detector-oscillator tube's oscillator "plate" and these work well, without modulation hum. Yet Model 66, which is similar to Model 60, DID use the extra choke to feed the 6A7.

I tried an experiment once, many years ago...I hooked up a Model 66 so that its B+ distribution was identical to Model 60, eliminating the choke. It had unacceptable hum. I then put the circuit back to its original design, and the hum disappeared.