[fields 100]

I am looking for for such set. Other than looks which I like, how is it performance wise?

[Steve Chambers]

I'm working on one right now. Electronics are done and it plays nicely. It has a tuned RF stage and push pull output. Plenty of volume and quite sensitive on the two bands. Has shadow meter. Careful with the schematic in Riders. Dots are missing that should be there to indicate a couple of connections.

[SteveJB]

I restored a chassis (have a rough cabinet) and love the set. The only issue I had was a worn tuning gear. It is some type of fiber that wears down over time.

I fixed it by removing the fiber wheel with a piece of rubber dial cord material and made an o ring connecting the two pulleys. It plays well and I love the look.

Now if I could repair the cabinet....

[Arran]

In terms of servicing, DO NOT power it up with the speaker unplugged, unlike most sets the B+ is not completely disconnected from everything with the speaker removed. As a result you will end up with an elevated B+ voltage going through several parts and potentially burning them up.
Also there is a 15 - 20 watt resistor that goes between part of the B+ circuitry and ground, this is what used to be called a bleeder resistor, it's very common for this resistor to fail. I've never had my Canadian 3118B working as yet but from what I have heard the 18/118 series chassis are a very efficient design and a good performer. However the 118 is an updated version of a model 18, so I think that's why is has some archaic things like that bleeder resistor and a rather unsafe B+ supply.
The Philco service diagram also has some errors in it in the power output stage, I think it relates to how the #42 driver tube is connected, refer to the diagrams with this same power output stage for reference.
Regards
Arran

[morzh]

If the bleeder is #60 (a big long wirewound), same one I had in 18. Not very elegant solution to lower B+, also increases current through the field coil and the filter choke, which does not add to longevity...

[Ron Ramirez]

Mike (Morzh), Arran,

Could either of you suggest an improvement to the power supply that would eliminate the bleeder? I'm not being sarcastic, really. I think it would be great for folks to see options in case they have one of these sets with an open bleeder resistor.

[morzh]

Ron

The way this radio is, unless the tubes will behave at higher voltage, the way it is now is the way.

If you look at 18 vs say 38-15, which have very similar tube Lineup in the receiving-detecting part, the sole difference s the Push-pull 42 stage in the 18 vs single 41 Class A in 38-15.

So they gave the Push-Pull 280V P-K whereas Class A needs only 200V. So they need to lower B+ for the rest of them in 18.
Hence the bleeder.

This said, the options were:

1. They could have used a separate winding on the transformer, better yet - bucking winding. Needs a second rectifier.
Expensive, costly.

2. Use a speaker with higher Field coil resistance. Possible if it exists. Still needs a bleeder but will dissipate less. A model H speaker has 3,200 Ohms.

3. Could use the lower voltage on the output tubes.
This is interesting as 42 tube shows 250V recommended B+ and so do other tubes in this radio, so I think, it is possible to find a compromise voltage where the pushpull still works well (how much power do we really need) and the other tube are still comfortable.
This would require lower rectifier voltage.
Again, at this point, unless the power transformer is gone, and one is looking to replace it, not an optimal solution.

4. Increasing the bleeder alone so it dissipates less will possibly stress the tubes a bit more, and the bleeder is still required.

5. Using tubes in PP with lower Plate votage and different output transformer. Like 90 or 111 do.


Unfortunately there are two ways to reduce the DC voltage: divide it, dissipating power (so do linear dropout regulators and resistors) or convert it.

The latter option will require a DCDC converter and presents the problems of other sorts, the very first being necessity to create a solid state high voltage DCDC which also will be noisy.

It is not easy to change the existing design.
Also the changes will violate the "original" status.

–---------

What I would try.

A speaker used in 18 (for 118 cathedral might not be an option) is shown to have 1.11 Ohm voice and 1.14kOhm field.

An H speaker from 111 (or 95, or..) has close enough voice and transformer to be used with the same pushpull, but has 3.2kOhm field.
That can be combined with higher value bleeder that will result in lower dissipation.

The performance will have to be evaluated after that, especially audio.

[fields 100]

Arran,

Thanks for the info. I did look at the schematic and the bleeder resistor. Hopefully, the modern equivalent is available. It is for sure way to go.

[morzh]

There are some.

http://www.mouser.com/ProductDetail/Vish...mN6Q%3d%3d

(6 wks lead right now if you are not in a hurry).

http://www.mouser.com/ProductDetail/TE-C...fmWWaNY%3d

Stock, 6.8K but will do just fine (+5% difference).


http://www.mouser.com/ProductDetail/Vish...VAVfbIE%3d

Stock, 6.2K, I think is also fine (-5% difference).

I would probably used the latter as you run at higher Mains voltage so a little drop in value could be even welcome.

[Mondial]

There is a specific reason for use of the bleeder resistor in the 118. This radio uses a class AB triode output stage, and as such the plate current drawn from the power supply varies greatly with audio output level. Because the field coil of the electrodynamic speaker requires a fairly constant minimum level of current to maintain the magnetic field, the bleeder resistor was added to supply this minimum current through the field coil under all volume level conditions.

All the Philco radios using the triode connected 42 "Super class A" output stage also use this bleeder, 16B, 650 etc. As mentioned previously, there is no simple way to modify the circuit to eliminate the resistor unless you replace the field coil speaker with a PM type.

[Arran]

I was not suggesting that the bleeder resistor was good, bad, or unnecessary, just that they fail regularly in 18/118 set, and it was somewhat of an archaic feature for a set of that time. I did not know that it was related to the design of the power output stage but that does make sense, what I remember reading was that bleeder resistors were used for filtering in some early AC sets, which means voltage and or current regulation. What makes these a bit of a pain is that there were a proprietary Philco part, 7500 ohms I think, which is not a standard value of a wire wound, so you need to pair up two resistors to replace it. I suppose in practice it's connected somewhat like a Zener diode, but working out a Zener substitute might be a tall order. However I don't like using "Sand" in a vintage radio without a good reason, or unless there is no other option like with a dead selenium rectifier.
Regards
Arran

[Ron Ramirez]

Interesting. Thanks, guys, for your insight. I have often wondered if those bleeder resistors could be replaced with zeners. Although I know this was the best the designers could do at the time, the use of those bleeders seems rather inefficient to me (waste of power and heat).

Mondial, just out of curiosity, how would you modify the circuit assuming the original speaker was missing and you needed to use a PM speaker?

[Mondial]

Ron, without the field coil, you would not need the bleeder to provide the minimum field current. You would simply remove the bleeder resistor and replace the field coil with a series resistance of sufficient value to provide 200 to 225 volts for the tubes other than the output stage.

This would be similar to what Philco did in the 38-116, where the output tubes run at full B+ supply voltage and the other tubes operate at lower plate voltage supplied through series dropping resistors.