[jjfritz5528]

Right up front, sorry for all the information in this post, but wanted to provide as much as possible.
I just finished getting my Model 512 working, thanks in a large part to Ron's updated schematic!
I was planning to test it with the matching speaker, but it shown open, so I used a Peerless 7-A, with good success.
I believe that this Philco Speaker is a Timmons Talker Metal case Cone speaker in disguise. It has the Timmons name stamped on the brown laminate in the picture. The speaker looks identical to the one in Buford's Classic Cones, without the red paint. It is also shown on the RadioMuseum.org site for the Model 512.
As I dug into this, I couldn't make sense of the wiring, although it is pretty simple. (A Note here, there was a sticker from 1932 indicating a Radio Shop did some repairs to this set, on the inside of the lid to the radio. Not sure what was done.)
I am assuming that the black cylindrical device is a transformer. I read 238 ohms across the top 2 pins and approx. 0.3 ohms across the bottom two. This doesn't make sense to me, since the speaker cord is connected to the top and bottom pins on the left side. I get no resistance readings on any other combination. i thought 238 ohms seemed too low, expecting something in the 1k range. AND based on the reading, I would have expected the speaker cable to be connected across the top terminals of the transformer. Any thoughts on what the resistance/impedance ratio should be?
I also don't get any reading on the 2 terminals at the right, on the brown phenolic. I'm assuming there is a coil inside there, but didn't want to start tearing into things before I get some info from the experts. If it is a coil, any thoughts on rewinding?
Any ideas? Thanks in advance!

[MrFixr55]

I believe that Philco may have bought the company that produced the Timmons Talker speaker.  The can may be an LC network.  Attached below is the schematic for the 1928 RCA 100A Speaker.  The filter almost looks like a Pi Network.  I do not know what the expected roll-off frequency was supposed to be.

RCA 100A Speaker.pdf Size: 297.37 KB  Downloads: 114

I will look to see if there is a schematic for the 512 speaker.

[jjfritz5528]

Thanks! I didn't see the attachment?!

[MrFixr55]

Hi John,

Just put it in. Sorry, I wrote in "Quick Reply", which does not allow for attachments. I had to do a Full Edit. It is there now. Hope that helps.

Interestingly, output transformers were not popular with radios of this vintage. These early cone speakers had high impedance, high resistance electromagnets (as opposed to the much lower impedance voice coil in the dynamic speaker. In battery sets, the speaker was placed directly in the plate circuit of the output tube. However, in your 512, the RCA Radiola 18 and similar Atwater Kent models of the time, LC coupling was used. A choke was in series between the output tube plate and B+. A 0.5uF capacitor was connected between the plate of the output tube and one terminal of the speaker. The other terminal of the speaker usually connected to the filament center tap either directly or through a resistor.

Hope this helps.

[RodB]

Yes, MrFixr is correct about the speaker being connected to the output tube via a coupling cap. I restored a 514 a few years ago and discovered the impedance of those old speakers is 2k-2.5k. The speaker on this one was in bad shape so I used a two watt universal line transformer which gave 2k pri to 8 ohm sec. Connected an 8 ohm modern PM speaker and it worked great. Even mounted great in the old enclosure.

[MrFixr55]

Hi John,

Looking at your post and the description, it would seem that the cylindrical device is not a transformer, but the filter that I described in my previous post (See the RCA 100A Speaker schematic). Seems to me that the bottom 2 terminals on the cylindrical device in your post are internally connected together. There is likely a choke coil between the 2 top connections and either one or 2 condensers between either or both top connectors and the bottom connectors, in the Pi network that. The square device on the right is the magnetic reed "motor" (That's what they were called) that drives the cone. normal resistance across the 2 terminals on the motor should be between 500 Ohms and 2K Ohms. I would not be surprised if it were open; I have a whole collection of headphones and an AK Model E speaker with open coils.

There are experts here who have rewound headphone and magnetic speaker coils. These are not exactly fun as there are many many turns of very fine wire. Disassemble the motor and carefully remove any insulating paper and look for corrosion. This is where the break will likely be. If the break is near the surface, then it is easily fixable. However, if it is all the way near the core, then if you can't remove the form, you will need to rewind or replace it.

[MrFixr55]

Hi Rod,
I am a fan of using a PM Dynamic speaker and an AA5 transformer for these old radios. The weak link as far as fidelity is concerned is the magnetic speaker and the grid leak detector. I have hooked up an RCA 100-a speaker using a transformer in reverse to a hi fi amp and the sound is about as poor as when used on the same vintage radio. There can be many reasons for this including play in the motor, etc, but even in good shape, these speakers do not reproduce music well.

However, for historical accuracy, it is nice to have a working period-correct speaker.

[jjfritz5528]

MrFixr55-
Thanks for all the good information. I know that the "motor" is open. Since this is my first venture into completely disassembling the speaker, I was treading slowly and looking for experts opinions.
I have re-wound a few Philco Oscillator and RF coils, but as you mention, I suspect this wiring would be even finer (smaller gauge). I am also aware of the dasteredly green area on the coils!
I wanted to try to save it's originality, but may have to fallback to a PM speaker replacement internally.
I've set it aside for the moment, but will get back to it and start looking into the motor.