Old parts from 100 years ago. Do capacitors & inductors age?

[nonost]

 Hi! I think that some of you may like to see this old stuff.

I'm dealing with this old speaker from the late 20s & early 30s. The speaker comes along with a "little" circuit and a transformer. Here the schematic:



The "little" circuit is a treble filter. Here's how it looks:



The capacitor reads near 8uF. Regarding the inductor, my meter can't measure Henries so I can't tell.

I've raised the tabs and desoldered the blobs in order to remove the top cover and take a look at it. Well, it wasn't what I was actually expecting...Here a pic:



There are plenty of legs there. It looks like there are 3 caps inside. I get three different readings: 4uf, 1.5uf and 1.7uf. The sum of the three kind of match (more or less) the 8uf value read from the outside (before removing the cover)

And this is the block's bottom:

I'm wondering if the 007 is for the cap (7uf) and the 002 for the inductor (2mH???). And the 450v for the voltage rating of the parts. It does make a bit of sense



There's a massive black dense goop filling the box. Something tells me that I'm not supposed to remove it. The thing is I don't know if the parts are ok or not. I don't know the kind of technology of that era. So I wonder if parts are out of spec or whatever.

About the transformer: the primary is 600ohms and the secondary 3ohms. I also wonder if it is ok or not. I can take a picture if you like!

[Rob Strand]

I'd be trying to match the pins up to the schematic.

I read 'grave' as low frequency.  The circuit is a low-pass crossover which sends only the low frequency band to the woofer.

First measure the DC resistances between the connection points.   That will let you know what connects directly to what.  it will also show where the inductor is in the circuit since that's probably going to be very roughly 2 ohms DC resistance.
It's will also tell what doesn't connect to what, or what connects via a cap.

You need to build-up the connection picture before you make cap measurements because you could be measuring capacitance *through* the inductor and that's going to give you "measurement" but not a useful one if you are trying to work out the part values.

I have little doubt measuring between points 40, 41, 10 will end-up matching the schematic.   Measuring across 40 and 41 should be the true cap value.  Measuring between 10 and 40 will give the inductor DC resistance and inductance.

If you go here, enter 22 ohms for the impedance.
https://speakerwizard.co.uk/2nd-order-butterworth-passive-crossover-calculator/

fc     C          L
650   7.9uF    7.8mH
2k     2.6uF    2.5mH
4k     1.3uF    1.3mH

Don't expect both L and C to match-up with the circuit, although there is a small chance.
The main point is it gives a good guide to the size of the values you should expect.
So the inductors are in mH - very common for crossovers, especially for 22 ohm (bigger L's and smaller C's compared to 8 ohm).

[nonost]

 Ey Rob!

Yeah. "Grave" is "low". Apparently is the same word in Spanish and French.

Before I opened things up I took measures and the inductor DC resistance was around 3ohms (between pins 10 & 40). Sadly I can't measure inductance. The capacitor was 8uF (pins 40 & 41).

What I wonder is if these were the stocks values or these parts are somehow defective.

I really appreciate your help, thanks a lot!

[Rob Strand]

The biggest mystery is the intended use of the speaker.  As an add-on woofer or as a stand-alone full-range speaker.

As an add-on woofer:  You would place this next to another full-range speaker (in a radio) and it would add more bass.
The filter would act as crossover.  There is no need for a high-pass crossover on the other speaker as it's expected
to have a poor bass response.   In this scenario an 8uF and around 8mH woofer inductor might in fact work.
When I see "grave" I'm thinking crossover.

Another use is the speaker is still operating as full range but the crossover acts as  "tone control" to knock out the highs.    In this case an 8uF cap is going to be too much.  You would expect a smaller cap and inductor.

If you knew the inductor value it would definitely help decide what a reasonable range of capacitance is, and then you could decide if 8uF is acceptable.

If you have a sine-wave oscillator and an oscilloscope or multimeter you can do some simple measurements and make a reasonable estimate of the inductance.   You don't need a precise measurement just a ball-park one.

With a multimeter set to AC voltage and a sine oscillator set to about 1V

Wire-up say a 100 ohm resistor and the inductor.


Measure the voltage across the resistor, VR
Measure the voltage across the inductor, VL
You want to adjust the frequency to make VR = VL  (note that doesn't mean VR or VL is half Vin)
- If VL > VR you want to decrease the frequency.
- If VL < VR you want to increase the frequency.
Once you get VR = VL
Measure the frequency 'f' using a multimeter or eyeball it from the sine-wave oscillator scale.
Calculate L = R / (2*pi* f)
Measure the resistor value for a little more accuracy.

There's many other ways to do it, like resonating the known cap with the inductor, either parallel or series,
however the above method can be done with just a multimeter (the meter doesn't even need to read accurately).

[puretube]

Odd looks: https://www.radiomuseum.org/r/philips_le_martien_2113.html

[amptramp]

Back in the twenties, many speakers were relatively high impedance so they could be connected directly to the plate of the output triode and the positive supply.



The speaker has a U=shaped magnet with a coil of fine wire that moves the cone through a wire pushrod.  This one is still in excellent condition and having a 1926 radio playing modern pop music is something its designers never imagined.  The rest of the radio works well and the power supplies replace batteries that were always a problem.

[nonost]

 
I don't know Rob...The thing is that the block is more complex than the one showed in the schematic. At least it's what I see. There are a few capacitors an at least two inductors. It looks that it isn't just an inductor & a capacitor in series, at least in my case. There's one additional connection that breaks the simple inductor & cap in series.

The inductors are 1r5 each one. You can see the three legs. The side ones (points 10 and 41) reads 3ohm, while reading between the middle one and sides says 1r5. It is in this middle leg where a cap leg meets and breaks the simple inductor + cap pattern. If I remove this connection it will end up like inductor + inductor + cap, which is the "same" as the schematic shows.

I can make the capacitance go from 1u5 to 13uf when reading at points 40 & 41 just by connecting or disconnecting some of the thinner legs to the inductors thick legs. The main capacitor legs are the two longer ones at the back.  It's a bit strange. I'm going to keep it as it was and try it out.

This speaker is a electrodynamic loudspeaker. I will upload a couple pictures later.

[Rob Strand]

The inductors are 1r5 each one. You can see the three legs. The side ones (points 10 and 41) reads 3ohm, while reading between the middle one and sides says 1r5. It is in this middle leg where a cap leg meets and breaks the simple inductor + cap pattern. If I remove this connection it will end up like inductor + inductor + cap, which is the "same" as the schematic shows.

The fact two 1R5 readings can add to give 3-ohm means there is either two equal inductors inside or a single inductor with a tap.   If you measure the inductance across each 1.5 ohms and then across the 3 ohm you can determine if there's two inductor or a tapped inductor.   Two separate inductors will give double the inductance across the 3-ohm points whereas a tapped inductor you would see upto four times the individual inductances.

It's becoming more clear the circuit isn't simple as the schematic.   That makes me think the intent of this thing could be to roll-off the highs, like a tone control except it used inductors and capacitors.   However at this point i'm still not overlooking the fact it could be crossover.

I can make the capacitance go from 1u5 to 13uf when reading at points 40 & 41 just by connecting or disconnecting some of the thinner legs to the inductors thick legs. The main capacitor legs are the two longer ones at the back.  It's a bit strange. I'm going to keep it as it was and try it out.

It's not strange to me since measuring the capacitance of a complicated circuit with many caps an inductors can be very hard to interpret.   Depending on how your capacitance meter works you can even get capacitance readings which don't match any capacitor values in the circuit - How?  well the inductances and other caps affect the overall *impedance*.  The cap meter measures impedance with the assumption there's only a cap there but the presence of the other L & C's changes the impedance and that changes the measured value - you can get cap values larger smaller than the real caps depending on how the circuit is connected.   Other cap meters might show less of an effect from the inductors because they use pulsed methods instead of impedance measurement.

It's pretty clear some of your earlier measurements could be measuring caps in parallel.    You might even be getting measurements of caps in series.    The problem is you don't know if the inductors are also present in these measurements.

You don't know what parts are connected to what inside.    You need to consider the caps, how many are there, and how they could connected together inside.

It's going to be very difficult to work out what's inside.   It's possible to guess a circuit then do a few measurements to confirm it but still some doubt.

If you have sinewave oscillator and an oscilloscope and wire it up like this,  (use R=4.7k)



but instead of the inductor shown you connect to two pins of the circuit.     

Monitor the voltage at the test point with the oscilloscope
while you manually sweep the frequency from say 20Hz to 20khz.
- Pure inductors will show a rising voltage with frequency
- Pure capacitors will show a falling voltage with frequency.
- Capacitors in parallel with inductors with show a rise in voltage upto some point
  then the voltage will peak at some resonant frequency and then start to fall.
  If you record the frequency it might be use later to calculate the inductance and/or capacitance.
- Capacitors in series with inductors do the opposite the voltage will fall then hit some
   minimum at the resonant frequency then start to rise again.

Repeat for many pairs of the leads until you can work out which connections are pure capacitors
and inductors.  Then from the resonant connection see if you can piece together what is in parallel
and what is in series.

The whole idea is to build-up a picture of the connections and the parts.   Then from that it might
possible to do some specific measurements to work out the values.

[MR COFFEE]

Have you looked at the possibility that the speaker has terminals for another winding besides the voice coil?

Many really old radio and organ designs didn't use a permanent magnet speaker, but, instead, used a second set of "speaker" terminals that connect to an additional coil (NOT on the voice coil former) that is energized with DC to create a magnetic field for the voice coil to interact with.

The electromagnet coil often served a dual purpose as a choke in the power supply. (In case you hadn't looked for that already)

The caps and inductors may serve a function not purely about "speaker crossovers" like those we see in more recent PM (permanent magnet) speaker designs.

[nonost]

I can't thank you enough, Rob, for taking your time.

Yeah, the circuit it's definitely more complex than the schematic. I need to get an oscilloscope right now, and also a new multimeter with inductance meter feature.

Hi Mr. Coffe. I think this is a more evolved kind of speaker. It's called electrodynamic:



The speaker is passive. You plug it into an amp and be happy. If I'm not wrong this kind of speaker was an improvement over the type of speaker you are talking about. It's quite similar to the ones with have nowadays. I have to take a couple pics.