[jrblasde]

Good afternoon, folks:

    I am restoring a Philco 49-906 and have run into a minor roadblock. Namely, I cannot align the FM circuit in full because the screws to adjust the first FM IF transformer are stuck in place. Peering into the top hole, it looks like the screw head is stripped. Has anyone run into this before, and how have you resolved it? I can imagine I need to desolder and remove the unit, open up the can, and somehow file a new flat-head groove into the screw head. Has anyone worked out a less invasive solution?

This is a Philco 32-4257 part (first FM IF can, 9.1MHz) and is labeled as TC-300A and TC-300B on the schematic.

[slford310]

I have taken them apart and was lucky once to get the screw out, but normally I don't even try anymore. Just pick up another IF Transformer and swap it out.

[jrblasde]

That certainly is a valid solution. I’ve not seen any of this particular part available on eBay. Are the IF transformers fairly interchangeable, so long as the new one is also designed for 9.1MHz?

[slford310]

They can be, but some are designed for point-to-point hook ups and some for a PCB. You should be able to find one that you can make work. I have purchased many IF Transformers on eBay over the years.

[jrblasde]

I’ve searched the internet high and low, but it would seem that there are no more 9.1 MHz IF transformers available (save for buying another radio with one included). My thought is now to get a 10.7 MHz IF transformer and replace the capacitors to create a circuit that resonates at 9.1MHz.

[Arran]

Are these FM IF transformers slug tuned or are they the kind with trimmers? If the type with trimmers a small drop of penetrating oil may help get them out if the end of the screw isn't peened over or something. I replaced the trimmer screws in a Philco AM IF transformer and they were a very fine machine tread, I don't know what pitch, I pulled some replacements out of one from a scrapped set, the FM ones may use the same thread pitch as the AM ones, but a thread gauge would be useful for finding out. 
Regards
Arran

[Chas]

Can a pix be posted of the offending IF can along with dimensions?

Some technician may have these as spare parts but un-marked...

Chas

[jrblasde]

Yes I’d be happy to add some photos. Silly me; I should know by now to start out a post with some photos! The IF transformer in question is a Philco 32-4257. It is the first FM IF transformer in the Philco 49-906 (also in the 49-905). The base measures 3/4” by 3/4”, and is 1-7/8” tall. It has a resonant frequency of 9.1 MHz. It is a slug tuned circuit; the inductors are adjusted, but the capacitance is fixed. The unit in question is Z300 on the attached schematic. The primary and secondary impedance DC impedance are both 1 ohm.

[jrblasde]

After some further research, I'm onto a new idea I'd like to test out. I saw a few different styles of IF can rebuilds designed to fit onto printed circuit boards that can be fit into the can of the gutted IF transformer. In one design, a guy used two side-by-side axial-lead inductors for the two coupled coils. In another design, a guy had two stacked PCBs, each with a radial-lead inductor. The two PCBs were stacked so that the bottom one has the inductor facing upward, and the top PCB has the inductor mounted on the underside to face the inductor on the lower PCB. In either design, the inductors are fixed. A fixed capacitor is then added in parallel to each inductor, with a capacitance of less than but close to the value needed for the proper resonant frequency. And finally, a trimmer cap is added in series to increase the capacitance until resonance is achieved at the desired frequency--in my case, 9.1 MHz.

    This sounds like an interesting experiment! After all, what is a transformer, if not two coupled inductors? I think I'd like to try the radial-lead inductor method. I will start by sizing the inductors. Looking on Mouser's website, for example, I am reviewing the data sheet for a 33 μH inductor with a minimum self resonant frequency of 9.0 MHz. Next, I need to select the capacitor for the LC resonant circuit. If f = 1/( 2*pi*sqrt(L*C) ), then that means C = ( 1/( 2*pi*f*sqrt(L) ) )^2. Thus, I need a total capacitance of 3.045 pF. I could use a 2.7pF fixed capacitor, and add a 1 pF trimmer for adjustment.

    The only thing I will need to sort out is the bandwidth and coupling of these two resonant circuits. I should probably start out on proto-board to refine the Q value. The bandwidth of an FM station is 200 kHz (with 15kHz dedicated to audio), so I will need to find the loaded Q value of these resonant circuits to then determine the critical coupling.

    Any thoughts? Has anyone tried to redesign an IF transformer with much success?

[slford310]

I have not but I’m very interested in what you have suggested and the results.  Please post it all

[jrblasde]

I'd be happy to keep everyone updated. To put words into a picture, here is a sketch of a schematic and a physical depiction of what I have described. I believe I will give it a go.

[RodB]

I think you will have difficulty tuning to resonance. Using a mechanical variable capacitor that varies one pf will be very touchy. That's one reason why most FM IF transformers are inductor tuned. This will indeed be an interesting experiment.

[jrblasde]

That is a very pertinent consideration. After thinking for a few minutes, I may have a remedy. Capacitors in parallel are additive, whereas series capacitance is the reciprocal of the sum of reciprocals. If I need a total capacitance of 3.045pF to pair with a 33 μH inductor, I could alternatively use a fixed 2.2 pF capacitor in parallel with a series connected 1 pF trimmer cap and a 1 pF fixed capacitor. This will result in an adjustment range of 2.7pF to 3.3 pF. Still not as fine of a range as it could be, but it does narrow it down a bit. I could also try to select a different inductor in order to give myself a bit more leeway in capacitor selection. Perhaps a 22 μH inductor instead of a 33 μH one. This would allow me to increase the total capacitance ever so slightly.

[jrblasde]

I realized that I had made a mistake in my calculations. I would need a 3.045 μf (not picofarads) capacitor to pair with a 33 μH inductor. Thus, I could have a 2.97 μF fixed capacitance (a 2.7 μF in parallel with a 0.27 μF), in parallel with the series combination of a 0.12 μF fixed capacitor and a 100pF trimmer capacitor. The total range of adjustment would be 2.9701 μF to 3.09 μF. I feel that this will be a much more narrow range of adjustment, and resonance will be easier to obtain. I will just need to be sure to get capacitors with low tolerances.

[jrblasde]

Back at home for the evening, I've started modeling the circuit in KiCAD. I've attached my final schematic that I will use to order parts for testing. I mentioned it earlier on, but inductors come with a nominal self-resonant frequency (SRF, as most data sheets call it). I thought it would be appropriate to select an inductor with an SRF near to the resonant frequency I am striving for, but I was wrong. At any frequency above the SRF, and inductor will no longer behave linearly. You want to select an inductor with an SRF above your intended operating frequency. I was unable to find a 33 µH inductor with an SRF above 9.1 MHz, so I stepped it down to a 22 µH inductor. I found one with a minimum SRF of 12 MHz, which will work for my application. Thus, I need a total capacitance of 3.729 µF. I will accomplish this by using a 3.3 µF fixed capacitor and a 0.33 µF fixed capacitor in parallel, combined in parallel with a series connected 0.12 µF fixed capacitor and a 100 pF trimmer capacitor. See attached schematic for the general setup.

I will order some parts, and test this out on proto-board. Recall that I will still need to determine the appropriate separation of the two inductors in order to obtain critical coupling. And we shall see if I have refined the range enough to achieve resonance without a fuss. I should be able to adjust from 3.630099 µF to 3.75 µF, which puts the target 3.729 µF roughly in the middle of that range.