Missing info on schematic - could you wing it?

[ChristofferB]

Hi! I'm considering building a replica of a small direction finding receiver from the 1960's, link here:
http://www.cryptomuseum.com/df/fu303/index.htm

It's a fairly straightforward design, a tuned circuit with a myriad of plug-in coils for bands, and some AF stages, with the ability to inject 1000Hz into the rx, to listen for direction (assume sound dies out OR becomes loudest there).

Here's the issue: the nice schematic that's available from the manual is missing some values for some of the components. See attached.



Specifically, two transformers between AF stages, an output coil (choke?) to gnd, a coil in the 1000Hz gen, and potentiometer P3 in the 1000Hz gen.

The output choke is probably just filtering, that might be less critical. The unmarked potentiometer is just attenuating the output from the 1kHz osc. so that's probably not critical either. maybe 10K or 100K would do..

The trouble is the AF transformers. They look to be standard hi/lo impedance transformers, like the ones you see in telephone equipment. Could I just wing it, and use some of those? I've never seen it before, so I don't know how critical it is.
Maybe some home-wound 10:1 transformers on a bit of ferrite rod would do..

I think I'll settle on AC128's instead of the OC75 transistors. They're pretty similar, they're not made from glass, and I have some already..
Thanks in advance!

[oldway]

Specifically, two transformers between AF stages You can find such little interstage transformer in old AM 6 transistors radio's


an output coil (choke?) to gnd No need of this coil, use only a magnetic High impedance earphone of 600R


a coil in the 1000Hz gen, and potentiometer P3 in the 1000Hz gen.May use a primary winding of an interstage transformer.

The unmarked potentiometer: 1K Lin seems to be OK

Maybe some home-wound 10:1 transformers on a bit of ferrite rod would do.. No, ferrite can only be used for high frequencies

I think I'll settle on AC128's instead of the OC75 transistors. Without comparing datasheet, I would say AC126 instead of AC128, but I am not sure about it.

[ChristofferB]

Specifically, two transformers between AF stages You can find such little interstage transformer in old AM 6 transistors radio's

Thanks! that's a good idea!

an output coil (choke?) to gnd No need of this coil, use only a magnetic High impedance earphone of 600R

If I wanted to use a high impedance crystal earpiece like the original, would the coil be needed? I'd like to keep it at least sorta original looking.

[Alex Nikitin]

it is also worth noting that the capacitor in the middle of the circuit is marked incorrectly - it should be 100uF electrolytic, not 100pF !

Cheers

Alex

[ChristofferB]

I'd completely missed that! Thanks!

Could one make a good guess on the impedance, or ratio on those transformers? I might have something similar lying around.

[ChristofferB]

I've found a couple of transformers, that seems to have a ratio of approx. 1:200, based on resistance of the wire (prim. and sec. has same gauge wire). might they do?

[Ian.M]

I suspect the primary resistance is fairly critical.  Also the unknown resistor feeding the 100uF decoupling cap for the first two stages will have some effect.   As it uses collector feedback biassing,there is *some* stabilisation of the operating point, but I would recommend breadboarding the transistor to be used with a 100K feedback resistor and a 4.5V supply and determining what value resistor in place of the inter-stage transformer primary, will get the collector to sit at 2V. (The optimum operating point would be Vce=1/2 Vsupply, but above 1/2 Vsupply it wont be unconditionally thermally stable)   Try all the transistors and take an average.   The original OC75 has about 20% more gain than your substitute and it looks like the collector current was intended to be about  1mA to 1.2mA, which would require inter-stage transformers with primary resistances of about 2K.  In the absence of other evidence, try 5% of the required collector resistance for the decoupling resistor (10% drop as two stages are loading it).

A good reference book for Germanium transistor circuits is the "Mullard Reference Manual Of Transistor Circuits".  Its fairly easy to find online

[oldway]

I've found a couple of transformers, that seems to have a ratio of approx. 1:200, based on resistance of the wire (prim. and sec. has same gauge wire). might they do?

Ratio was more or less 20/1.
Why not modify this circuit and use PNP silicium transistors ?
This old germanium transistors are very fragile and sensitive to heat.

[ChristofferB]

I suspect the primary resistance is fairly critical.  Also the unknown resistor feeding the 100uF decoupling cap for the first two stages will have some effect.   As it uses collector feedback biassing,there is *some* stabilisation of the operating point, but I would recommend breadboarding the transistor to be used with a 100K feedback resistor and a 4.5V supply and determining what value resistor in place of the inter-stage transformer primary, will get the collector to sit at 2V. (The optimum operating point would be Vce=1/2 Vsupply, but above 1/2 Vsupply it wont be unconditionally thermally stable)   Try all the transistors and take an average.   The original OC75 has about 20% more gain than your substitute and it looks like the collector current was intended to be about  1mA to 1.2mA, which would require inter-stage transformers with primary resistances of about 2K.  In the absence of other evidence, try 5% of the required collector resistance for the decoupling resistor (10% drop as two stages are loading it).

A good reference book for Germanium transistor circuits is the "Mullard Reference Manual Of Transistor Circuits".  Its fairly easy to find online

Thanks, that sounds like the way to go, but finding transformers with that precise requirements might be difficult.
If the AF circuit was modernized slightly to Si-transistors, as oldway suggested, could the transformers be done away with entirely?

I just noticed from the hi-res photos of the board, that the 15K resistor in the detector circuit clearly is labeled "1.5K" so that must be a schematic error as well.

[Ian.M]

Modernising it will be near impossible without reworking the  design completely.   Do you have any idea of  the gain of the amplifier stages?  Silicon transistors tend to have much higher gain and a fraction of the leakage which invariably requires full rebiassing.  Also the 1KHz oscillator will probably have more distortion.   

There's really little point in redesigning it for silicon and to eliminate the coupling transformers - you might as well use an OPAMP circuit.  As long as Germanium diodes were used in the detector, performance would be similar!  However if you do go down that road, you'll probably need potential divider biassing, and to insert emitter resistors to throw away enough gain.  You might as well flip the rails and the diodes and use NPN transistors.  The oscillator may need redesigning for thermistor stabilisation.

How I'd fiddle with it to get something working - try Eagle LT44 interstage transformers which are still readily available as 'clone' parts.  Their primary resistance is only about 1K for the 'blue' variety and 500R for the 'yellow' variety, so a pair of LT44 blue, with a 470R decoupling resistor should be right in the ballpark.  Tweak the base bias resistors to get the operating point you want.

[ChristofferB]

Yeah, that sounds right. I think I might just build up the circuit dead-bug on copper clad, then it's then fairly easy to experiment, and then I might just resort to swapping transformers until performance is decent.

[Ian.M]

@ChristofferB: bump:  You replied while I was still posting - I accidentally hit post half way through the first paragraph and edited in the rest of it.



Personally, I'd use solderless breadboard, initially with a simple tuned circuit and diode detector (leave out the vertical OA81, 30 pF cap and 1.5K resistor till you've got the bias oscillator working) and get the AF amplifier working first.  The transformers I suggested have tapped secondaries so you can get a 4:1 (8dBV) variation in AF amp gain by tapping down if you need to.  To match modern headphones, you could use an Eagle LT700 output transformer in place of the inductor in the output stage collector circuit.

[ChristofferB]

@ChristofferB: bump:  You replied while I was still posting - I accidentally hit post half way through the first paragraph and edited in the rest of it.



Personally, I'd use solderless breadboard, initially with a simple tuned circuit and diode detector (leave out the vertical OA81, 30 pF cap and 1.5K resistor till you've got the bias oscillator working) and get the AF amplifier working first.  The transformers I suggested have tapped secondaries so you can get a 4:1 (8dBV) variation in AF amp gain by tapping down if you need to.  To match modern headphones, you could use an Eagle LT700 output transformer in place of the inductor in the output stage collector circuit.

Sorry! The forum usually alarms you of that!

The LT44 is available pretty cheaply, you're right. That's the plan, then.

Thanks for the help!

[Ian.M]

Do check the primary resistance when you've bought them as the wire gauge and resistivity used for Chinese LT44 clones will be all over the place - if its significantly under 1K, you will need to adjust the decoupling cap feed resistor.   It may need some tweaking anyway.

[ChristofferB]

Hey, and sorry for kicking a very buried topic, but since it's the exact same topic I didn't think there was reason to have two of them swimming about.

I have a last question to ask on this matter, somewhat dissimilar, somewhat not:

I've decided for re-implementing all the transistor amp stuff with op-amps, two stages to be exact, schematic below; I have no idea what sort of gain I'm going to need!

I was thinking x100 in the first stage (disregard gain set resistors and op amp type on schematic for now) and then a variable in the second with a trimmer, but I

honestly don't know. What would be a good first order decision to work from?



thanks!