A few questions about re creating vintage amplifier using ac187/188 K germanium pairs

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi. I want to build a low power germanium stereo amplifier and i got a little stuck...
So the story goes like this: A few years back was repairing a vintage radio and had it hooked up to one of my crappy speakers which had a decent bass for what it was and i was really surprised by the "warmth" and power of that tiny amp. I don't know if it was because all i used at the time were cheap class d board or because of the tapped volume control but I liked the sound and I didn't even know germanium transistor were a thing at the time... Then I started to look for a circuit using these transistors (ac187/188k) and I found an interesting amplifier which is riz transiwatt 8 and it uses these transistors. It doesn't have a tapped volume control but it has a tone control.
Here is a circuit:

https://www.google.com/url?sa=t&sou...Vaw1bdam7IFN2Tf_iu8zoPBSu&cshid=1607085821221

Now here are my questions:

1 what do r13 and r14 do. They seem to be trim pots and i don't know how to adjust them.

2 where am I supposed to find 13M ohms volume pot?

3 is there a substitute for ba103 diode because NOS part costs 20 dollars!!!

4 I see that under each transistor it says another transistor name. I think that is other compatible transistor. Now what do I do. I can get NOS parts or use bc108 ? (I already have NOS output transistors)

And finally is there some suggestions about the pcb layout?

Thanks in advance and sorry for the wall of text. Also if anyone has a better circuit for these transistors feel free to tell me.
 
Last edited:
1. These set the bias for the output stage. Ideally the positive leg of c14 should be 1/2 the supply voltage.
2. No idea but it likely has something to do with the filter on the input.
3. The datasheet says its a silicon diode so anything with a similar rating should work. The 6v rating leaves me scratching my head though.
4. Its really up to you if you want a bc108 or b115.

I'm just a novice here :)
 
AC187/AC188 are well known to me in 60,s plastic small portable AM radios as output devices .


Old school biasing for Germanium BJT,s used both an upper and lower resistor attached to the base to bias them unlike the modern silicon versions .


The "B" in BA denotes its silicon as opposed to "A" which is ( early ) point contact ) later silicon junction.


As its a standard diode of its time why don't you check out old 60,s transistor radios usually in a lot of them.


If you want to keep it Germanium then don't install the silicon BC range apart from the fact it might not function correctly even though the spec ,is near it ,the circuit compensated (usually ) for the leakage in them.


I cant access Google I have it blocked on my network port(s).
 
Administrator
Joined 2007
Paid Member
I suspect the volume control would be 1Meg rather than 13Meg and even 1Meg is high and yet consistent with using a crystal cartridge whcih would be common at the time.

For modern sources you can use a 10 or 22k and reduce the balance pot and the 51k by the same sort of ratio.

Why not initially build it with all silicon which apart from the output stage it is anyway.

The BA103 comes up as silicon... could it be? The package suggests it could be mounted for thermal compensation. If it is silicon then a 1N4148 should work. It should be in thermal contact with the output stage though.

I suppose it could be Si given the 0.47 ohm emitter resistors.
 

Attachments

  • Screenshot 2020-12-04 143735.jpg
    Screenshot 2020-12-04 143735.jpg
    127.5 KB · Views: 353
  • Screenshot 2020-12-04 143831.jpg
    Screenshot 2020-12-04 143831.jpg
    109.3 KB · Views: 331
Thanks. Here is only thing i was able to find about this amp pretty much. It is a post on Croatian DIY audio
where guy restored one of these amps (there are pictures of the inside here)

Forum audio samograditelja • Pogledaj temu - RIZ TRANSIWATT 8

This amp is for some reason extremely rare and pretty much impossible to find. Also I don't think that diode is used for thermal compensation because on the pictures of the inside i only see output transistors on the "heatsink" and all the other semiconductors are on the pcb.
 
Many circuits on silicon. You can use any medium power complementary pairs in your circuit. TYPE 31/32С, TYPE 41/42С etc. are available.
€ 0.23 SD 882 Complement to 2SB772
Transistor 2SD 882 NPN, 30 V, 3 A, TO-126
High input impedance to work with vinyl crystal head. Use the usual 10-100К.
In series D1 (any silicon), turn on a 27 ohm resistor (selected).
 
Last edited:
Many circuits on silicon. You can use any medium power complementary pairs in your circuit. TYPE 31/32С, TYPE 41/42С etc. are available.
€ 0.23 SD 882 Complement to 2SB772
Transistor 2SD 882 NPN, 30 V, 3 A, TO-126
High input impedance to work with vinyl crystal head. Use the usual 10-100К.
In series D1 (any silicon), turn on a 27 ohm resistor (selected).

So you are saying i can use silicone transistor for testing. That would be great because I am kind of scared of destroying old germaniums.
Also as far as i know bias is different for silicone transistors ?
 
Administrator
Joined 2007
Paid Member
Yes you can use silicon. The big differences are that germanium transistor begin to turn on with as little as 150mv B-E voltage vs around 600mv for silicon. The transfer characteristic will be worse with silicon I expect (more distortion) although you can get around that by increasing the bias current.

Certainly for basic testing you can just substitute silicon here and although the bias current will be non existent the amp will still work OK like that.
 
Administrator
Joined 2007
Paid Member
I've just done a quick simulation but am a bit unsure of the results. The gain is extremely high, the response poor and drive ability is best not mentioned.

I suspect R10 is incorrect and should perhaps be 2k2.

This is as drawn. Nearly 50db gain and response rolling off from a couple of kHz.

Now with 2k2.

.asc file attached for anyone wanting a play.
 

Attachments

  • Untitled1.jpg
    Untitled1.jpg
    476.5 KB · Views: 266
  • Untitled2.jpg
    Untitled2.jpg
    458.9 KB · Views: 253
  • Transiwatt 8.asc
    3.9 KB · Views: 42
  • Untitled3.jpg
    Untitled3.jpg
    442.4 KB · Views: 266
Administrator
Joined 2007
Paid Member
Perhaps :)

With 22 ohm in place full output is reached for just 14mv rms input. I never dabbled much with ceramic cartridges but seem to recall they put out a few hundred millivolts or so. It just seemed a bit to high gain. No getting away from that lack of HF though.
 
Philips manufactured tubes since early radio days ( approx 1927), I didn't like their radios internally as the internal wiring was - black or black or black .
Likewise the capacitors were coated in a --yes a black tar like substance giving no indication as to the type or value.


They used rubber insulation that that ---yes disintegrated after many years leaving bare wires , on the other hand they were quite innovative in circuit design .


First place Hitlers troops visited was Hornemann in Eindhoven.
 
If anyone know a better circuit using these pairs feel free to tell me. I don't know anything about this amp. I was just looking for a circuit that is using these transistors and preferably but not necessarily has a tone controls.
I don't know how it sound or anything. Only place where I heard these transistors was in a old radio and I liked the sound. Probably not real hi-fi but warm and enjoyable.
I have two vintage speakers that I really like and would like little amp for them using these transistors.
 
There are some mistakes. The BC214 is a PNP and should probably have been BC184- that was the NPN complement, but BC108 is fine.
The AC187/188 pair are good ones to use for a Ge output stage. I would, however, use a germanium transistor for the bias stabiliser in the same way that silicon transistors are done. It would mean getting hold of a second AC187 (NPN) which would (should) give better thermal stability than a silicon diode, because you need two Vbe compensation while a Si diode would only give one.

A rather better arrangement might be to use a PNP input transistor (BC558) and a BC547 driver in the classic 4-transistor arrangement. Make it 5 with a bias stabiliser. I guess transistors weren't used back then as they were pretty expensive.

I would not recommend a silicon output stage like TIP31/32 for lowish power output. The AC187/188 had a minimum gain of 50 (80 for the PNP) at 1A. If you want to use silicon, and I don't believe you do, I would recommend at Zetex (now Diodes) type which also have high gain at 1A or so (like ZTX650- ZTX750) These can do 2W with a heatsink, about the same as the AC187/188 with a heatsink.

You can use a JFET or a buffer circuit to give a high impedance input which will be able to drive a lower resistance to your circuit as Mooly suggested.
 
Ok, supposing you *actually* want to build it after all, I´m digging in "as if" you will.


Not a bad amp for the era, when I started (1969) amps like these were in millions of homes.

This was the "premium" housing: separate amp on its own and 2 detachable speakers, most common would have been exact same guts mounted "all togeter" inside a "combination grammophone": amp, speakers, turntable, most often an AM radio tuner and sometimes an FM one.

Same era tube equipment (which would have been all over the place) would have used (in Europe) a stereo 6BQ5 based amp, so 2 x 5-7W for the "good" ones and stereo or mono ECL82 based ones (2 x 2W RMS) for the "economy" class, meaning that a Solid State 2 x 4W RMS amp would have been a perfect competitor.

Germanium was still popular for power devices, I built many Hi Fi 12-15W amps using AD161/162 and hundreds of battery powered Guitar amps using AC187/188 well into the 70´s, because they provided a significant power advantage when fed from lowish 9V DC, from 6 x carbon zinc "C" batteries, go figure.

Hi. I want to build a low power germanium stereo amplifier and i got a little stuck...
.........

Sorry, coffee break is over, will continue answering your questions later, and suggesting a couple circuit corrections, including what I consider printing errors.

Typically missing decimal diots/separators so 8 2k actually means 8.2k (notice space between schematic printed 8 and 2 digits) and so on.
Same with "1 3M" volume pot, not the same as "13M", where I guess it "must" have said "1.3M" but dot was missed.

In all, a decently made amp for the era, but for some reason germanium was being phased out and that´s lack of performance compared to silicon :rolleyes:

EDIT: a not so small point: you will need to design and make a PCB for it ... will you?

Even if you make it on perfboard or veroboard, you will still need to make a layout.

Not insurmountable, but like anything else it implies some "learning curve" .. typically a couple Months and 4 or 5 "practice" designs, ... or this same one but be ready for 4 or 5 iterations (restarts - corrections) until you are more or less happy with it.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.