Any leads would be appreciated.
Someone said BC549 or 2n3904 for the BC169B.
I can't find any equivalents for the 2n4062.
Schematic of the amplifier they're to go in attached:
Q1,2,3,5,6 and 7,8 respectively.
Someone said BC549 or 2n3904 for the BC169B.
I can't find any equivalents for the 2n4062.
Schematic of the amplifier they're to go in attached:
Q1,2,3,5,6 and 7,8 respectively.
That is a very generic type circuit and tbh pretty much 'anything' is going to work in there. The BC169B is only a 20 volt device but they are high gain. My old trusty Towers book suggests BC184L as a replacement along with the BC214L for the 2N4062. In practice though and most small signal types will work just fine.
@Mooly Great - thanks so much, that's good news. Probably since it's a guitar amp it was meant to have higher gain, but it's quite a noisy amp, especially around the 2n4062s, so I thought it would be worth a try to swap those out.
A bit dark sounding unless the treble is maxxed out, so maybe I need to change the electrolytic caps too.
A bit dark sounding unless the treble is maxxed out, so maybe I need to change the electrolytic caps too.
All you can do is try but I would be surprised if different transistors altered the noise floor by much. I think you'll find the reverb stage runs at very high gain (because the reverb unit itself is very lossy) and that is where a lot of noise will come from. Q5 is configured for a very high gain by the looks of it with a 10k collector load resistor and 75 ohm (R26) in the emitter.
Looking closer and the gain of the stage around Q5 is over 40db meaning that for just 10mv of signal applied to the base you get over 1 volt at the collector. So a massive voltage gain and all from one transistor.
Looking closer and the gain of the stage around Q5 is over 40db meaning that for just 10mv of signal applied to the base you get over 1 volt at the collector. So a massive voltage gain and all from one transistor.
I agree with Mooly 100%
To be honest, your supply is 43 volt, so going with a 50 V transistor (common) with high-ish beta will work fine. Your noise is determined by other parts, so transistor type will not affect that much at all.
Look at 2N5087 as a possible, beta ranges from 250 to 900 - so good enough easily.
-Chris
To be honest, your supply is 43 volt, so going with a 50 V transistor (common) with high-ish beta will work fine. Your noise is determined by other parts, so transistor type will not affect that much at all.
Look at 2N5087 as a possible, beta ranges from 250 to 900 - so good enough easily.
-Chris
At least I would replace the input emitter follower by some JFET source follower. Generally in the old times many designs emanated with quite poor noise figures - nobody cared about. I just "repaired" a Marshall Combo from 1988 with "unbearable hum" - which seemed to be a slight exaggeration to me at the first glance. These old babies often hummed, even with master volume muted. But ok, I accepted the challenge and found out a humming loop caused by improper pcb design. Relocating the gnd return pin of a supply bulk cap solved the problem completely - those were the days of the "geniusses"...
Anyway, I share moolys opinion, any standard BJT will do. Probably otherwise there is nothing to "repair" with that mean circuitry.
😉
Anyway, I share moolys opinion, any standard BJT will do. Probably otherwise there is nothing to "repair" with that mean circuitry.
😉
When you know the dependence on the volume and reverb settings, you know what contributes most: the high-gain stage after the reverb unit or the input stage.
For the input stage, transistor current noise will be dominant, so a transistor with high hFE and not too much 1/f noise could help somewhat, BC550C or so. Preferably with a 1N4148 in antiparallel with its base-emitter junction to protect it against emitter-base avalanching. As suggested by bucks bunny, an N-channel JFET is better.
For the input stage, transistor current noise will be dominant, so a transistor with high hFE and not too much 1/f noise could help somewhat, BC550C or so. Preferably with a 1N4148 in antiparallel with its base-emitter junction to protect it against emitter-base avalanching. As suggested by bucks bunny, an N-channel JFET is better.
Last edited:
That's in the middle of a low-gain power amp. That's not your hiss.
Replace Q1 with any fresher NPN. Something we don't see in home-amps, but do see in stage-amps, is stray or strange inputs at the B-E junction, makes damage, "emitter-base avalanching", which can lead to a 'functional' part with very raised hiss level. One early Heathkit stage-amp is notorious. This Ampeg looks better protected but maybe it has a bad history.
No, we don't "improve" well-loved classic guitar amps without a lot more thought than we have here.
Ah so, that much huh? That's interesting. The reverb has a footswitch to take it out of circuit, but that doesn't affect noise at all. The reverb increases hum when you turn it up but not hiss. Is the reverb return transistor in circuit all the time, regardless of whether it's turned up/on? Bummer!
Thanks for all the comments, @anatech @bucks bunny @MarcelvdG @PRR I'm learning a lot. As I said elsewhere, someone recently told me my Akai tape deck from the same era would benefit from newer transistors, so that's why I asked.
I would love to know how to do this if it's feasible.
I dropped these to 47k and it helps some. But now I get a little AM radio, as expected, lol.
On my "scope" the noise in the beginning seems fair - 60db s/n ballpark(?), which is fine - but it gets noisier by the time it's at the power stage. Is there a better way to bypass the reverb return part of the circuit?
This particular Ampeg was a low-end model and was never especially well-loved -- it's far from a Fender ab763, so I'm sure it could have been a better design. It makes a fine bass amp because the treble doesn't have to be above 7.
I still have to track down the sources of hum. It won't be a high-end amp, but it's dripping with 70s mojo so I like it and it's worth at least getting in original shape.
Turning the control marked Lin Reverb (VR4) to minimum should remove that section from the circuit. If you remove C16 as a test then everything reverb related is removed.
There are a few decoupling caps that are worth looking at such as C22, C24 and others. There are quite a few decoupling bias networks and supplies. A scope should help show where hum is getting in. 100/120Hz 'buzz' suggests injection from the rails, 50/60 Hz would be more like induced hum.
If I understand you correctly, the VR4 setting has no impact on the noise. Is that correct?
Could you also check the effect of the VR1 setting? Preferably with a guitar connected straight to the amplifier input.
Replacing an NPN with an N-channel JFET usually requires a bias circuit redesign, but in the case of Q1, it doesn't. Just check what N-channel JFETs in a suitable package with an IDSS of at least 2 mA and breakdown voltages over 30 V are still available, buy one and connect it there, drain to where the collector used to be, gate to base, source to emitter. Whether it is useful to try this depends on the outcome of the VR1 experiment.
Could you also check the effect of the VR1 setting? Preferably with a guitar connected straight to the amplifier input.
Replacing an NPN with an N-channel JFET usually requires a bias circuit redesign, but in the case of Q1, it doesn't. Just check what N-channel JFETs in a suitable package with an IDSS of at least 2 mA and breakdown voltages over 30 V are still available, buy one and connect it there, drain to where the collector used to be, gate to base, source to emitter. Whether it is useful to try this depends on the outcome of the VR1 experiment.
Last edited:
Okay, maybe I misunderstood you in #4. The reverb control VR4, oddly enough, does not add hiss. So if turning it down or up does not change the hiss, I guess Q5 is not really a key source of the hiss. I will try lifting that cap. VR4 dials in quite a bit of 60hz hum.
I have a DIY listening probe (capacitor & resistors) and I am looking at the signal on an oscilloscope VST in my DAW. I hear hum on some of these caps, but I am not clear on where it's just voltage without the audio signal. I will try to find the first spot with audio and hum.
There is certainly hum on C22/C24 - but I tried paralleling in another cap with no effect. Will try that again now.
That's correct.
With the guitar (or test tone) connected, dialing up V1 increases the hiss; even more so, dialing up V2 (treble) increases the hiss (as would be expected).
Then Q1 is the main suspect, assuming the resistors around it are of a reasonable quality (carbon film or better, not antique carbon composition resistors).
They certainly are indeed antique carbon comp resistors. Are there certain key resistors I should try swapping out with modern ones? I want to minimize chances of lifting a trace on the PCB here so only change out what is necessary and/or j-hook new components in.
I don't think so 🙂 Q5 runs at high gain and so would have been an obvious candidate... but if turning the reverb down does not change the hiss then its not that. The hum (60Hz) does make sense because that super high gain makes it easy to pick up stray hum. If you remove C13 you isolate that stage from the delay line. Does it hum now?
The carbon comp resistors could be an issue. Obvious ones to swap out would be those in the bias networks to all the transistors.
You might in the end be chasing things that aren't actual faults, rather it is just the way it is.
I think you're right - it's not a matter of drifted or outdated components, as I had hoped. I'm probably going to embrace the lo-fi, and maybe mod Q1 to a JFET just for fun to experiment.
With the reverb, that white wire that flies across the board to the Reverb on/off switch probably doesn't help. Maybe I will shield it.
Also I tried an experiment, using another preamp going into this amp, boosting highs, and I think I can get better high frequencies and a bit less noise at high volume by driving the input with a boost pedal or something, juicing the gain a little. If anything can recover some highs, it would help.
🤷♂️ It does sound good as a bass amp. 😎