| Electronics > Projects, Designs, and Technical Stuff |
| Help me understand and troubleshoot this early 70's circuit |
| << < (2/2) |
| Maxxarcade:
I ordered some BC546B and BC546C to test out, which should be here in a couple days. I'll report back with the results. Worst case scenario, I'll just add a series resistor to the B+ rail. But it's nice to be able to understand the problem either way. |
| bson:
Saturation is the same for modern transistors, at the same range of Vbc at 0V (Ic/Ib starts falling off) to 0.4V. Try to find specs for the original transistors, it may be that something like an MPS42 or 2N2222 is a better substitution due to beta, SOA, max Ic. It's worth noting that a 2N3904 is very easily destroyed if overloaded, it won't smoke or get hot - it merely warms up slightly, then behaves erratically (i.e. not like an NPN BJT). |
| Maxxarcade:
I tested with BC546C, and there was no change in the behavior of the whistle. In fact it seemed more muffled. It was definitely Q6 and Q8 that were generating the whistle sound, so I think I need to focus on those. I also noticed that if I touch the back of the board, and my finger bridges the B+ trace to the base of Q8, it will start to whistle until I let go, and varies with the amount of pressure. But I'm not sure if it's the correct sound, or another part of the circuit making the noise. The BC546C did improve the overall gain when used in Q1, Q2, Q4 and Q5, but a 2N3904 at Q7 was much better at generating good white noise. Lowering the voltage to the trigger input (connector pin 4/input of R25) prevents the whistle from starting at all. It basically behaves the same as using pin 3. |
| Maxxarcade:
I'm getting warmer now. I found that the circuit works fine if I put a high value resistor (1Meg to 3.9Meg seems to work best) between the base and collector of Q8. Without the added resistor, there is only about 0.14 volts on the base of Q8. When I add the resistor and the base gets around 0.6v DC on it, it starts working. But go even slightly too high (0.65v or more) and the frequency goes too high. So to me it looks like there might just be a bias issue with Q8? I tried different values for R21 a while back and didn't have any luck, but manually adding the perfect amount of DC to the base seems to do the trick. It's probably not the ideal way to fix it, but it's another option besides adding series resistance to the B+ rail. I suppose I should also try the board in the machine so that it's running on it's original power supply. For all I know it might work fine in there, and I'm going through all this for nothing. But right now I'm working from home, and I only have the board here with me. |
| floobydust:
Q8 base 0.14V seems wrong, something is leaking, like Q6 staying on a bit, or C9, C8 leaky or R22/R9 bad. If Q6 is off, Q8 should be on with 0.6V on the base, and the opposite with Q6 on, Q8 should be off (when over 7V input at C15). The zener D5 18V to regulate VCC (and make noise lol) so it should not change a lot under different power. Unless pin 6 B+ runs off to other things in the game. I thought it would only go to the sound switches. P.S. What is the audio amp IC used? |
| Navigation |
| Message Index |
| Previous page |