sam's 300W or eti-466 :setting quiescent current
I've put together the amp. at :
I follow the setup procedure at cambie's site( but I reference all component id as in Sam's site) . The first part (shorting the e,c of Q8 and output) works out fine but I had some problem in setting the quiescent current.
Things I had in my setup differ from the original :
1. I use a 3K pot instead of 2.2k (I start at 2.2k and adjusting downward)
2. I use BD139 instead of BC549 (I found this easier to mount to the heatsink)
3. I had skipped the Zobel network.
4. I use 7.5 ohm instead of 10 ohm across the fuse holder (I should measure a voltage of .75V if everything is OK, am I right?)
As soon as I throw the switch, the voltmeter reads about 2.5V, ( Q. current around 330mA) and keep rising. I turn down the pot and the voltage rise higher. If I turn up the pot, the voltage swings down but as I reached 2.8k (max of my pot) the voltage rises beyond 3V. (Q > 400mA). Seems I should use pot + R22 > 7.5k. Is it save to do so? The problem is not only unable to adjust the Q current but seems the setup is unstable. (Or will it be stable if I can achieve the norminal Q current). The DC voltage between output and 0V is about 5.0mV
I had test all components before soldering and check my pcb and soldering points.
As suggestions ?
Tks a lot.
Last night, I tried to increase the pot value (or more precisely the resistor between b of Q8 and pot). I raise it to about 75k (by steps of about 2-3k a time). I finally manage to keep the voltage across my test fuse (7.5 ohm in my setup) to about .85V thus giving a Q. current of 120mA. Is that good enough? Would this be OK cause the original design had a close symmerty between b-e, b-c of Q8.
I also found the 1k 5W resistor to be uncomfortably hot. Would this be normal. I did not tried with any sign injected or speaker connected. I'm worried that they might blow up when work with a load
Anybody kind enough to give me some advice?
The BD139 is much lower gain than the original BC549, and will require higher base current to turn it on. Therefore REDUCE R22 (on original schematic) 6k8 to maybe 3k3 instead of increasing the pot value.
You said you skipped Zobel network - you mean C15/R47? - BIG NO-NO!! This provides load at high frequencies, and can easily lead to instability. It doesn't take much to upset this amp.
The 1k 5W resistors R20 and R25 will run reasonably warm, and are not effected by signal levels.
It seems that using 3.3K does not solve the problem. I just try using the bc549 and fall back to the original values (expect the pot) Now I can adjust to a Q. current of about 110mA /w pot at 2.8k. Hope that is close enough (or should I raise R23 higher to get closer). Now I have to find out if the BC549 is thermally coupled to the heatsink. I think I'll try to get the thermal epoxy tomorrow. But now I found that R44,46 (5k6 at B of Q16,17) burning hot. I don't know what might happen when I actually load the amp. Any comment on that?
Have you reconnected the zoebel network?? If not, I would guess that the amp is oscillating supersonically (which this amp is prone to do if shortcuts are taken) - this would explain both the hot resistors and high quiescent current (the trimpot at maximum should be capable of reducing current to zero). I would also recommend shorting the input whilst setting up.
I had the resistor and cap reconnected (does Zobel include the inductor?) but I don't have the inductor. Its is difficult to get enamelled wire here but I might be able to find some ready made.
I had now R22+pot ~ 7k5.
Could R19-32 be the trouble-maker? Can I test if they are carbon or wire wound with a DVM? When I blew one of the test resistor (fuse replacement), I see some wires-like stuff inside.
Finish it at last. I had it driving my friends 18" speaker. They said it's ok but still needs tuning. It's running a bit warm (very comfortably warm, I guess just a few degrees about room temperature). My friends told me I have to check for distortion and stuff like that.
After it makes a sound, what do we have to do on it? According to my ears (which is not too smart I must say) they sound OK. Any suggestion on procedure to tune it?
|All times are GMT. The time now is 01:30 PM.|
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio