I shoot for 1 to 3 mV (when warm) across each emitter resistor in a pro PA amp where there are 3 or more outputs in parallel. It is NOT the optimum bias for a class AB amp, but it will get distortion down to levels where you can't fault the power amps.
This is a BIG pro amp. 5kw per channel at 1 ohm, 14 pairs of bipolar output devices with 8 pairs of mosfet step switch devices. rails are +/-80 and +/-160V and it is used for running many sets of subwoofers so it gets the crap beat out of it.
I don't have a way to run it much past 400-500 watts. My load bank will go to 800 watts but I don't have AC power service in my shop that can handle it. My shop's AC is shared with the office next door unfortunately. (the down side of cheap rent.) So I have to error on the side of caution. I need it fixed and fixed right one time.
I trust DJK's information. I will start there and see how it goes.
Zc
I don't have a way to run it much past 400-500 watts. My load bank will go to 800 watts but I don't have AC power service in my shop that can handle it. My shop's AC is shared with the office next door unfortunately. (the down side of cheap rent.) So I have to error on the side of caution. I need it fixed and fixed right one time.
I trust DJK's information. I will start there and see how it goes.
Zc
Pink-noise test it (instead of sine wave) at high power. Use an 8 ohm dummy load one channel at a time. That will keep the AC demand under control and still let you test it to clipping. With most of these amps, the bias tracking is intentionally overcompensated (when it runs hot the idle bias drops a bit). "Light " load at high output will test the bias stability, clipping behavior, and proper operation of the rail switches.
With the exception of one bad LM339 comparater on one pre-amp board. I have both output stages up and running again. I set Bias to 30mv for starters. and WOW this baby is rock solid on the bias. if you set it to .30mv it doesn't even flicker a 10th! But i haven't got to do any testing at that bias yet as tracking down the issue with the pre-amp board has taken a lot of time.
One curious thing about this amp that i have never seen another amp do. with bias set, the AC Line idle current is very low. I have not measured it with a digital meter but the analog current meter on my variac is less then 1 amp which is the first mark on the meter and im sure the meters resolution down that low isn't very good...anyway..
With no speaker load connected to the amp but a DVM connected to the output. a sine wave signal feeding the input of the amp. if you vary the input gain. or the gain of the signal generator. As the output of the amp approaches .8vrms output, the amp starts to draw current from the ac line. it seems to have peak current draw around 1-1.5vrms. at this point the amp is drawing 4-5 amps of current from the AC line. scope shows a clean clean sine wave. no oscillations or visible distortion. once you go past this point, the current drops back down to idle levels and you can run the amp up to the point of clipping without any current drawn.
Both channels exhibit this behavior.
I need to repeat this test with a load connected and see if it does the same thing? but i thought it odd. without a load it shouldn't be drawing any current at any output level.
Zc
One curious thing about this amp that i have never seen another amp do. with bias set, the AC Line idle current is very low. I have not measured it with a digital meter but the analog current meter on my variac is less then 1 amp which is the first mark on the meter and im sure the meters resolution down that low isn't very good...anyway..
With no speaker load connected to the amp but a DVM connected to the output. a sine wave signal feeding the input of the amp. if you vary the input gain. or the gain of the signal generator. As the output of the amp approaches .8vrms output, the amp starts to draw current from the ac line. it seems to have peak current draw around 1-1.5vrms. at this point the amp is drawing 4-5 amps of current from the AC line. scope shows a clean clean sine wave. no oscillations or visible distortion. once you go past this point, the current drops back down to idle levels and you can run the amp up to the point of clipping without any current drawn.
Both channels exhibit this behavior.
I need to repeat this test with a load connected and see if it does the same thing? but i thought it odd. without a load it shouldn't be drawing any current at any output level.
Zc
Last edited:
I found a spec sheet for the amp that states a 2amp AC line current draw per channel at idle. wow that has to be a lot of bias. at 30mv across the emitter resistors I am not even drawing 1 amp from the ac line. could too low of bias cause the weird behavior i am seeing?
I have many phone calls into Crest/Peavey about the Bias spec for this amp but as of yet they have not responded.
Zc
I have many phone calls into Crest/Peavey about the Bias spec for this amp but as of yet they have not responded.
Zc
Oh yeah i hadn't thought about that. I spoke with the Crest service manager and he stated he wasn't sure why the 10001 exhibits the odd line current behavior but stated that it is normal for that model....so i tuned it up and out the door it went. just knocked out a 8002 power supply rebuild and i am on to a 9001 with a charred PC board...yea fun...
I am working on a 9001. The clip and temp lights stay on The The both pre amps and power amps work fine. All low and high voltages are good. I tried the both chs in another amp they worked great. The problem is on the Back Octal input board and or the front Display board. I am not getting the return signal on pin 3 of IC1. But I am getting it on the next side of C10. I replaced the cap, Ic and checked all the surrounding resistors. Still not getting the signal on pin 3. There is a .2dcv on this pin. If I install a .01 cap instead of the original 470pf, I would get the signal on pin 3.
Does anyone have any thoughts on this.
Regards,
Saj.
Does anyone have any thoughts on this.
Regards,
Saj.
The crest amplifiers have a High impedance protection circuit. I have found that dust, dirt and especially fog juice that collects on the power supply board around the protection circuits can cause them to stay in protect mode even though the amplifier is working fine. One amp I worked on was so full of fog juice that it had soaked into the circuit board edges and I never could get it to work right.
I got everything functioning except its staying in the temp/ dc protect mode. the ouput relays would obviously not come on. Whenever I remove pin6 on J3 and J4 The relays would come on and everything function normally except there is no signal on the signal leds.
Does anyone have any thoughts on this.
Does anyone have any thoughts on this.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.