Help me bias my amp...Please

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

scott2000

Well-known member
Joined
Feb 21, 2015
Messages
3,896
Location
Sunny...Sometimes Florida- USA
I was under the impression after some reading online that  around 500mvdc at the junction of the bases of Q8 and Q9 would get me in the ballpark but I'm not sure if I was probing correctly.. Is this a good place to check? Would 500mv be correct???

Also,

Are the 36 ohm emitter resistors for q10 where I should be measuring across too??? Or the .43 ohm emitter resistors?
What voltages should I look for?? Around 20mv???

All this with no input signal?????


Once again, I appreciate ANY direction...

I've touched Q10 from probing and smoke came from somewhere on board and nasty hum almost blew my speaker....... Everything on the board is checking out fine now...???..... Need to find out what blew????Didn't see exactly where smoke came from....... :-\

There's some kind of thermal sensor on the heat sink..... it's open so I am guessing that's not good .....


 

Attachments

  • Amp bias !.jpg
    Amp bias !.jpg
    193.6 KB · Views: 53
Thanks!

Looks like Just the thermal switch took the hit....... I had an old QSC amp that had one I scavenged..good find....it's ..$16 at Mouser!

I'm going to wait for some direction before I put the boards back on. The test points/areas/spots I think I need to probe are really hard to get to and I think I may put some leads on before hand just to set everything.....

I do have a scope...... Would be nice to use it. /// I haven't been able to figure out where to use that for biasing either.... :-[

 
scott2000 said:
I was under the impression after some reading online that  around 500mvdc at the junction of the bases of Q8 and Q9 would get me in the ballpark but I'm not sure if I was probing correctly.. Is this a good place to check? Would 500mv be correct???
Q8 and Q9 are part of the current limiting output stage protection, so will have 0V at idle.
Also,

Are the 36 ohm emitter resistors for q10 where I should be measuring across too??? Or the .43 ohm emitter resistors?
What voltages should I look for?? Around 20mv???
when operating with modest class A bias current you should see diode drops (0.5-0.6V) across base-emitter junctions.

The 36 ohm will have the base-emitter drop of Q11 plus the class A bias current times the 0.43 ohm  emitter degeneration resistor.

20 mV across the 0.43 emitter resistor makes 46 mA of class A bias, not low but should not release smoke.  So total across 36 ohm will be one diode drop + several mV (hard to predict exact voltage).  When setting class A current (with Q Bias trim) measure voltage directly across the 0.43 emitter resistor.  I would expect something between 10 and 20 mV across 0.43 ohm resistors.
All this with no input signal?????


Once again, I appreciate ANY direction...

I've touched Q10 from probing and smoke came from somewhere on board and nasty hum almost blew my speaker....... Everything on the board is checking out fine now...???..... Need to find out what blew????Didn't see exactly where smoke came from....... :-\

There's some kind of thermal sensor on the heat sink..... it's open so I am guessing that's not good .....
Depends on how thermal sensor is connected... thermal cut outs often disconnect PS rail... That amp also had DC protection  Q17, but unclear what it connects to perhaps an output speaker relay?

Try not to blow stuff up... that makes repairs harder and more expensive.  8)

JR
 
Yes.... I'm paying for my tuition into how things break for sure...... I wish I would've left well enough alone but, I wanted to make sure I didn't have any issues down the road from bias not being right....

This all started from me getting random crackles....

Thought it was a transistor....nope..
Leaky caps maybe???? Went ahead and replaced them....nope


Thought it was the .43 wire wound resistors because they were 40% out of tolerance........nope....learned that my fluke meter isn't so accurate down that low but a cheapie LCR meter is????? At least I have some extra resistors......

Turns out I had a bad input pot......

Then the slip of the probe......

Blew the thermal switch open..... fine, I had one.......

Now the amp won't even turn on...... This big ole transformer couldn't have broke from my mistake????? I did get the worst sounding hum coming through that channel that just about threw my speaker driver across the room.......

fuse is fine....

I need to figure out how the ac relay board works....Doesn't help that the manual and schematics are not the same amp....... Similar but definitely not the same.... a few Different values and placements  of components.....didn't see that hidden triac under the board....Maybe I'll get some luck with it....??

Thanks so much for the advice.....!

Hopefully I'll get another shot at biasing this..... :-\

 

Attachments

  • ac board1.jpg
    ac board1.jpg
    214 KB · Views: 6
an old trick when troubleshooting amps is to wire a 100W light bulb in series with the line cord (old school incandescent please). The lamp will current limit at about 1A if amp is trying to draw excessive current.

A variac also helps while troubleshooting.

JR
 
JohnRoberts said:
Q8 and Q9 are part of the current limiting output stage protection, so will have 0V at idle.

If you are referring to the annotation that says "500mv?", I would expect the voltage drop from the two 0.43 ohm resistors to show up there. 1/2 volt implies over 1/2 amp through the sampled transistors.

Or am I missing something?

Gene
 
I think I had it backwards.......

I revised this drawing....I was just thinking ballpark but I'm still on the drive there it seems.....

....I'll get my head around it soon enough

Been working on my truck wiring most of the day....

Have a coolant sensor reading -40* but the computer is correct in voltage.... TBIs are great if everything jives....Mine's not...

Thanks!!
 

Attachments

  • Amp bias !.jpg
    Amp bias !.jpg
    193.1 KB · Views: 7
> Have a coolant sensor reading -40*

Check wires, then replace the thing that screws into the block. It has gone open-circuit, and most thermistor readers round "infinitely cold" to -40F.

> This big ole transformer couldn't have broke from my mistake?????

Little mistakes can break big things. Yesterday I pulled two people from a totaled car. Some minor diving mistake led to upside down in the woods, and not obvious from the road. I'm thrilled that both persons were OK. The car will never be the same again, and was worth some more than your puny punt-anchor.
 
Holy smokes! That's crazy!

Good thing you were there and everyone was ok..... Scary stuff.....

Yeah the sensor is a new delco but, who knows. My buddy did run a scan a year ago and showed me the same thing. I've since replaced it and just put a new one in. Finally got a bluetooth obd1 scanner (highly recommend) and saw the reading...... The wires look fine from sensor to ecm and the voltage was spot on at the ecm in closed loop so I'm guessing somewhere between there and the aldl port.......
Haven't gotten that far to figure out where the ecm runs out to......

Thanks again!!!
 
scott2000 said:
Yeah the sensor is a new delco...

Delco implies probably GM, OBD-I means '95 and back, throttle body means not so far back to the dark 4-BBL carburetor days, likely similar to my '91 350 small block.

In mine there are two temp sensors, one for the ecm, and one for the dash gauge. And I believe they are one-wire sensors, meaning that they make ground through the block. You didn't do a nice, proper plumbing job on the threads, with plenty of insulating teflon tape, didja?  :D

If the ecm thinks it is -40F out, it will idle eye-burning rich, and get MPG on the order of a sherman tank.

I've had mine since '98 and am very familiar with it, and since this is a severe topic drift, feel free to PM me if you think I can help.

Gene
 
> my '91 350 small block.
> In mine there are two temp sensors, one for the ecm, and one for the dash gauge.


Where is the one for the ECM? (A 93-95 mongrel, if it matters.)
 
PRR said:
Where is the one for the ECM? (A 93-95 mongrel, if it matters.)
I don't remember offhand, it was at least a decade ago when I messed with those sensors to fix a low-reading temp gauge..Changed the wrong one, and it turned out to be the gauge anyway. Still reads low, but in my mind, that is the new normal.

Supposed to be tolerable weather here tomorrow, I'll go check my Vehicle Maintenance notes in the garage

OK if I PM you, to reduce the topic shift here?

Gene
 
Gene Pink said:
I don't remember offhand, it was at least a decade ago when I messed with those sensors to fix a low-reading temp gauge..Changed the wrong one, and it turned out to be the gauge anyway. Still reads low, but in my mind, that is the new normal.

Supposed to be tolerable weather here tomorrow, I'll go check my Vehicle Maintenance notes in the garage

OK if I PM you, to reduce the topic shift here?

Gene
This is a huge veer but my (aging) car has two temp sensors and I think the one driving the the ECM is faulty. Car appears to run lean when cold.  While the dash temp gauge is fine... I guess I could swap them...  ;D  (Odd that detroit would use two of something when one would do it. They are some of the sharpest pencil designers.)

The check engine light has been on for a couple years, I checked and the engine was still there.  8)  (actually a neighbor plugged up his code reader and it says some emissions who-hah isn't hahing...).

JR
 
I got the amp working again.....

It turns out that it was just Q10 that was bad..... I messed up by thinking that the thermal switch was supposed to be closed but it was an open switch by default....Changing it with one from another amp had me chasing my tail because it was a closed switch...... :-[

Now I have the original issue of the random cracklies but on channel one now.....The fiasco that started all this repair was because of random cracklies on channel 2.....Sounds like lightning striking in the distance and how it can sometimes send little static through the power......

Replacing the bad input pot seemed to make the issue go away on channel two and I had replaced channel one at that time too.....Channel two is still quiet........... My slip of the probe may have cause some damage I haven't found....Or may the bias pots are going bad too???Same type of trimmers as the inputs I replaced....

I may just shotgun all the transistors.....It's a pretty easy thing if I can find the right parts...... It's pretty random but irritating.....

My questions are, when trying to get around 10-20ma at the .43ohm emitter resistors for biasing , am I doing this at idle with no signal????

Are all the .43 ohm resistors supposed to read similar????

And I have 43mvdc at channel one speaker terminals and 16mvdc at channel two's....Is this normal or acceptable???? Is there something I should be looking at???

Thanks for all the help!!!!!!!

PRR said:
> my '91 350 small block.
> In mine there are two temp sensors, one for the ecm, and one for the dash gauge.


Where is the one for the ECM? (A 93-95 mongrel, if it matters.)

Gene's should be similar but it's older than mine by a couple of years....


My sensor that goes to the ecm is at the intake manifold by the water neck for the upper radiator hose....

The gauge sensor is at the exhaust manifold......between 1&3 plugs.....

I was reading my manual and was told to remove the sensor plug and jump the two terminals to see if the reading would show normal.....It does.......

Getting close.......
My Mongrel.....lol......93 GMC Yukon.....
 

Attachments

  • ab amp channel.jpg
    ab amp channel.jpg
    178.2 KB · Views: 4
> Are all the .43 ohm resistors supposed to read similar????

In circuit, probably not. In fact flip your red/black probes and it probably changes: you are read the transistor or stray charge.

> 43mvdc at channel one speaker terminals and 16mvdc at channel two's....Is this normal or acceptable????

I'm too foggy to be sure. It's "acceptable" for an old low-price PA amp, yes. It seems to be about what you'd have for a poor-match input pair.

> ......between 1&3 plugs.....

Thanks. (FWIW: two sensors because you don't want the owner to know too much, but the ECU should follow even slight surges. It does seem like there could be a cheaper hack, but thermistors are really cheap by the million, and GM does not dread tapped holes.)
 
PRR said:
two sensors because you don't want the owner to know too much, but the ECU should follow even slight surges. It does seem like there could be a cheaper hack, but thermistors are really cheap by the million, and GM does not dread tapped holes.)

That explains why they look identical except for the pins in the connectors>>>>>

....It even lists as the same "coolant temperature sensor".... You have to read the fine print in the application details (does it even mention it?.......comments help I guess).......) 

https://www.summitracing.com/parts/ado-213-928

https://www.summitracing.com/parts/ado-213-77

Thanks for all the help everyone!!!! The amp cracklies are gone for now..getting 15mv at the .43 resistors.... seems good ..... don't want to risk breaking anything more if I don't need to.....

Sure would like to know how to scope it though to see the bias results..... be a good learning experience.......

Better not though/// :-\
 
scott2000 said:
That explains why they look identical except for the pins in the connectors>>>>>

I got out there today to my notes, and my sensor AC/Delco numbers are the same as yours. The single wire one is up by the thermostat, and my notes say that is for the gauge. Leaning over the radiator from the front, it is about 3 inches at about 8 o'clock from the center of the thermostat.

The other one can't be seen by me, as my vehicle is a Vandura 3500, a van conversion to a 14' boxtruck, I'll need to remove the doghouse in the cab to get to the motor, to see anything.

Meanwhile, the attached may be of help.

Gene

 

Attachments

  • GM thermistors.jpg
    GM thermistors.jpg
    112.8 KB · Views: 5
scott2000 said:
Now I have the original issue of the random cracklies but on channel one now.....The fiasco that started all this repair was because of random cracklies on channel 2.....Sounds like lightning striking in the distance and how it can sometimes send little static through the power......
Dirty jack? Cold solder connection? Does the sound change with percussive troubleshooting (tap components with a pencil eraser)?
Replacing the bad input pot seemed to make the issue go away on channel two and I had replaced channel one at that time too.....Channel two is still quiet........... My slip of the probe may have cause some damage I haven't found....Or may the bias pots are going bad too???Same type of trimmers as the inputs I replaced....

I may just shotgun all the transistors.....It's a pretty easy thing if I can find the right parts...... It's pretty random but irritating.....

My questions are, when trying to get around 10-20ma at the .43ohm emitter resistors for biasing , am I doing this at idle with no signal????
yes, idle, no signal.
Are all the .43 ohm resistors supposed to read similar????
There will be some variance due to slightly different Vbe voltage. In production all transistors will come from same batch so will be reasonably close.  The 0.43 Ohm emitter degeneration resistors actually force the transistors to share current evenly. Without the 0.43 Ohm resistors even a few mV lower Vbe would hog more of the current with bad result.

At higher output current levels the few mV Vbe difference is inconsequential, and sharing is good.
And I have 43mvdc at channel one speaker terminals and 16mvdc at channel two's....Is this normal or acceptable???? Is there something I should be looking at???
0.04V DC is not going to bother your speakers...  If it bothers you chase the different voltages around the circuit to find the difference, but don't blow anything up.
Thanks for all the help!!!!!!!
we all participate in your success (or failure). I'll take success.  8)

JR
Gene's should be similar but it's older than mine by a couple of years....


My sensor that goes to the ecm is at the intake manifold by the water neck for the upper radiator hose....

The gauge sensor is at the exhaust manifold......between 1&3 plugs.....

I was reading my manual and was told to remove the sensor plug and jump the two terminals to see if the reading would show normal.....It does.......

Getting close.......
My Mongrel.....lol......93 GMC Yukon.....
 
Back
Top