Solid State Amp runs hot

Ivan Funk · From: Hamburg Pennsylvania, USA

Solid state amp question:

Leslie 760 rebuild: It was running hot.
I replaced all the electrolytic capacitors in the power supply and amp boards.
Also replaced the power resistors on the amp board.
All voltages test points test good.
The amp has zero hum and sounds really good.

I’m getting 50 mV drop on the emitter resistors. (Resistors are 0.5 ohm, 5 watt)
According to my poor match skills this is 100 mA - or 4.2 watts on a 5 watt resistor?
The resistors get very hot. The bias was even higher before the rebuild. Seems too close to the edge of the temperature limit of these wire wound resistors.

Do you think this is normal or is there a fault somewhere?

DC offset on the outputs is about 50 mV if that is relevant.

Donny Hinson · From: Glen Burnie, Md. U.S.A.

Those ceramic/wire wound resistors are made to run hot, that’s why they use them! I wouldn’t be concerned about it unless they discolor or start smoking. But if you’re worried about it, you could always parallel a couple of 1-ohm resistors.

Ivan Funk · From: Hamburg Pennsylvania, USA

Thanks for the quick reply Donny,

The resistors are new and therefor smell a bit when hot , but no discoloration and no smoke (yet).

Maybe I'll let it run a while and see if the smell lessens?

Do you mean two parallel resistors to share the load / heat?
What about upping the wattage rating of the single resistor from 5 watt to 10 watt?
Is one method better that the other?

Thanks,

Stephen Cowell · From: Round Rock, Texas, USA

Those resistors (emitter resistors) are also meant to serve as fusible components... if you put 10watters in there the next meltdown might burn the board.

Ah, the joys of SS Leslie... Mercotac! I put the crossover module into mine so I could use the combo pedal with it... these have the 11-pin cable right? Does your customer really have one of those SS Hammonds? Bought one, fixed it up... got it working completely... and ended up junking it, just not great-sounding. Still have the Leslie spinamathing and the TrekII combo pedal though. Would be a good candidate to put a real bottom rotor and a CS800 amp into.
_________________
New FB Page: Lap Steel Licks And Stuff: https://www.facebook.com/groups/195394851800329

Ivan Funk · From: Hamburg Pennsylvania, USA

Thanks Stephen,

I also have a Leslie Combo Pre-amp II so I can run anything into it.

I'll leave the resistors stock specs.

I did let it run for a couple hours and it seems stable.
They just run hot I guess. Nothing is getting hotter with time.

No oscillation.

Do you think 50mV DC offset at the output is acceptable?

Yes the Leslie 760 is 11 pin. It has an active crossover and is bi-amped for upper and lower drivers.
Here is the full schematic:

Stephen Cowell · From: Round Rock, Texas, USA

50mV is not ideal... be nice to dial that out somehow. With all that transformer isolation between stages your offset has to be coming from your final section (unclear if it's bass or treble, the amps appear identical anyway). If only one side has the offset I'd start comparing quiescent voltages between the two channels... get a gnat's ass ohmmeter and measure those bias resistors carefully. Not clear what you replaced in the amp but as you know stuff can be hurt and still sorta work. Not going to sound that great with DC going through the speaker!

Your skem shows a 9-pin... I think I have a 720 with 11-pin... I had to add the active crossover your skem shows, I bought the TrekII LCO-1A and added it. The side-channel 6x9's are silent, but at least the other three channels (woofer, rotosonic mid, and tweeter horns) run.

The Mercotac is a way to run speaker connections through a spinning mechanical contact... it's got mercury in it! Would not be allowed now of course. The Rotosonic drum has 6x9's in it and is driven by a separate channel... I'd much prefer a regular wooden drum for the 15" woofer... but I understand that the bass frequencies don't get much use from spinning anyway.

You should know about TrekII and Captain Foldback if you don't already:

https://www.trekii.com/

http://www.captain-foldback.com/Leslie_sub/pinouts.htm

He says there can be problems swapping around 9-pin Leslies... sounds like you don't have that issue.
_________________
New FB Page: Lap Steel Licks And Stuff: https://www.facebook.com/groups/195394851800329

Ivan Funk · From: Hamburg Pennsylvania, USA

Thanks Stephen,

Agreed it has to be something in that final stage.

The amp was fairly noisy and had a good amount of hum when I got it - thus the rebuild. Now it's quiet, hum is gone, and it sounds good. Goes a lot louder than a 122 or 145.

The two amps (Bass and Treble) are slightly different circuit designs.
I replaced all the power resistors around the final stage on both amps. Replaced any and all electrolytic caps everywhere.
All the 5 watt resistors are 1% except the 430ohm (2 per side) are 5%.
Just for the heck of it I replaced the 2N3055s power transistors also.
The sound improved and the noise lessened with each stage of these restorations.

The bias was way high before I started and there was DC offset. Both are improved but not totally solved.

The bass channel has -40mV DC offset.
The Treble amp is around +50mV DC offset.
Emitter resistors measure about 40mV across.

I can't figure out if this is just the way it was designed or if there's something off somewhere.

How do I get the emitter resistor bias down from 40mV to 20mV or less?
Is my math correct?

Here's a photo of the boards just for fun:

Ivan Funk · From: Hamburg Pennsylvania, USA

Thanks Quentin,

I looked those up on Mouser.
$63 a piece. I don't think I'll go that route.

Update: I got the DC offset on the treble speaker way down to 20mV. Bass side is still around 60mV. Good enough for now I guess.

Then something went bad in the crossover circuit. Low volume and distortion. All the transistors have the correct voltages on them.

To troubleshoot I bypassed the crossover and put a signal straight to the bass amp. Loud and clear. Sounds fine. So I think there's some fault in the crossover somewhere.
Anyone have any ideas?

Ivan Funk · From: Hamburg Pennsylvania, USA

(I was looking at the wrong part, those TO220 resistors are $11. Still too high though)

Quentin Hickey · From: Nova Scotia, Canada

Sorry. I deleted the my other post for te sake of not stepping on any other toes. Distortion means the transtors are ether in saturation or cutoff. Crossover distortions refers to class B biasing which this is not. Your dc biasing and ac load line are two diferent things. If a capacitor is open or short this can cause yourac liad line to swing way out of whack. You know for sure the transistors are good and are the correct ones? What I would do is power this whole thing off of an isolaion transformer and use an oscilloscope to signal trace where it is going off the rails. It's pretty ard to trouble shoot without a scope and signal generator. Maybe someone in your area has those....

I will add, make sure you have grounds where you are supposed to and that they are all common. Double check your resistances. Double check your base to collector and base to emitter diode drops on all of your transistors, but yes get a scope!! That will be your best friend
_________________
www.fessendensteelguitars.com
www.quilterlabs.com
www.goodrichsoundcompany.com
www.daddario.com/nyxl
www.bjsbars.com

Quentin Hickey · From: Nova Scotia, Canada

Any time you want to lower your base bias you need to change the series resistors so the fraction gives you what you want going into the base. For example you have vcc 10vdc. You want 1v going to the base. Your resistor.network needs to be a 9top:1bottom fraction. You would use a 10ktop:1kbottom
Higher total resistance will give more current limiting. That's important.
The equation is Vbase=(R1/R1+R2)VCC

I still recommend the to220 power resistors. Well worth the cost.
_________________
www.fessendensteelguitars.com
www.quilterlabs.com
www.goodrichsoundcompany.com
www.daddario.com/nyxl
www.bjsbars.com

Ivan Funk · From: Hamburg Pennsylvania, USA

Thanks Quentin for all the great info.

I do have a scope and a signal generator.
The input capacitor on the crossover was bad.
Replaced it and it's all working again.

Now I'm back where I started with slightly high emitter bias and some DC offset at the output.

I think I understand what you're saying about base bias and the resistor network.

How do I lower the current through the emitter resistors? (circled in red below)

Quentin Hickey · From: Nova Scotia, Canada

That's kind of t wrong way to look at it. Components will only pull the current they need. Thy don't just use all current available. If the resistors in that node are pulling excess current that you are concerned about than you need to look toward what is happening at the load. Double check everything to the right of point R. Something could be partially shorting. Point R is sort of a virtual ground, ts not really ground since it is floating. This is a balanced network where the 2 transistors in seriers provide lots of swing (42V to -43V).
Check ahead of point R. Use a temp sense gu to see if you can find any other heat sources. How does everything look when you trace it through with scope end to end?
_________________
www.fessendensteelguitars.com
www.quilterlabs.com
www.goodrichsoundcompany.com
www.daddario.com/nyxl
www.bjsbars.com

Ivan Funk · From: Hamburg Pennsylvania, USA

Excellent.
Makes sense.
I will see if I can find a short / partial short after point R.
I wonder if the blocking caps in the feedback loop are going. That could cause the offset also right?

When I trace end to end it all looks good. Just amplifies the same waveform. ...unless I'm using the scope wrong.

Anyway I'll do some more hunting this afternoon.
I think I'm getting close.
Really appreciate the help!

Ivan Funk · From: Hamburg Pennsylvania, USA

Update:
I’m getting closer and not closer.

Emitter bias down to 35mV both amps. Good enough.
I re-flowed some solder joints around various ground points.
No clue why it came down.

Anyway it still gets hot but not as hot so that’s good.
The new resistors smell slightly like a candle burning. But no discoloration or smoke. Maybe the smell will burn off eventually. Also when everything is assembled, the bass rotor (when spinning) cools the amp boards considerably and the resistors don't get hot at all.

Treble amp DC offset is way down to 4mV
Bass amp DC offset is still high at -150mV

Could it be that the speaker is part of the feedback loop so will always have some offset?
I don’t really know what I’m talking about obviously.

I’d call it good enough but I’m curious to solve the offset mystery.

Stephen Cowell · From: Round Rock, Texas, USA

IMO 4mV is OK... 150mV is outrageous, enough to hear and perhaps warm the VC.

Here's a method... heat/cool components and watch the offset change. My guess is that this should show you where the problem is.

Good luck!
_________________
New FB Page: Lap Steel Licks And Stuff: https://www.facebook.com/groups/195394851800329

Ivan Funk · From: Hamburg Pennsylvania, USA

Thanks Stephen,

I'll try that.

What I don't understand is - why the bass amp has the speaker as part of the feedback loop and the Treble amp doesn't?
I'm trying to rule this out as part of the DC offset problem.

Quentin Hickey · From: Nova Scotia, Canada

My guess would be negative feedback so it doesn't overload the system. Where as the horn is alot less likely to
_________________
www.fessendensteelguitars.com
www.quilterlabs.com
www.goodrichsoundcompany.com
www.daddario.com/nyxl
www.bjsbars.com

Ivan Funk · From: Hamburg Pennsylvania, USA

Ok. Makes sense.
I've seen that on tube amps.

I tried to trace the offset back from the speaker.
It appears in the B- 42v right after R86 (at the base of Q14)
Does this mean that Q14 is pulling too much current heating up R68? That's the resistor that gets the hottest.

I don't understand what happens at the point where the B+ and B- meet.

Thanks for all the helpful info.
I should find a back to basics course and better learn how these circuits work. Tube amps seem more simple to understand (to me).

Quentin Hickey · From: Nova Scotia, Canada

_________________
www.fessendensteelguitars.com
www.quilterlabs.com
www.goodrichsoundcompany.com
www.daddario.com/nyxl
www.bjsbars.com

Quentin Hickey · From: Nova Scotia, Canada

Its setup to minimize distortion with the base to emtter diode drop when passing the virtual ground or 0V crossing. The resistors at the bottom will take the brunt because it is always on where as the upper transistor goes into cutoff when the signal dips below zero crossing.
_________________
www.fessendensteelguitars.com
www.quilterlabs.com
www.goodrichsoundcompany.com
www.daddario.com/nyxl
www.bjsbars.com

Ivan Funk · From: Hamburg Pennsylvania, USA

Thanks for the great explanation.
Now I understand why the bottom resistors are hotter than the rest.

I still cant find, or understand whats causing the DC offset.
There aren't that many components in this side of the circuit and they all test good.

If the DC offset is -150mV should I be looking for a partial short on the B- side of things?

Ivan Funk · From: Hamburg Pennsylvania, USA

Update:

I swapped the position of two 2N3055 in the bass amp and the DC offset came down from -150mV to -58mV
Interesting.

One more question:
The over all specs for the Leslie 760 state that the Bass amp is 50 watts and the Treble amp is 40 Watts. Does the difference in emitter resistor values in the two amps account for this difference in amp wattage?
( 0.3 ohms and 0.5 ohms)

If so - can I replace the 0.3 ohm with 0.5 ohm in the bass amp? Would that bring it down to 40 watts?
The goal would be to lower the heat a little - the 0.3 ohm resistor gets the hottest.

This thing goes crazy loud so I don’t think lowering the bass amp output a bit will matter.

Tony Prior · From: Charlotte NC

Similar- The Fender Hot Rod series amps have 480 ohm 1 watters off the 6L6- G, they are mounted OFF the PC BD because they get darn hot ! One of the issues with these amps is during assembly, some of these got mounted right on the PC BD and caused pc bd burn damage. YEP, they get HOT !
_________________
Emmons L-II , Fender Telecasters, B-Benders
Pro Tools 12 on WIN 7 !
jobless- but not homeless- now retired 9 years

CURRENT MUSIC TRACKS AT > https://tprior2241.wixsite.com/website

Ivan Funk · From: Hamburg Pennsylvania, USA

Thanks Tony,

Update: I got the bias down to around 50mA and the DC offset down to around -50mV.
It runs a bit hot but not as hot as it did before the rebuild.
The sound is great and there's zero hum.
I put it all back together there's no "hot" smell when running.
I'm calling it done.
Thanks everyone for all the help.

Forum Index > Electronics	Goto page 1, 2 Next	Next oldest topic :: Next newest topic