diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Solid State (http://www.diyaudio.com/forums/solid-state/)
-   -   Amplifier Design for Fractional Ohm Loads (http://www.diyaudio.com/forums/solid-state/160964-amplifier-design-fractional-ohm-loads.html)

SlunkBoy 9th February 2010 06:09 PM

Amplifier Design for Fractional Ohm Loads
 
I recently cobbled together a plain vanilla Class AB amplifier with an opamp front end and a push-pull Darlington output stage in order to drive a fractional ohm resistive load. The opamp is a TL082 and the output pair is comprised of a TIP120 and a TIP125. The supply rails are +/- 8 volts with about 20,000 uF of filtering on each rail.

I understand that circuit resistances that can be ignored under normal conditions become significant when driving very low resistance loads and I suspect I am losing a considerable amount of power in the various leads connecting the power supply to the transistors and subsequently to the load and back to ground.

Given these conditions, I am still curious as to why the amplifier can only manage output swings of about 1.5 volts before clipping. Very little heat is being dissipated by the amplifier and the Darlingtons should be capable of 8 amp pulses. While I am not able to accurately measure the resistance of the load since it is so low, the .1875 ohms suggested by dividing the 1.5 volt peak output swing by the 8 amp collector current limit of the output transistors seems reasonable.

While I have tried paralleling the output transistors, the necessary inclusion of current sharing resistors (.47 ohm was the only handy value) has made for a difficult comparison.

Does it stand to reason that the collector current limit of the output transistors is causing the clipping?

Are hardier transistors and thicker, more direct leads the solution?

Any insight would be much appreciated.



Sincerely,
Nick

Mooly 9th February 2010 06:29 PM

Darlingtons suffer from a 2 vbe volt drop and at high current low supply voltage that's significant. So you could lose 1.5 v swing or more there.
Would have to see a circuit :)
Driving low impedance loads is fraught with difficulty... the transistors will exhibit severe gain droop at those levels you talk of. I suspect you need far higher drive currents than the TL082 an provide and multiple parallel pairs of outputs.
You need to scope along the wiring to see if you have power losses... or use a DVM and drive the amp at DC and measure volt drops along the wiring etc (ohms law) but make sure the heatsinking etc is up to the job.

event horizon 9th February 2010 06:31 PM

1 Attachment(s)
I think you need to seriously think about increasing the amount of storage capacitance :D

Here is a nice simple way of working out the ripple voltage.


From what i can see, that is likely to be the culprit. In other words the voltage is sagging so much that you end up with damn near no output.

You are also likely to blow those transistors if you expect each one to carry more than the maximum current rating. Be careful as if you increase the capacitance you'll have the capability of doing that.

This is all of course assuming your transformer can supply enough current.

SlunkBoy 9th February 2010 07:01 PM

Mooly and event horizon,

Thank you both for your replies. I allowed for the opamp and biasing losses in the design and the same amplifier driving a 8 ohm load is capable of reasonable voltage swings without clipping.

From what I've been able to measure with a voltmeter, the power supply seems solid, though I suppose it could be suffering from instantaneous sag. I'll hook the scope up to the rails tonight and investigate.

The gain droop idea seems to be the most promising avenue to pursue. I had been assuming a gain of about 2,000 for the Darlingtons, which the TL082 should be able to drive, but I hadn't taken into account the fact that that gain drops at higher currents. My design currently has the opamp driving the output transistors directly, but I'll incorporate a drive transistor tonight and see if that improves things.

Thanks again for the great suggestions.



Sincerely,
-Nick


All times are GMT. The time now is 09:36 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2