diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Solid State (http://www.diyaudio.com/forums/solid-state/)
-   -   Odd LM3886 Application Question (http://www.diyaudio.com/forums/solid-state/163240-odd-lm3886-application-question.html)

general 15th March 2010 07:09 PM

Odd LM3886 Application Question
 
Hey Folks
Odd application of a decent audio amp chip!

I'm trying to drive some largish magnetic coils at about 30-50 Hz. Impedance at this frequency is about 4 ohms, and I thought I'd use a LM3886 chip to get the signal I needed as I had one kicking around. Now, here's the trick. I need about a 5A offset, and then another 2A sinusoid at 30-50Hz on top of this.

So, I built the Single Supply circuit as shown in the application notes, but left out the in and output caps (see above regarding requiring a DC output). Tried balancing my input voltage on the + and - pins but I still get quite a crazy rail voltage going out, which I'm sure the 4700uF cap would smoke but unfortunately I need that DC bias.

I'm having a bit of trouble balancing the "0V" point using the diagrammed 91k, 100k voltage divider into a 2N3904 transistor. Anybody with a good suggestion on balancing this, or otherwise, I would love to listen.

Thanks in advance

dangus 17th March 2010 12:51 AM

Maybe a single-ended amp (of sorts) is better. Create a constant current source (or sink) and add the sine wave to the reference input. Or look at Nelson Pass's class-A amp designs.

tomchr 17th March 2010 12:58 AM

I suggest running the LM3886 off of a split supply (e.g. +/-20 V). Then use an op-amp connected as a summing amplifier to create the DC offset and add the signal.

~Tom

dangus 18th March 2010 07:00 AM

Driving such a load from the LM3886 will require more than one of them. 5A + 2A comes awful close to the current limit spec, if you refer to the sacred text, I mean data sheet. A constant current of 5A through the LM3886 will make it rather warm, too.

Here's a circuit that should work if scaled up a bit. I'm too lazy to troll through a power transistor data book and find something with an appropriate SOA, but you'll obviously need a hefty power transistor, possibly several in parallel, probably a driver transistor as well, and heat sinking.

This application also resembles an AM transmitter modulator. What I remember from my ham radio books is that you'd stick a transformer secondary in series with the plate supply of the power amp, and drive the primary with an appropriate audio amp. That might be an easier approach... use an off-the-shelf audio amp to drive a 1:1 or 2:1 isolation transformer. Saturation could be a problem with a normal AC transformer due to the large DC current, though; maybe it'll need a gap or something.

Eva 18th March 2010 08:23 AM

Have you considered PWM? (Sort of class D amplifier). It allows to control coil current, both AC and DC, without wasting a lot of heat.

Given the operating frequency range (30-50Hz), a high PWM frequency is not required (say 25khz to avoid audible noise) so fast switching is not required either.


All times are GMT. The time now is 07:43 AM.


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