THAT1512 / Transformer 'blind test' mic preamp

[DaveDC]

Not sure how your scope works but can it do a trigger view with the trigger source set to line? That might reveal your mains waveform safely. With no further clues, I suspect your mains is flat topped unless your transformer is saturating. It should not saturate with no load, I’d think. I could be wrong though.

 

[jamesk]

@DaveDC Yes, good idea, I'll do this when I get the chance, possibly tomorrow.

It's driving me a bit mad now because I can't see what's wrong with the design, it should work. I've built lots of linear power supplies before and never had so many problems. Anyway, it's the name of the game with DIY electronics!

 

[DaveDC]

You are using a proper scope probe, yes?

 

[jamesk]

@DaveDC I'm using a proper scope probe, yes. I checked and although my scope has line triggering, I can't view the source. I checked just the transformer again, though and that's definitely what the unloaded secondary output looks like with nothing connected so I wonder is that's just how my mains looks, as you suggested. I'll open something else up when I get the chance and scope the transformer secondary to confirm.

I've worked out why the voltage is high, though. I checked the datasheet for the transformer and the 18V is the full load rating. With no load, it outputs 21.4V, so that's my 30V peak I'm measuring. Perhaps I should have gone for the 15V version, which outputs 17.8V with full load. That would give me 64V input to the +48 regulator at full load and 21V to the 17V regulator. That gives 4V headroom for the 17V regulators, which is 1V more than the datasheet specifies as the minimum. When designing it I didn't consider that the 15V was a full-load rating so thought this would be cutting it fine, but it looks like it would have been okay.

 

[jamesk]

I should just swap the transformer, actually. The regulators are going to be dissipating a stupid amount of power, otherwise, as you said.

 

[DaveDC]

Probably a good idea to dial back the transformer voltages. Right now your supply is about 55% efficient which is awfully low. But good to know that you have an explanation. Now it is a simple matter of adding and checking the remaining circuitry.

 

[jamesk]

@DaveDC Yes, I've ordered a replacement regulator so will test the supply once that arrives. If it's all working correctly, I'll order the replacement transformer. Then maybe I can finally get on with the preamp itself.