lm317 and 337 regulators.

Hi and thanks for reading this post.

I am trying to upgrade several lm337 and 317 regulators in an audiolab 8000av preamp as they have a reputation of being noisy.

Can you recommend any better or lower noise types.

Would a simpler alternative be to place a small 0.03uf wima polypropylene capacitor across the output and adjust terminals which may smooth out some of the ripple.

Many thanks in advance. John.

 

Thank you

 

Felix is the voltage regulator expert around here.

There are some very good articles on the use of 3-pin voltage regulators over on acoustica website:

http://www.acoustica.org.uk

I found these very helpful and I recommend you give them a read.

 

Definitely - lots of great info there.

 

They are great regulators and I've often seen them work as well or better than fancy ones in practice.

Bypassing the adjust pin with a cap and also changing the adjust resistor for a zener (or precision voltage reference) is a good upgrade that I found made a measurable reduction n HF noise on my active xover. I'm not sure the bypass is even necessary if using the zener method?

 

In the D.I.Y section there is a thread on my modified Arcam Alpha 5 CD player. This has a pretty good explanation of what's involved with the LED/Zener method.
I'd start by adding a capacitor across the voltage set resistor as Tenson suggests. A electrolytic of 10uF or higher will do nicely. Be sure to check the polarity.

 

Hi Thanks for the replies, much appreciated.

I will definitely give the zener a try.

With this audiolab preamp, the transformer has 6 secondaries the smallest is 5 volts going up to 17 volts ac for the largest.

But the rectifier consists of 4 motorola 1N5404 diodes which are 3amps 240 v
Is there any reason why they used such large diodes, also the instruction manual says to use 1 amp fuses. I am thinking of replacing with the 1N5404 with 50 volt schottky would this be okay.

Many thanks. John.

 

I've used Schottky 11DQ10 on a number of my projects (1.1A 100v) including CD players, pre-amps, DACs etc.

But wait and see what the others say as they are much more experienced than me!

 

Welcome, John

50v didodes are very marginal for this use.

Rectifier PIV needs to be 2root2 (=2.83) x AC voltage minimum. For 17VAC that's 48volts. Probably ok but if your mains drifts high, very marginal, and losing the rectifier risks damage downstream. While you'll very likely get away with it, I'd stick with 100v - even though that rules-out some schottkys. (Personally I'd conservatively call a diode of 50v rating as fine for up to 12VAC, but be wary above that)

3A rating over 1A makes sense to a Manufacturer because it virtually costs no more but is a lot more robust at peak curents (e.g startup: full forward voltage applied to teh rectifier, but the caps are empty = high peak inrush limited only by the secondary winding resistance).

So what to use if not 50v schottky rects? Well anything described 'soft recovery' is first-call for minimum noise from the rectification process, and tends to work better than 'ultrafast' from this point of view. HTH.

ETA: One from experience for low-currrent uses up to 1A max forward current then UF4007 are a fine replacement for the usual 1N4007. Still 1000v/1A rated, and stupidly-fast 15nS recovery time minimises stored charge, so its quiet - and very very cheap!

 

Hi Thanks for the reply.

I love the uf4007 and they are my default diode of choice Did use them in a dac a few years ago and they were brilliant as well as so cheap Could not believe the price/performance ratio. Might not be suitable for this application, although there is a one amp fuse on the mains input?

I am looking at the UF5408 as a replacement, any thoughts on this. There is a guy here: http://forum.metroamp.com/viewtopic.php?f=4&t=16739
who does like them.

Man thanks. John.