Gainclone Power Supply

[iandrum]

Hi,

I'm new to the DIY thing, but eager to learn and hopefully produce something I can be proud of.

I'm looking to build a gainclone amp with +-35V power supply based on the pages below:
http://sound.westhost.com/project19.htm
http://sound.westhost.com/project04.htm - however only using one rectifier/set of capacitors to provide a single supply

To make sure I don't do anything stupid, I want to have every planned out properly before I do anything and make sure I know what I'm doing.

At the moment I'm getting a little confused over component selection and as much as I'd love to be cheap and reuse things I have lying around it's probably not the best idea.

In my searching around I found a kit which I think maycontain a suitable rectifier and capacitors, as well as having the benefit of a PCB.

For the power supply, would the kit shown below be suitable with a 120VA 2x25V transformer?

http://www.quasarelectronics.com/3186.htm

Any other comments or advice would be much appreciated.

[yusuf]

Should work. However, small capacitor bypass along with larger one would have been better. You may also want to check other options like audiosector or digi01. Whatever kit you choose, ensure that they have sufficient track width to handle required current.

[AndrewT]

Hi,
you can hardwire your own PSU. You don't need a PCB, that just adds cost and makes it look neat.

All the info is in the ESP site.
transformer + rectifier + smoothing caps for dual polarity.

Optional:-
Add interference attenuators at the input and at the transformer.
Add spike attenuators at the transformer.
Add snubbers at the output of the PSU.
Add rail fuses after the smoothing.
Use a separate PSU for each amplifier channel, i.e. extra rectifier + smoothing.

BTW,
most cap banks perform better when made from parallel sets of smaller caps. Sometimes the same capacitance is cheaper when using small caps. eg. 8off 4700uF to give +-18.8mF is probably cheaper than buying a pair of 20mF caps. (but 10mF and 15mF are fairly cheap).

[iandrum]

If I go with individual components, would the following be suitable for the rectifier? I read somewhere that it needs to be rated for 35A. Does this refer to the Ifs current on the specification?

http://cpc.farnell.com/jsp/search/pr...p?sku=SC2KBB10

Also, for the transformer, is there any reason I can't use a transformer with 2x115V on the primary instead of 1x230V (obviously connecting the two coils together)?

[phn]

I think this--a 100 pct Gaincard clone--is a better beginner project. http://www.geocities.com/rjm003.geo/...io/diy_gc.html

[AndrewT]

Originally posted by iandrum
If I go with individual components, would the following be suitable for the rectifier? I read somewhere that it needs to be rated for 35A. Does this refer to the Ifs current on the specification?

http://cpc.farnell.com/jsp/search/pr...p?sku=SC2KBB10

Also, for the transformer, is there any reason I can't use a transformer with 2x115V on the primary instead of 1x230V (obviously connecting the two coils together)?

Hi,
the 115+115Vac is effectively the same as 230Vac.
Are you on 240Vac? that will increase the input AND output voltages. Make sure you connect the phase of the two primaries correctly. Use a mains light bulb for all your initial testing and any future modifications (not for high power testing).

Continuous forward current is the spec you should check.
A 25A or 35A bridge rectifier will be suitable for a dual polarity supply. The PIV should be at least 3times the secondary Vac. so use 200V.

[iandrum]

Actually, surely this bridge rectifier would be needed.

http://cpc.farnell.com/jsp/Component...?sku=SC2KBB20R

The voltage output of the transformer is +-25V, which means 50V across the bridge, so the 40V rating on the previous post is insufficient.

Also, is the quoted secondary voltage of transformer a rms value? So the maximum voltage across the bridge is 1.4*50=70V? Thus an 80V rating is ok?

[iandrum]

Ok, so this would be the appropriate bridge rectifier? 200V 35A.

http://cpc.farnell.com/jsp/search/pr...questid=263413

Having realised that main will be 240V and not 230V, this will mean that the DC output will be over 35V (approx 36.5V by my calculations). Obviously I don't want to fry my ICs.

Is this likely to be a problem, or do I need to use a different transformer, or find another way to reduce the output slightly?

Thanks for the help so far, I've learnt a lot already.

[AndrewT]

Hi,
the peak voltage on the smoothing caps can approach 41.3Vdc when the 230/25V transformer is unloaded.
As one adds amplifiers/loads the average DC voltage drops and the ripple increases.
When the quiescent current is set that finally defines the DC voltage and ripple.
See
http://www.diyaudio.com/forums/showt...62#post1171062

The normal operating voltage is likely to be around 38Vdc.

[iandrum]

I've obtained all the components I need and built the power supply. As you said, it's output is pretty much +-38V (the +ve did seem slightly higher than the -ve on my quick check though).

The amp chip data sheet says:
Supply Voltage |V+|+|V−| (Input Signal) 84V

This means it can safely take +-42V without dying on me, so the power supply will be fine for it, right?