Small Power Regulated Power Supply

[martin clark]

Another thing - since the o/p Z of an LM317 depends heavily on current draw, isn't all this fine tuning a bit pointless anyway?

Fair point, but that depends on how sensitive your load is to supply Z. Some circuits simply are not.

When I wrote the 317 page quoted above I suppose my underlying assumption is that what the 317 is good at is ripple-rejection. The suggestions are purely about maximising that parameter along with load rejection for what is a 'jellybean' part.

The 317 is not a very-low-noise reg; if that matters I generally use a different approach, or at least some sort of clean-up after the reg. But in bang-for-buck, the humble LM317 is pretty good.

Ditto with other aspects of linearity and good behaviour, for which the 317 is not the last word. But I agree very much with AndrewT's comments on jbau's investigation.

[teleman]

I've just finished building the PSU as in the circuit above, (1st posting) but thought I'd ask for a little tech. detail before i power it on to....

In the circuit diagram above, the diodes are reversed for the + & the - rails.is this correct?

secondly,should the "clamping" diodes (not shown) between the Adj.pin & Vout also be reversed for the - supply rail?

I am rather confused..so any help/advice would be welcome!

Thanks.

[mjf]

hello.
only look at the datasheets of the regulators..............there are the examples - with the diodes(figure 3) ,and so on.
http://www.datasheetcatalog.org/data...r/DS009063.PDF
greetings...............

[teleman]

No worries,I've found it!
thanks

[teleman]

Continuing the saga..part 2

I finally got down to testing this PSU (please refer to the above schematic) after I managed to get a suitable casing.The case actually made it very easy for me to bolt the regulator ICs on either side providing ample cooling at the same time.

I've constructed the circuit exactly as given,but have added additional diodes-1N4002-between v adjust & output of each regulator IC

I've also added 200mA fuse to each secondary ac supply rail...which I'm not sure,if it's a good or a bad idea!??

Q:There's a slight drift on both rails..but seem to almost lock on either 8.99 or 9.02 volts!! Why is this?

Any clever thoughts or suggestions?

Thanks!

[mjf]

hello.
.......if the 200ma fuses blow you can go up to 500ma (transformer rated 500 ma, iirc).
the slight drifting could be caused by slowly drifting mains up and down.......
or a little drift with temperature.......
solid alu case!
greetings

[teleman]

Thanks mjf.In case you've wondered,I'm using a 9-0-9v trafo instead.There's still unequal voltages ...since after an hour or so it does vary considerably for instance,+9.05,-9.10 or some other voltages no matter how much I adjust it!
Does not seem to settle to a fixed voltage! I did soak/run it for many hours to no avail.

C1b & C2b are 0.1uF/450v caps? I've followed the original circuit,bar for the 2 additional diodes across reg.o/p & voltage adjust trimmer!
I did change the elects caps ie:2x2200uf to better ones....didn't help much
Is it ok to have these fuses in the secondary? I've got 2x200mA slow ones.

The casing is actually from an old S.M.PSU for a 12v cigarette lighter.So don't think this should be a problem, Do you?

[teleman]

After many hours of tweaking,it's now showing some signs of improvement & has stabilized considerably! albeit for very slight fluctuations in the 2-5 mA region. I'd like to test it with some sort of resistive/capacitive load before I actually plug this into one of my projects,as I don't want to fry many hours of hard work work!

What value resistive/ capacitive load should I run this with?

Any suggestions? Thanks

[teleman]

Taking lgreen's suggestion,I read through the data sheet for LM317, although the whole sheet refers only to LM117?.Anyway,here are few points I would like to clear up before connecting to a load.

Acc.to the data,the clamping diode between adj/o/p is only necessary if the cap is > 10uF & o/p voltage is > 25v! So would having one for a 10uf/9v result in malfunction?

It is clearly stated that,with no load the o/p voltage would rise,'cause the regulator(s) require an absolute min.of 10mA load current!

So,now I understand the reason for this rise in voltage on idle!
Here's a quick recap on the different/additional comp.I've used.

1.Additional Clamping diode between trimmer pot/IC adj. pin)

I.Trimmer pot=5k
2.C3a/C3b=10uF/25v
3.C5a/C5b=470uf/25v
4.C6a/C6b=0.1film cap
5.2x200mA Slow blow fuses on the ac secondary (These do have a resistance of 9ohm/fuse)

I have a 2 ch.digital Oscilloscope,which has a max.20 d.c range. however I'm not any good at using these generally.Can someone please give some guidance/advice on what to look for when measuring dc voltages.Yes, I've 2 probes too!

[mjf]

hello.
annother theme.........the tab-heatsinks of the regulators are connected with pin 2 (mid pin) . do you isolate the regs against the aluminium case.........to prevent a short?