Electronics > Beginners

RPS with from 0-30V and Current variable up to 3 A design Circuit Please....

<< < (6/8) > >>

BravoV:
With all excellent replies, advises, tips and even circuit, I guess the OP will not create, yet another new thread nagging for a new 723 based 30V 3A psu again, as this is the 5th one or 6th already ?  :-//

Its just sad to see all goodwills and helps going to be wasted again if this happened.

j395:
why not use something like LD1085 - much more robust for beginners due to built-in overcurrent/overheat protection ?
to decrease heating, use cheap buck or buck/boost module as pre-regulator, LD1085 works at 1.3V dropout - only 4W wasted at 3A

rstofer:

--- Quote from: j395 on January 11, 2019, 11:47:52 am ---why not use something like LD1085 - much more robust for beginners due to built-in overcurrent/overheat protection ?
to decrease heating, use cheap buck or buck/boost module as pre-regulator, LD1085 works at 1.3V dropout - only 4W wasted at 3A

--- End quote ---

It doesn't meet the "spec" in that the minimum output voltage is around 1.8V - see Vo Table 3  or it could be as high as 2.5V from Table 4 but I didn't pursue the difference in #18 versus #25.  In any event, the device can't get anywhere near the "spec" of 0V.

Without a pre-regulator and in the situation where Vi-Vo is about 25V, the device can only deliver about 0.5A based on Short Circuit Current.  But we want, perhaps, 35V as Vi and 0.1 as Vo so call it 35V across the regulator.  The maximum output current won't be much = Figure 7.

With a pre-regulator that can get Vi down to a reasonable level, the device can deliver the full 3A required by the "spec".  But that adds a bit of complication.  Transformer taps would be better (in my view).

https://www.st.com/resource/en/datasheet/cd00001883.pdf

"Specs" have consequences.  It's a lot easier to write 0-30V than it is to achieve the 0V end of the spec.  Same with the 0-3A.  It's hard to deliver 0.1V at 3A from a fixed secondary transformer that is delivering enough voltage to produce 3A at 30V.  I haven't seen a circuit for a linear supply served by a switching mode pre-regulator but I haven't really looked.

rstofer:

--- Quote from: BravoV on January 11, 2019, 10:07:45 am ---With all excellent replies, advises, tips and even circuit, I guess the OP will not create, yet another new thread nagging for a new 723 based 30V 3A psu again, as this is the 5th one or 6th already ?  :-//

Its just sad to see all goodwills and helps going to be wasted again if this happened.

--- End quote ---

But maybe some of the responders, like myself, are learning something about the problems created by the "specs".  Yes, I have spent some time wandering around, bouncing off of other schematics and different approaches to the problem and, as a result, I have a new appreciation for the problem.

If there was a simple design that had a decent bill of material, it would be a 'sticky' and served up as the universal answer to a DIY power supply.  Alas, it isn't simple!

As I see it, there are two choices:  First, come up with a more realistic spec or, second, plan on finding a transformer with secondary taps.  I don't know how hard it would be to wind a custom toroid.  Maybe buy one, unwind the secondary and then rewind it with taps.

There are a lot of commercial supplies that advertise to meet the spec and they tend to be in the $50-$60 range.  I don't know that they actually meet the spec but I would tend to just buy one rather than try to build one.


dmills:
Hum, 2 * 9V dual secondary transformers, 4 bridge rectifiers and filter caps (That will only need to be 16V or so rated), and 4 NPN pass transistors plus 4 diodes and some small beer stuff, build an output stage class G style, almost right out of Doug Selfs book on audio power amplifiers (You only need half of it!)?

At 3A with 12V across the pass device (more or less worst case with zero output voltage) you are looking at only 36W in the heatsink which is easily within the SOA of a single TO3P pass device, I would have thought a 1.5C/W convection cooled job would be fine.

I think you could even wrap this around a 723 as the controller if you really wanted to, but there is nothing really wrong with a long tailed pair, Vas and some passives.   

The trick is that you hang a current source (a few mA or so) off the highest rail, then use a string of zenners to switch on the higher taps as the voltage comes up.

I am not sure it is economic given what you can find a lab supply for on the used market, but I don't really see the difficulty, I might design if for shits and giggles.

This is the basic concept :



Zenners are 3.1V, transistors are probably darlingtons, current source is a few mA via the usual PNP transistor and a couple of diodes and resistors sort of affair.

Regards, Dan.

Navigation

[0] Message Index

[#] Next page

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod