My take on a discrete shunt voltage regulator

Status
Not open for further replies.
Jump to Latest
#463
Just the individual boards. The price was higher for compound boards.

Yep, the wiki here on diyaudio seems to be stuck in some sort of weird state that no one can edit it anymore. After a few days of waiting I just created a web page that I can have some control over.

If any moderators are reading this, who can I contact to get the wiki fixed?
 
#466
ikoflexer said:
Just the individual boards. The price was higher for compound boards.

Yep, the wiki here on diyaudio seems to be stuck in some sort of weird state that no one can edit it anymore. After a few days of waiting I just created a web page that I can have some control over.

If any moderators are reading this, who can I contact to get the wiki fixed?
Click to expand...

Ok, thanks. I'll use your web page! 🙂

I know there is some discussion on these pages about what to use as a pre-regulator but is there a basic 'ideal' for rectification and capacitance to sit before the reg?

Thanks
 
#469
Magsy said:
I know there is some discussion on these pages about what to use as a pre-regulator but is there a basic 'ideal' for rectification and capacitance to sit before the reg?
Thanks
Click to expand...

Oh, I don't know, everyone seems to have their idea of what ideal is. I have used both CRC and CLC before the regulator and the final ripple was below 100uV. More like 10-30uV. It's basically the limit of what I can measure. I'd recommend you have a look at psud2

PSUD2

a great little program that you can simulate a psu easily. For instance you can aim for about 100mV input to the regulator, and see what CRC that you have may give you such a result.

Of course, if you have access to an oscilloscope you can just build it and test it. There is lots of info in this thread and the simplistic salas thread, the only disadvantage being that it's a lot of reading 🙂

Anyway, you can always ask more direct questions here too.
 
#471
😎 good news, having stuffed the board last night, finally got a bit of time this evening to power it on. used one LED only because my LEDs have about 2v Vf. For R2 I used a 10K variable resistor, set initially to about 500R. R9 and R11 = 30k. R7 = 470. The load was passive, 470R. Don't take these values as set in stone. I'm only giving an example of what I did. OK, I set the negative input voltage to about -28V and let it go. Then slowly turned R2 until the current limit went to about 400mA. The output voltage stayed stable at almost -16V.

On the scope, the same result, hard to tell exactly how large the ripple is, but definitely below 50uV. This is it, worked from the first try, the board is good, remote sensing and all.

Onwards we go. Next thing on the agenda, getting definite pricing from the pcb manufacturer.
 
#472
Hi,
I have been ill last week so didn't see the reactions till now.


how about just one regulator across the +ve and -ve power pins.
Then use a virtual ground between the supply rails for a zero volt reference for the input signal.[
Click to expand...

@AndrewT
It's an idea. But would the -ve (when using +ve CCS) not suffer from the ripple on -ve, as it will not have a CCS there?

Regards,
Marcel
 
#473
Marcel, OK, here's what I think about your proposal. In principle you can have just one shunt element connected between the positive and negative rails. But, then there must be one reference+amp to drive this shunt element, and this part of the circuit also has to be connected between the two rails. The results is that the voltage between the positive rail and the negative rail is kept at some constant value, but not the voltage of each rail wrt to the ground. In fact I would expect it to be quite unstable.

In my opinion this is not how you should address grounding problems. Instead, opt for a properly designed pcb and then properly hooked up to the load, which in your case are opamps.
Click to expand...

Hi Ikoflexer,

I understand what you say. The Shunt will keep it's voltage (say 30V) but you can't tell where it will keep this voltage in the headroom between the two CCS'es. Will probably depend on the CCS'es, will be very hard to get the exact same current/voltage for both of them. Not a problem for the opamps power supply itself I think, but it will shift/ being undefined compared to the signal ground then. This could be a problem.

The reason to come up with this was the thinking if it would be possible to put 1 shunt mosfet per opamp directly at the opamps. Hoping that with the shunt directly on the opamp the use of capacitors can be avoided. 2 shunt elements there will take to much room and would require a power gnd (that I don't want).
Other thing I asked myself was, if it could help making the output impedance ever lower.
As in my experience this indeed is very important to get a good sounding result.

I will look at the setup of Hifinut.

Thanks for your reply,

Marcel
 
#477
Not too late. Price coming up, should be less than $10 a piece. I'm trying to see if the pcb manufacturer agrees to making a small number so we can test it.

Edit: What I mean is, being the first time dealing with this pcb manufacturer, it would be nice to test a sample of their fabrication. I've already tested the board as made by me.
 
Last edited:
#478
I've updated the site with pricing information and more about the design. If any of you wants to change the number of boards the only way to do it is to delete yourself from the list and then add yourself again with the new number. I will also start working now on a document about how to assemble and test the regulator. The document will be on the site when finished and I'll post a link to it.
 
#480
Telstar;2044404 It would be good if you could test the produced pcb ;)[/QUOTE said:
Of course, no other way would be acceptable. Let's wait for a bit for people to make a solid decision now whether they want it or not, and then we'll close the GB registration and go ahead with the order. That'll give me some time to work on the assembly document and BOM.
Click to expand...
 
Status
Not open for further replies.
Top Bottom