advertisement


Did anyone build a "SSR03 Super Regulator Power Supply"

fred sonnen

fred sonnen

pfm Member
as above: Did anyone build a "SSR03 Super Regulator Power Supply"?
I will do and perhaps I will have some questions befor and durung the assembling....

greetings
F.S.
 
I will too in a few weeks. I've built more than a dozen of the Andrew Weekes Super Regs which are very similar and there are ways to improve both which I will be trying. Post your questions here and I'll try to help when I have time.

John
 
Thank you
I will build a 15V PSU. I can choose between a LM431 and LM329 SSR03-solution.
What are the pros and cons of these?
The difference are only four components...
How can I influence the delivery capability of current?
Is that even possible?
I am planing to use a 230/18V and 50 - 80VA Transformer
 
Last edited:
I didn't build a SSR03 but a Sigma22 ( https://www.amb.org/audio/sigma22/ ). Quite straightforward to build and a fully discrete design. I use it to power my Project Phono RS and it works a treat. Support through the AMB forum is first rate should you run in to any issues so maybe an option to look at instead of a SSR03.
 
fred sonnen said:
Thank you
I will build a 15V PSU. I can choose between a LM431 and LM329 SSR03-solution.
What are the pros and cons of these?
The difference are only four components...
How can I influence the delivery capability of current?
Is that even possible?
I am planing to use a 230/18V and 50 - 80VA Transformer
Click to expand...
What current do you need? What current does it deliver?
 
Hi Fred,

First off is the question about choosing between the TL431 and the LM329. There are two key factors to consider, Dynamic impedance and Noise voltage. Comparing the graphs the dynamic impedance of the TL 431 at the 6.95 volts of the LM329 is about the same at 0.7 ohms but maintains this to a slightly higher frequency though the subsequent filters before the opamp cut in long before these frequencies. Not a lot to choose between them. Noise wise the TL431 is about twice as noisy as the LM329 when set to the same voltage of 6.9v. As long as you are wanting a regulator output voltage of more than 6.95 volts then I would go for the LM329. It is also simpler in implementation and doesn't need care with load capacitance to ensure stability.

The max current of this regulator is set by the capability of the LM317/337 tracking pre regs. Their absolute max current is approx. 2 Amps. If they aren't used then the pass transistors are capable of more than 10 Amps continuous as long as the current sources feeding their bases can deliver the approx. 200mA of base drive they will need. This is a lot and will compromise performance, particularly if the load current is not consistant. You can replace the LM317's with FET cap multipliers and use Darlington transistors in place of the output transistors to improve performance and reduce the base drive needed but careful design and a good understanding of actual load current is a must. As things stand 1.5 amps is a sensible max current of the design.

If at all possible can you measure the load current requirement over time of whatever you are planning to drive with these regulators? You'll not get a sustained 3 Amp delivery from 50VA 18 volt transformer. 300mA would be more realistic.

John
 
john.luckins said:
Hi Fred,

First off is the question about choosing between the TL431 and the LM329. There are two key factors to consider, Dynamic impedance and Noise voltage. Comparing the graphs the dynamic impedance of the TL 431 at the 6.95 volts of the LM329 is about the same at 0.7 ohms but maintains this to a slightly higher frequency though the subsequent filters before the opamp cut in long before these frequencies. Not a lot to choose between them. Noise wise the TL431 is about twice as noisy as the LM329 when set to the same voltage of 6.9v. As long as you are wanting a regulator output voltage of more than 6.95 volts then I would go for the LM329. It is also simpler in implementation and doesn't need care with load capacitance to ensure stability.

The max current of this regulator is set by the capability of the LM317/337 tracking pre regs. Their absolute max current is approx. 2 Amps. If they aren't used then the pass transistors are capable of more than 10 Amps continuous as long as the current sources feeding their bases can deliver the approx. 200mA of base drive they will need. This is a lot and will compromise performance, particularly if the load current is not consistant. You can replace the LM317's with FET cap multipliers and use Darlington transistors in place of the output transistors to improve performance and reduce the base drive needed but careful design and a good understanding of actual load current is a must. As things stand 1.5 amps is a sensible max current of the design.

If at all possible can you measure the load current requirement over time of whatever you are planning to drive with these regulators? You'll not get a sustained 3 Amp delivery from 50VA 18 volt transformer. 300mA would be more realistic.

John
Click to expand...

A series pass transistor could be added to the LM317 to sort the tracking pre reg issue out but frequency compensation issues would almost certainly need changing to allow a Darlington at the output...
 
john.luckins said:
Hi Fred,
First off is the question about choosing between the TL431 and the LM329.
...
If at all possible can you measure the load current requirement over time of whatever you are planning to drive with these regulators? You'll not get a sustained 3 Amp delivery from 50VA 18 volt transformer. 300mA would be more realistic.
John
Click to expand...

Hallo John, all the best for 2018 ;-)

Thank you so much for your explanation!
As I will use 15V SSR03 output I will take the LM329 solution.
As for the current. I will not change the design of the SSR03.
If I choose a 230/18V Transformer to get 15V out of the SSR03, is there a practical difference whether I use 50VA or 80VA Transformer?
 
Happy New Year Fred...

The practical difference between 50VA and 80VA are due to the load current they see. You really need to know what this will be to know for sure either is sufficient. I would always go for the largest VA rating practical as the percentage regulation will be better, meaning less voltage drop under load. Unless using choke regulation the current peaks as the rectifier diodes conduct will be more limited by the series resistance of the lower VA rated transformer. All other things being equal I have always found a larger rated transformer to sound better. I can't fully explain why that should be.

John
 
So far I did some progress in regard of brining parts on the pcb ;-)
For some parts I am uncertain about the installation direction:
Farnell Nr. 1572146 and 1215508

I do need to solder 4 parts ( LM317TG, LM337BTG, D44H11, D45H11) which need a heat sink.
How do I connect these parts?
Do I need thermal paste?
And do I need some kind of isolation between the heat sink and the active parts?

an advice would be nice ;-)
Thank you
 
Last edited:
Soldering is done.
I have a 18v transformer. Output of the ssr03 is planned to be 15v.
Now I want to test my work �� what parts do I need to do so?

Are there measuring points on the board availability to measure individual blocks of the board?
 
fred sonnen said:
So far I did some progress in regard of brining parts on the pcb ;-)
For some parts I am uncertain about the installation direction:
Farnell Nr. 1572146 and 1215508

I do need to solder 4 parts ( LM317TG, LM337BTG, D44H11, D45H11) which need a heat sink.
How do I connect these parts?
Do I need thermal paste?
And do I need some kind of isolation between the heat sink and the active parts?

an advice would be nice ;-)
Thank you
Click to expand...

You will need TO220 mounting kits which you can buy including silpads and the plastic insulating bushes. If you get silpads you don't need thermal paste.
 
You must log in or register to reply here.


advertisement


Back
Top