JSR-03 parts calculations, options, etc.

[raq]

Hi all,
Hi P-A,

I hope you're all doing great.

I'm finally back to building one of the JSR-03 kits I got back a little while ago, and I'm hitting the same brick walls as before. I guess I don't find the instructions at all intuitive... I strongly believe one PDF well put together document explaining the kit construction and options is a must, especially as this is for money, not some online freebee.

For the first PS I am trying to build (there's going to be 4 overall), I need 7.2V out. Amperage could be either about 0.5Amps, or twice that. For the sake of calculation, let's assume 1Amp.

Sans that (good PDF manual), here are a few questions I have:

1. Needing 7.2V out, DZ1 has to be 3.6V - is that correct? I decide to try a blue LED which I measured to be very close to 3.6V - I just need to insert this LED instead of DZ1, right? What is the correct polarity of doing this? I don't need to reverse polarize the LED, do I?

2. What is the value of Vref in this case? Can it be anything? I have some very quiet LT1034 1.2V and 2.5V, which are two zeners in one package, either 1.2 or 2.5 and 7V - can I use these (I just want to use one reference in the package)? Can I use a LED? How do I connect a LED? Can this just be a series voltage reference, a shunt, or either? Which resistors need to be removed in case of using a shunt Vref?

3. What are all the outputs needed for? In case I just want to use this conventionally, which X output do I use? And then, which R and C are needed? (R15/C13? R16/C14)?

4. Could you just briefly explain which Voltage reference, Zener, LED, etc. are needed and how they are calculated? I understand that DZ1 is half Vout (due to R7,8,9), but what about Vref at + input of opamp? What about the red 5mm LED? Or the other LED which does not have a label (!)?

Thank you very much in advance for your assistance.
Best regards,
Radu.

[peranders]

I'll get back to you after the weekend since I'm surfing on an iPhone. Meanwhile you can check especially the JSR01. You have there several links to super regulators and how they work but I will help you when I'll get home.

[raq]

Per-Anders,

Thank you - I'll be on stand-by for your input. In the meantime - enjoy your time away wherever you are! (vacation?...)

Best,
Radu.

[peranders]

1 One or two red LED's can bu used or a blue LED or a zener.

2 Use 2.5 V reference if you can choose. R4-R6 aren't used. A LED can be used and must be biased in forward (should be lit)

3 If you don't want to use the sense inputs just short them on the solder solderside.

4 You can use any voltage reference as long as the inputs of the opamp will be within it's common mode limits (see datasheet of the opamp). AD825 has a common mode range of 1.5 V over gnd and 1.5 V under supply voltage. The reference voltage should be well over 1.5 V. The gain must be set to 7.2/2.5. How it's calculated is written in my text. See the section "Circuit description" and "Feedback network"


You have lot's of info about the super regulator with detailed explanations on the net. One good source is here
Super Regulator Links

[raq]

P-A,

So, I gather that while DZ1 definitely has to be half Vout, the Vref (IC2) voltage is a function of the amplification factor the opamp is set to run at (R8/R9+1).

That being said, wanting to run this at 7.1V, I can use the blue LED for DZ1, and some voltage reference, provided adequate R8/R9. If I were to make one of these a trimmer pot, which one should it have to be? I am guessing R8. Or: trimmer cannot work, as R7 has to R8 II R9?...

Please let me know. Thank you for your continued support.

Best,
Radu.

[peranders]

If your application isn't critical with exact voltages I'll recommend you to have fixed values. The best choice for trimpot location is at the voltage reference if you plan to use a LM431 or similar. If you plan to use a voltage reference with a fixed voltage then you have to tune the gain and this should be carried out through adjusting the gain a little. Don't let the trimpot to be alone at a position. Put the pot in series with a fixed resistor.

[raq]

Dear P-A,

I think I got it to work - I am really happy, and thank you much for your help.

There seems to be one problem, though, and I am not sure how far reaching or whatever. At first startup, the red 5mm LED would light and then turn off after half a second. The series resistor to it was 4.9k, value that should have provided roughly 1mA current through. Suspecting too little current thorough to keep it lit, I chnaged R13 to 1k.

With these values, both the red LED and the blue LED (the 3.6V reference) turn on at startup, then after half a second, they both turn off. The voltage out seems to also stabilize at a lower voltage than when R13 was 5k, from 7.18V to 6.93V.

Do you know what could be happening? I guess there are some transient phenomena taking place before the kit stabilizes, but what I don't know if this is functioning correctly at all.

My assumption is that both LEDs would need to be lit - isn't that a fact?

Please let me know what you think. Upon your request, I can send a schematic with the values I used for everything.

Thank you,
Best regards,
Radu.

[raq]

P-A,

A little bit further information. After my previous post, I changed the opamp from MAX410 to LME49710, and both LEDs started lighting. The voltage out seemed to be off calculation, at a little bit over 6V (expected to be one volt more than that). In any case, with the MAX opamp, at least the voltage seemed to be closer to what I calculated (MAX410 gives 7.18V out, LME49710 gave 6.1V). Then I let it sit a few minutes (having something to do), and now I started it off again, and neither LED is lighting (!). At the same time, the voltage out went up to around 6.4-6.5V.

I don't understand how a circuit can be so fluctuant in operation... I am assuming I am keeping it very close to some threshold, and it keeps on bouncing right above and right under the threshold every time I put it on.

Please let me know what you think. Here are the values I used (I might send over a schematic later with values marked up for easier reference for you): R3 = 3.5k; R7 = 650R; R8 = 1.87k; R9 = 1k; R13 = 1k; opamp fed from the output, R11 used; voltage reference is LM1034/2.5, with 2.5V reference and 6uV noise; with the value chosen for R3, it should be biased at almost 1mA, which is well within it's specified operating conditions).

Thank you,
Radu.

[raq]

P-A,

Alright, so I think I got it to work. I have it wired to my default lab PS, based on an LM317, and was feeding it what it will be fed when it lands in my DAC, that is a little bit under 10V (~9.5V). I believe that was too little room for drop-out for this regulator. As soon as I lifted that voltage to well over 10V, the regulator started working fine with both opamps (LME49710 and MAX410) - both LEDs are lit and voltage out is the same, and as calculated, 7.19V.

Thank you for everyhting, and sorry for the multiple posts - at least I can say it has been solved and without much trouble for you (thanks to the difference in time US / Sweden)...

Let me know if you want more details on my current setup. I politely suggest you'd build a database with verified values / opamps / voltage references - wouldn't that be super-useful to newbees? I think people would be very grateful if you'd do that.

Best regards,
Radu.

[peranders]

I'm about to collect variants of output voltages and now it's quite easy since the cad program I use supports it.

The circuit must have proper input voltage in order to work good so when you are in the grey area you can expect anything.

Could you send me the complete parts list for both configurations.

[peranders]

Hi all,
Hi P-A,

I hope you're all doing great.

I'm finally back to building one of the JSR-03 kits I got back a little while ago, and I'm hitting the same brick walls as before. I guess I don't find the instructions at all intuitive... I strongly believe one PDF well put together document explaining the kit construction and options is a must, especially as this is for money, not some online freebee.

Since you can download everything for a small fee I regard this as a freebee but I see what you mean. Info about the super regulator can be found at many places so if you want deep analysis you can get it. The reason why I choose not to have pdf's is that it's easier for me to write whenever I have got some spare time.