JSR03 going a bit nuts on me

[raq]

P-A,
I hope you're doing great.
I haven't been using my JSR03 for some months now, but as I'm going back to my DDDAC1543 project, I put it back in action but get some weird behavior.

I initially used it with C13,14 and R15.16 in place, and I am pretty sure I wired my load to X6. Important to note - the solderside jumpers were not jumped/soldered.

Funny/bad thing - I do not remember the exact reasons I did this, but it worked just great. With my values - which you might remember - I was getting a rock solid 7.1 or 7.2V out.

Yesterday, in an impulse to clean up (how many times the impulse to "clean up" totally messed something up for yah, huh?...) I took out both caps and resistors, and soldered the jumpers together.

Now I'm getting a different voltage out every time I fire it. I tried to tune it to work by varying the input voltage. But it would definitely not land on the expected voltage out.

If you're wondering, both H1 and DZ1 (where I have a blue LED) are lit in operation.

Shall I just put the Rs and Cs back and take the jumpers out? I would still need to understand what I am doing.

In a whim, I think I'll swap the opamp for a known good tonight.

Thank you very much for your help, as always, P-A.
Best,
Radu.

[peranders]

Just wondering, but why remove R15, R15 and C13, C14? The will be disconnected if you'll short them??

It sounds that something is floating.

Voltage at the inputs of the opamp?

[raq]

P-A,

I see your point, but don't forget I've been using this with jumpers not in place. Therefore, whichever way I was using the circuit, the Rs and Cs in discussion needed to be there.

Can you expand on the current sense outputs? How am I supposed to use them? For some reason I decided to use them when I put this together, but I cannot figure out how.

Thank you for your help. I will also be able to measure voltages on the opamp later on tonight, my time (US CET).

Best,
Radu.

[peranders]

The Sense inputs are for removing wire resistance, especially important if you have long wires and/or want to have low output impedance at the load. The 220 ohms resistors are just a protection so those inputs not will be floating (not good) if you forget to connect them.

If you'll check Walt Jung's articles you can see how you wire up in the best way.

[raq]

P-A,

Here are the voltages I read early this morning (so early that it was more like 'your time' than 'my time').

My Vref is 2.5V, which I see at both ends of R7 (I used 649ohm for that, as my R9 = 1k, R10 = 1.87k).  For Vref I use an LT1034/2.5

Opamp IN- is at 3.6V, while IN+ is at 1.78V. I got 6.72V at OUT. I am feeding it from the output (using R11) with 10.72V.

Needless to say, that is also what I'm getting out of the regulator.

My DZ1 is a blue LED with 3.76V Vdrop.

For reference, R3 = 3.48k, I am using R10 and C9, and my opamp is a MAX410.

I am going back to my homework of refreshing my memory with Walt's articles. It's been a while. In the meantime, please give me your input based on these numbers.

Thank you,
Radu.

[peranders]

It seems that your opamp is broken. You should have 2.5 V at your +IN. The transistor connected at the opamp output could also be broken. Do you have 0.65V across base and emitter?

What did you do exactly?

[raq]

Well, P-A, good question.

Firing this up was part of a major assembly of my DAC, so if messing this up sounds crappy, consider it in the context of much more work, the majority of which went absolutely flawless (this is therapeutic for me, so I'd feel better, just go with it. ok? :))).

In any case, in a nutsheaven, having some CRCRC filters in my unregulated side and registering some larger than expected voltage drop over the series Rs (fairly high current draw...), my entire DAC didn't fire up at all, initially (I had something like 2.6V out of the superreg., for 5 something in).

Therefore, I had to take all of the series Rs out, so I would have enough Vunreg for the superregulator to operate. Vunreg became a little under 12V. Vout what I said before (little over 10V).

If at this point  the opamp died... I don't know. It also is possible it has been damaged by weeks of carrying the little kit around with me (ES discharge or something).

I will replace the opamp tonight and let you know of my findings. You assumption that the opamp is bad coincides with my sense out of all this.

Thank you for your help.

Best,
Radu.

[raq]

P-A,
What should IN- voltage be for 7.2V out? Per my calcs, exactly the same as IN+ voltage... is that correct?

(I have a 1.87k/1k voltage divider of 7.2V => 2.508V @ In-, which is exactly what I see over Vreg, my 2.5V reference).

Thanks!
aR

[peranders]

If the opamp works and you have feedback you should have the same voltage at both inputs of the opamp, only some mV difference.

If you have 2.5 V reference you should have 7.2/2.5 gain => 2.88 gives you 1.88 in quota in your feedback network which you seem to have. Remember to have at least 11 volts in.

[raq]

P-A,

The situation is back under control. The opamp was busted, I guess. It's true that after replacing it, the SR went nuts for a good 10-15 seconds of regaining stability (auto-oscillation?) , during which my blue LED (DZ1) went on and off, off and on, flickered, blinked, etc.

And then - perfect, solid operation. I ran it for hours and hours yesterday on and on - no problem whatsoever. rock solid. I am not sure what happened exactly. Maybe a defective opamp drove this into a non-stability zone, possibly oscillating. The new good opamp canceled those effects, but it took some time to stabilize?... I'm not sure.

It's all good now. Running like a charm.

If you have any good insights into what happened, please let me know. I'd be very interested.

Thank you,
Radu.

[peranders]

Do you have a controlled start-up behavior now?

[raq]

It exhibits the most stable start-up behavior now.
aR

[raq]

P-A,

I have to take that back...

This thing is flipping on me every time I turn around. I thought the CCS current might be too small for my output demand (390mA), so I jumped it up to around 25mA (R14 = 47.5ohm).

It didn't make a change. It would usually be solid when not attached to load, but when I put it at work,regulation ceases and it puts out around 12V (I'm putting around 15V in).

I do not know what else to try.

Please let me know what you think.

Thank you,
Radu.

[raq]

...thinking that the 15V feed might have damaged the op amp when regulation was not happening (although I never measured over 12V @ OUT, where the op amp feeds), I swapped my initial choice of MAX410 for LM49710, which can withstand up to +-17V VCC... (MAX410 maxes out at 12V VCC).

Now the regulator seems to... regulate, but I am getting 6.3V out, which does not whack with the mathematics of this, as far as I understand it (i.e.: I did not change any passives with the change of op amp; the passives require everything to concur on getting 7.2V out).

Can you make sense of this?

Thank you,
Radu.

[peranders]

The voltage at the opamp pins? Can you measure those?