LM51561H: continue debugging

Marv In

Part Number: LM51561H
Other Parts Discussed in Thread: LM51561, LM5156

Tool/software:

Hi there,
this is an answer to Moritz in LM51561H: Output Voltage unstable - Power management forum - Power management - TI E2E support forums . since the topic was locked.

Hi there,

thanks for your continued help. I was on holidays, now I am back.

I tried playing around a bit with the values around the optocoupler, didn't help. I checked the Power Stage Designer, but can't find anything about optocouplers.
We are using the LM51561HPWRR, isn't tat one without hiccup already? I'll go through the schemetic once again today and let you know if I find anything.

kind regards

Marvin

11 months ago

0 Moritz 11 months ago

TI__Mastermind 18705 points

Hi Marvin,

LM51561 ist the hiccup version.

Did you try to implement the PNP at softstart already?

Best regards

Moritz

0 Marv In 11 months ago in reply to Moritz

Prodigy 40 points

not yet, I first have to order a PNP, or find one in my old boxes.
and the problem with using another version of the IC is, that we already bought 1000 of the LM51561. so i would prefer to get this one to work.

0 Moritz 11 months ago in reply to Marv In

TI__Mastermind 18705 points

Hi Marvin,

Ok, Then please try the LM5156 IC (no hiccup) and the softstart PNP as mentioned in previous thred first.

Best regards

Moritz

0 Marv In 11 months ago in reply to Moritz

Prodigy 40 points

the resistor and capacitor in comp are flipped on the evalboard. does it make a difference? and c25 is C_HF?

0 Marv In 11 months ago in reply to Moritz

Prodigy 40 points

i added an old BC557 and C_HF. I can now draw 8mA of current before it collapses. which is an improvement.

0 Moritz 11 months ago in reply to Marv In

TI__Mastermind 18705 points

Hi Marvin,

Marv In said:
the resistor and capacitor in comp are flipped on the evalboard. does it make a difference?

No, this does not matter.

It is good that you can see an improvement. Can you please share some scopeshots again?

Best regards

Moritz

0 Marv In 11 months ago in reply to Moritz

Prodigy 40 points

Its a really small improvement but not really.
I tried again:
-C_SS increase, 2nd increase
-SS_PNP added
-R18 increased
-C_HF added
-U2, U3, D4 exchange with original parts from EVM
-r14 lowered

no change in the behavior
we will try non hiccup after the holidays, but we would prefer to use the ones we already have a 1000 reel of.

0 Moritz 11 months ago in reply to Marv In

TI__Mastermind 18705 points

Hi Marvin,

Yes it would be good to try out the non-hiccup version.

Can you please send waveforms of the behavior you see now?

Best regards

Moritz