24V to 500-1000V, 20W floating DC-DC converter

Jul 21, 2020 104 Replies

This circuit is developed from conventional rail splitter balancing circuits by me in the refurb of a set of Owon oscilloscope batteries on my bench - these were not balanced as supplied, where failing cells produced early loss of capacity.

The natural deadband is a 'defect' of any linear voltage follower, and is used to advantage here.

The circuit, as published and maintained on VE3UTE service, is in the public domain. Note that the .plt file will not plot in LTspice installations that do not support unicode text. The OP amp used there has an Iq ~50uA, similar to TSX631 parts actually used in my battery refurbs.

You will need to breadboard and test any circuit that you use in this or similar applications. Familiarity with the effect of gain, offsets, tolerances and noise should be examined. Ground plane layout is advisable.

All measurements attempted on high-z nodes require care to take loading effects into account - best to avoid them where possible - there are plenty of low-z nodes or ~0V branches that are less succeptible to measurement error. It's slow DC, but scope it anyways to avoid surprises.

Note that some of the ICs listeded are reverse polarity protected up to ~18V by their spec sheet. In applications where cells may be removed or inserted in a string, or where fault interruptors may function, the single-fault abnormal conditions must be examined carefully.

A battery is always live and ready to screw you around. R5 is there for a reason - silicon shorts on a battery are no fun.

RL

e:

ote:

group. A search of TL431 came up with this thread recently. So, i am high- jacking this for a quick question:

ondary side only

ay.

page 7

of the pass transistor (TO-220 PNP)?

owever, the constantly on buffers/drivers drain out the batteries. I am in fact trying to figure out how to window out the middle part. When the batte ries are fairly balanced, i want to turn off the buffers/drivers.

cuits are necessary.

to 36V.

cuit your own design, or pull from somewhere? Just wondering if there is pa ttern or copyright issue for production.

OK, thanks. I'll quote you for credit. My version is almost identical, ex cept for the opamp feedback. I just use open loop for now.

You don't have to warn me about batteries. I already burn my fingers and B MS boards many time doing prototype.

I typed "low power op amp" into google, and got a link to the ST micro site.

formatting link
clicked on a link in the grid it took me here

formatting link

Seems they could benefit from some SEO.

Jasen.

Open loop won't give you a uA drain window, around the balance condition..

RL

rote:

te:

gle group. A search of TL431 came up with this thread recently. So, i am hi gh-jacking this for a quick question:

secondary side only

l way.

of page 7

be of the pass transistor (TO-220 PNP)?

it

ght

. However, the constantly on buffers/drivers drain out the batteries. I am in fact trying to figure out how to window out the middle part. When the ba tteries are fairly balanced, i want to turn off the buffers/drivers.

circuits are necessary.

up to 36V.

circuit your own design, or pull from somewhere? Just wondering if there is pattern or copyright issue for production.

xcept for the opamp feedback. I just use open loop for now.

BMS boards many time doing prototype.

Yes, this and the heavy duty lm358 is killing my batteries. I owe you the "legg resistor".

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required