Active pre-charge/inrush control
- Thread starter boo
- Start date
izeman
1 GW
thanks to ME? i'm the one asking questions. you're the ones helping meteklektik said:
i thank YOU!Well - this circuit's been around a while but it's your posting and questions that got these changes on the table so fechter could tune it up...
Any good engineer can tell you that good feedback is invaluable - it brings shortfalls and oversights into focus.
Like it or not, you're 'in the loop'....
Any good engineer can tell you that good feedback is invaluable - it brings shortfalls and oversights into focus.
Like it or not, you're 'in the loop'....
Beta testers are always going to have some issues to sort out. Glad to get the feedback.
It's like being a crash test dummy. :wink:
It's like being a crash test dummy. :wink:
nicobie
100 MW
Sure wish somebody would do a run of these boards with instructions and a parts list like was done with Fechter's meanwell clone current limiter.
izeman
1 GW
what do you think about that design? it's a fechter 3a version. as i now found an on/on switch i can now switch GATE to GND when off.
let me know if i've done something wrong.
my idea was to make it as compact as possible, while being able to solder DRAIN and SOURCE directly to large copper traces. the whole design is single sided as there was no pros for making it double sided.
the only issue i see is cooling, as the fets don't have direct contact to the heat sink on the back side.
let me know if i've done something wrong.
my idea was to make it as compact as possible, while being able to solder DRAIN and SOURCE directly to large copper traces. the whole design is single sided as there was no pros for making it double sided.
the only issue i see is cooling, as the fets don't have direct contact to the heat sink on the back side.
That looks good.
If you want more heat transfer to the back, use a bunch of little through holes (vias) and fill them with solder.
You could also add a piece of aluminum across the tabs on the top side to add surface area.
If you want more heat transfer to the back, use a bunch of little through holes (vias) and fill them with solder.
You could also add a piece of aluminum across the tabs on the top side to add surface area.
izeman
1 GW
thank fechter. so i will start ordering boards.
is anyone interested in a board, or an assembled pre-charge switch? if i order from oshpark again i will have to order 3 boards. so i have a spare.
is anyone interested in a board, or an assembled pre-charge switch? if i order from oshpark again i will have to order 3 boards. so i have a spare.
nicobie
100 MW
I'd like to have one. Already made would be nice, but I can put one together if I have the parts (or list) and simple instructions. I already have the on/off switch.
Nick
Nick
izeman
1 GW
do you have an on/on switch or an on/off? you need the first one for this circuit to work well. otherwise voltage decline will be very long.nicobie said:
I was thinking you might jiggle things a bit to pick up a little extra heatsink area at the cost of moving the switch Gnd pad away from the board edge...
Not a big deal, but with only two FETs a little extra heatsink won't hurt...
Maybe something like:
Just a thought...
Not a big deal, but with only two FETs a little extra heatsink won't hurt...
Maybe something like:
Just a thought...
Another thought is to make the layout for 3 or 4 FETs but you can install fewer if you don't need that much current.
izeman
1 GW
i moved all parts (except the zener) to smd (only 2 resistors and 1 cap) and added holes under the FETs. this is the new design.
i hope i haven't forgotten anything, or cut a trace that was there before
i hope i haven't forgotten anything, or cut a trace that was there before
izeman
1 GW
sorry teklektik and fechter. i saw your posts after i posted my new one. i will consider your thoughts and update my design. will got to bed now.
btw: the whole pcb will be bolted to the massive alloy frame box (with a little spacer. so there will be more than enough mass to spread the heat - if there is any. i don't think that it will get hot at all @50a peak.
btw: the whole pcb will be bolted to the massive alloy frame box (with a little spacer. so there will be more than enough mass to spread the heat - if there is any. i don't think that it will get hot at all @50a peak.
No problem at all - looks like you already freed the heatsink space I was looking for. Cool.
(I wasn't enamoured with my parts layout - just focused on clearing the space)
Ya - extra FET real estate for 3 or 4 would certainly enhance the desirability.
Sort of depends if selling a few is really that important or just a means to defray a one-off cost....
Check the no-heatsink power limits below.
(I wasn't enamoured with my parts layout - just focused on clearing the space)
Ya - extra FET real estate for 3 or 4 would certainly enhance the desirability.
Sort of depends if selling a few is really that important or just a means to defray a one-off cost....
Check the no-heatsink power limits below.
izeman
1 GW
i changed some things in the design but sticked to two FETs. they are very well cooled, and i don't think i will need more.
i hope the 0.35u thick copper will be able to handle the current. so i added copper to the top and bottom layer, and both are connected a) with thru holes and with the components. i may add some thick copper wires as well?!
i ordered three of them for $12 at oshpark.com. and some SMD parts at the local electronics supplier.
i will post some pictures once i have the boards.
i hope the 0.35u thick copper will be able to handle the current. so i added copper to the top and bottom layer, and both are connected a) with thru holes and with the components. i may add some thick copper wires as well?!
i ordered three of them for $12 at oshpark.com. and some SMD parts at the local electronics supplier.
i will post some pictures once i have the boards.
izeman
1 GW
pcbs have been shipped today. i hope to get them within a week or two.
nicobie
100 MW
izeman said:
Sorry I missed your post.
What is an on/on switch?
izeman
1 GW
"normal" switches have an ON and an OFF position. so if you switch it to ON the circuit closes.nicobie said:
with this circuit to work best you will need a switch that toggles between two position. so it has 3 pins. on for the switches signal and two inputs.
A SPDT switch.
izeman
1 GW
in this case the german name is easierfechter said:
ein/einschalter. i don't even think that there is a abbreviation for it (which btw is totally atypical for germans)
nicobie
100 MW
fechter said:
Thanks! It was the on/on that fooled me.
I'd of called it on/off/on or SPDT.
Many SPDT only have 2 on positions, and no off.
izeman
1 GW
damn. i wanted to write something like "on/off/on" is a 3 position switch and i'm talking about a 2 position on. but then decided against it because i thought it would be a bit smartass, but now that you said it ...amberwolf said:
You really want the two position version. Functionally, it will be on/off.
Similar threads
