Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

400v power supply design

Status
Not open for further replies.
D

dhanvi

Newbie level 1
Joined
Oct 24, 2019
Messages
1
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
21
Hi all

i need to design power supply which is having input as 400Vdc and input power of 7.4Kwatt. Now it will have 50V dc output voltage. how to calculate inrush current limiter resistor value and capacitor bank before conversion. Please help me in suggesting way forward.

Thanks & Regards
C.Dhanvi
 

7.4kwatt on 400V is 18A. Assuming AC fed PFC the caps need to ride through the input AC zero crossing regions without drooping too much. How much droop is ok? Depends on your application but maybe 10V?.

L=CV' so for example 10,000uF would droop about 9V in about 5mS (18/10000u*5m). So 10,000u is a rough ballpark starting point I think. Search "PFC bus capacitance".

There may be other regulatory requirements like riding through missing cycles.
 

for 400VDC in - you need little bus capacitance, perhaps only 20uF even at 7.4kW, therefore just about any inrush limiter will do, say 10 ohm 10 watt, giving 40A peak for a very short time ( < 20uS ) and <1mS to charge the bus, then close a relay on the 10E resistor.

We have developed many 6.5kW systems for 400V - 600V to "48V" 65VDC max at 120 amps DC continuous

kind regards, EP.
 

Hi,

But mind the powerdissipation at 7.4kW load.
It will be about 4000W @ 10 Ohms.(edit: corrected 40W --> 4000W)

Klaus
 

0.5 C V^2 / time of pulse = 16 watts for ~ 100mS, a 10 watt resistor will handle the 40A peak decaying to 0 over 100mS without issue ... for 20uF, 10E, 400vDC
 

Hi,

True, but this ignores the load current.
At 7.4kW output power you maybe have 20A input current ... continously.
Oh .. and even worse: My previous mind calculation was wrong. It is 4000W!

P = I x I x R = 20A x 20A x 10R = 4 000W

Added: this means a voltage drop of 200V, thus my calculation stiĺl does not fit to real behaviour. Most probably the circuit won't be able to output 7.4kW with a 10R resistor installed.

Klaus
 

yes but at start up the down converter isn't running yet - hence just charge impulse ....
 

Hi,

I agree ... just for the startup ...
but some times later it will start to operate...

Klaus
 

Hi,

This is what I missed to read. Sorry.

Klaus
 

When I hear 400V capacitor bank I think PFC. But maybe that's not what they were asking.



When it comes to inrush limiting a PTC is ideal if there is a relay to switch it out. If the relay fails it self protects (removing the need for a protection fuse). And it's specified for joules (good luck finding joule ratings for resistors).
https://www.vishay.com/docs/29165/ptcel.pdf

PTCs are also ideal for fast capacitor bank discharge.
 

also 7.6 kW off single phase mains ( 400VDC ) is 33 amps AC @ 230Vac

the question of how much C to follow a 7.6kW PFC stage is for another thread I think ...

PTC's are not favoured today much as they go high R as they get hot - thus you cannot get a reliable quick re-start of a PSU - or discharge of cap bank, once they get hot - modern practice is the good ol resistor with relay .... we have made const current discharge modules with mosfets for cap banks, up to 20A 400V - with heatsink ...
 

Funny because that's exactly what I think the benefit of a PTC is - it self protects in the case of quick-restarts or multiple fast discharges. Whereas even a well specified resistor could be driven to failure under certain use case scenarios.

And I can 'cash-in' the knowledge of its self protection as both faster charge and discharge under normal circumstances (our PFC's detect start-up, slow charge is only a user nuisance).
 

Our users tell us it is a nuisance - on other peoples gear, 2 or three starts will send a ptc to high res taking several 10's of seconds to cool down if you're lucky. For faults PTC's ( see polyfuse ) are very useful - limiting fault currents and sitting there hot.

However for psu's that must start within one sec - different story. Then there are ptc failure mechanisms - they tend to go open.

We put thermal fuses along side our start up R's - so that under true fault - the R's get hot and melt the fuses - no mess, no bang, no overheat.
 

Ok so 2 or 3 one second starts? Size the PTCs for that (parallel for redundancy if you want). Then you're done. Abnormal cases are handled and there is no fuse to blow.

With resistors the 2 or 3 number is irrelevant. You also need to handle abnormal cases without blowing the fuse. How many is that? This forces you to substantially de-rate the resistors and you still need a 400V fuse.


Our last design cycle I had us replace fuse+R with PTCs and found it to be a nice win in terms of size and performance (probably similar cost).


Discharge is even better. If you're relying on an external part to thermally protect your discharge element (R or fet) you have to limit watts to a rate the thermal protection can track. This can be the limiting factor.

PTC's can be sized to discharge as fast as you want ((like a couple seconds for our PC bank) under normal circumstances while still being protected in abnormal conditions.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top