Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Help needed with Soft Start Circuit for 12V High Current Load

Status
Not open for further replies.

Tango

New Member
I am trying to design a soft start circuit for a 12V high current load. The 12V power source is a large 240Ah LiFePO4 battery with a BMS that is cutting out because the initial load of the ARB Twin compressor is drawing a peak in-rush current of nearly 200 amps. The BMS is rated at 100 Amps constant or 150 Amps momentary.

I have made an attempt to spec components based on a simple mosfet, relay and power resistor design but it's not working. Ie I introduce the circuit into the load but it makes no differnce and the bms continues to kick in and disable the battery as soon as i turn on the ARB Twin Compressor.

I'm not sure if it's dodgy components (because I roasted them when soldering) or incorrect component specs.

I basically wanted to restrict the load current to 80 Amps until the relay is closed. So i chose 3 x 0.05 Ohm resistors. I've tried both 10 and 22 uf caps and when i test on the bench, i do get about a one second or two second delay before I hear the relay switch but when i install into the actual load (it's in my 4wd), as soon as i turn on the ARB Twin air compressor the BMS kicks in. This seems to not even be after one or two seconds but immediately as if the greater than 150 Amps is still being drawn.

I'm hoping someone can at least confirm the design is sound or not?

1614764691530.png


C1
50YXM10MEFRTA5X11 CAP ALUM 10UF 20% 50V TH
https://www.rubycon.co.jp/en/catalog/e_pdfs/aluminum/e_yxm.pdf

D1
1N4002-E3/54 DIODE GEN PURP 100V 1A DO204AL
https://www.vishay.com/docs/88503/1n4001.pdf

Q1
FQD1N80TM MOSFET N-CH 800V 1A DPAK
https://www.onsemi.com/pdf/datasheet/fqu1n80-d.pdf

R1, R2, R3
FCSL110R050FER RES 0.05 OHM 5W 4320 WIDE
https://www.ohmite.com/assets/docs/res_fcsl.pdf?r=false

R4, R5
RNC55J1003BSRE6 RES 100K OHM 1/8W .1% AXIAL
https://www.vishay.com/docs/31025/erc.pdf

U4
DG85D-7021-75-1012 12V dc Coil Non-Latching Relay SPNO PCB Mount, Single Pole
https://docs.rs-online.com/3ffe/A700000007295110.pdf
 
Last edited:
I don't see any obvious problem with the principle.

However, if it trips with the resistors in line, then the BMS must still be seeing an excess load.

I'd suggest trying with resistors alone to avoid any unwanted interactions, until you can get it run for a second or two without a trip.
 
A load current of 80 A through 0.15 ohms is a power dissipation of 960 W, or 320 W per resistor. That's a lot. Forget the BMS - what is keeping the resistors from flying off the board like jumping beans?

ak
 
Are you sure everything is wired correctly?
 
oops quoted the wrong person... orginal message deleted and correct person quoted.. I'm such a newbie! :)
 
Last edited:
I don't see any obvious problem with the principle.

However, if it trips with the resistors in line, then the BMS must still be seeing an excess load.

I'd suggest trying with resistors alone to avoid any unwanted interactions, until you can get it run for a second or two without a trip.

Ok that's a good idea.. i'll give that a try. I can only think it's ignoring the whole mosfet/relay circuit and going straight through the resistors. I have a few things to check but i wanted to make sure the design was ok.

Thanks

Edit. The problem is if I use those resistors for any longer than the brief in-rush period I imagine they will blow...
 
Last edited:
BTW There is definitely still a high load getting through (193 amps!) as per clamp meter on the load side between above circuit and the ARB Twin Compressor shown below...

1614784144108.png
 
A load current of 80 A through 0.15 ohms is a power dissipation of 960 W, or 320 W per resistor. That's a lot. Forget the BMS - what is keeping the resistors from flying off the board like jumping beans?

ak

I agree and initially i was looking at expensive components! However I was able to find a resistor (or three in series to get my required 150mΩ) that look like they would cope with the in-rush current. Here is an exerpt from the spec sheet... but maybe/hopefully I'm missing something?

**broken link removed**
 
I was able to find a resistor (or three in series to get my required 150mΩ) that look like they would cope with the in-rush current. Here is an exerpt from the spec sheet...
The inrush power rating is based on a 10 ms pulse followed by one full minute of cool-down time. That's 100x less energy than your circuit.

ak

ps. One gold star for posting the datasheet links without being asked.
 
The inrush power rating is based on a 10 ms pulse followed by one full minute of cool-down time. That's 100x less energy than your circuit.

ak

ps. One gold star for posting the datasheet links without being asked.


Ah ok I see thanks..
1614785866079.png


Ok, i'm thinking that what i was hoping the spec of this little resistor was too good to be true. I'm assuming what you're saying is that it is not capable of handling what i am expecting it to (by a factor of 100). Would you mind explaining exactly why? Ie how does the spec not measure up to my requirement?
 
I guess the compressor is designed to operate from a lead-acid car or truck battery capable of supporting that current when cranking the engine. You don't say if you are starting against pressure or not, that might make a big difference.
As this thing has two motors what about starting them in sequence, that would at least halve the inrush to start with.
Another possibility would be to use one or more of those crazy big capacitors car audio fans use to provide or assist in providing the inrush.
I fear a successful inrush current limiter may just stop the motor starting altogether :(
 
I guess the compressor is designed to operate from a lead-acid car or truck battery capable of supporting that current when cranking the engine. You don't say if you are starting against pressure or not, that might make a big difference.
As this thing has two motors what about starting them in sequence, that would at least halve the inrush to start with.
Another possibility would be to use one or more of those crazy big capacitors car audio fans use to provide or assist in providing the inrush.
I fear a successful inrush current limiter may just stop the motor starting altogether :(

The ARB Twin Compressor is basically two compressors plumbed together (shared air tank) with two separate fused motors/pumps. If I remove one of the fuses then the remaining compressor fires up ok and pulls 121 Amps which is under the max burst load of 150 Amps which the baterry's BMS can handle.

The battery is a 240 Ah LiFePO4 and if I can overcome this in-rush current, should be able to cope (I hope) with the load for a few hours not that i will be using it anywhere for that long. It will otherwise only be running for seconds at a time occasionally over a number of hours if that. It feeds a small air cylinder so only kicks in when cylinder is low and then pumps until it get to 150 psi.

While I will usually be using it while the car's ignition is on to inflate my tyres, I have specifically not plumbed it into my ignition as i want to use it while the car is not running for other things like an air gun and aerating my bait tank (I spend a lot of time in my 4wd on the beach fishing!)

When I test, I remove all air pressure from the system to ensure the compressor turns on when i flick the switch so there is next to no load (no air back pressure) at this time.

Staggering the start might be an option. The compressors are rated at 40 amps each so i might just get away with it (one observed single load of 121 amps plus hopefully less than 30 amps for a single compressor under normal load when the second kicks in).

I'd rather get this working if I can?

Until I found what i thought were the right resistors, I was looking at huge power resistors which are both expensive and big!
 
The inrush power rating is based on a 10 ms pulse followed by one full minute of cool-down time. That's 100x less energy than your circuit.

ak

ps. One gold star for posting the datasheet links without being asked.

Ah ok got it :) 100x comes from the 1 sec cap charge time so yes, I am expecting way too much from those little resistors. I did wonder when you compare with size of power resistors rated much higher than those and they still weren't suitable! I just got excited when i saw their in-rush rating which i hadn't seen on all the other resistors. Thanks for the education wrt the assocaited time spec. :)

Anyway, back to trying to find a resistor or resistors that won't break the bank or take up half my cargo. Otherwise am going to have to find another solution...
 
how does the spec not measure up to my requirement?
From the point of view of the datasheet, 1 second is the same as continuous, and the continuous power rating is 5 W. The issue is heat. Think of the heat of an electric hotplate. Now compress that heat down to the size of a fingernail. That's hot.

The resistor has both a power rating and a max operating temperature rating. The smaller the part, the faster it heats up. This part can handle 5 W continuously, or 600 W ***very*** briefly followed by a looong cool-down period. IOW, at 600 W it reaches its max operating temperature in 1/100th second.

ak
 
Anyway, back to trying to find a resistor or resistors that won't break the bank or take up half my cargo. Otherwise am going to have to find another solution...
To put things in proportion:
If you wanted one rated to take the full load continuously, you would need one like this - 1KW rating:

1KW_Res_800.jpg


For a very short switch-on surge limiter, you may just get away with using a good number of this type in a series-parallel array; they are normally used for current sensing, but are quite heat resistant. A fan could help as well.

ShuntRes_800.jpg


[There is a three phase IGBT power module under the board, for anyone trying to figure out the connections].
IGBT_Module.jpg
 
You can get some pretty hefty resistors on ebay for not a lot of money. For example 0.01Ω 100W resistors for ~US$2. Two is series will (probably) do what you need. They will get hot but only for a short time.

Mike.
Edit, to be extra safe you could series/parallel them to up the wattage.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top