Active pre-charge/inrush control

OK, here we go...

Theory of operation:
When the switch is turned on, R1 feeds voltage to the FET gates, which will be clamped by the zener diode.
Initally, the controller (-) will be at the same voltage as controller (+).
As the FETs turn on, the controller (-) line starts to pull down, causing current flow through the capacitor.
The dV/dt of the controller voltage will be constant to match the zener current. This will cause the current flowing into the capacitors of the controller to be constant during the precharge. Once the controller is fully on, current ceases to flow through C1 and the gate voltage climbs to 12v. Turning the switch off shorts the gate, immediately turning off the FETs and slowly charging C1.

C1 needs to be non-polar and rated for the full pack voltage.

Some approximate math: Lets say the pack voltage is 62V, so the voltage across R1 is 50V. This makes the zener current 50uA. With 1uF for C1, this will allow a dV/dt of 50V/second on the output.
If the controller capacitors total 1,000uF, the current would be a constant 50mA during precharge, for about 1 second.
If the controller capacitors total 10,000uF, the current would be 500mA.
Inrush Limiter 3.jpg

If one wants to use this without a switch and just disconnect the battery, the schematic below should work.Inrush Limiter 3a (no switch).jpg
 
fechter said:
OK, here we go...

Theory of operation:
When the switch is turned on, R1 feeds voltage to the FET gates, which will be clamped by the zener diode.
Initally, the controller (-) will be at the same voltage as controller (+).
As the FETs turn on, the controller (-) line starts to pull down, causing current flow through the capacitor.
The dV/dt of the controller voltage will be constant to match the zener current. This will cause the current flowing into the capacitors of the controller to be constant during the precharge. Once the controller is fully on, current ceases to flow through C1 and the gate voltage climbs to 12v. Turning the switch off shorts the gate, immediately turning off the FETs and slowly charging C1.

C1 needs to be non-polar and rated for the full pack voltage.

Some approximate math: Lets say the pack voltage is 62V, so the voltage across R1 is 50V. This makes the zener current 50uA. With 1uF for C1, this will allow a dV/dt of 50V/second on the output.
If the controller capacitors total 1,000uF, the current would be a constant 50mA during precharge, for about 1 second.
If the conroller capacitors total 10,000uF, the current would be 500mA.
. I think you could make a few mild changes to this to turn a contacter on.
 
I think my problem is my board itself. I made it so it's like a capacitor and thus when I turn it on, a huge surge of current is dissipated between the Source and Drain when it just opens, thus frying the board. I put the bottom of the board as the source and the top of the board as the drain and since it is high power, I made it a large pad, hence a Capacitor when it's off.
 
I don't think the capacitance of the board will be a problem. If you do the math on it, it is actually quite small compared to the gate capacitance and other things in the circuit. I do think having the negative leads coming off in opposite directions would be good to help even out the current sharing, which could explain why the one on the left blew first.

I think the problem was overdissipation. The datasheet for a IRFB4310 is here: http://www.mouser.com/ds/2/200/irfb4310zgpbf-72754.pdf

This one is just as an example but others are similar.

The maximum continuous drain current is over 100A.
The maximum pulsed drain current is 560A.
This seems like it should be plenty to charge a capacitor. But if you go down the page to Fig. 8, Maximum Safe Operating area, I think is where we find the answer. Just for example, let's take our 60V pack. If the charge rate is too fast, we end up outside the safe operating area on the right side of the graph. Maximum power dissipation is 250W. At 60V and 10A, we'd blow that baby for sure even though the FET can handle 560A.

In the circuit I posted above, the current will be constant, and in a fairly extreme case is limited to 500mA, which would be a dissipation of 30W, well within the safe area.

I forgot to ask; what voltage and controller you are using?
 
fechter said:
I don't think the capacitance of the board will be a problem. If you do the math on it, it is actually quite small compared to the gate capacitance and other things in the circuit. I do think having the negative leads coming off in opposite directions would be good to help even out the current sharing, which could explain why the one on the left blew first.

I think the problem was overdissipation. The datasheet for a IRFB4310 is here: http://www.mouser.com/ds/2/200/irfb4310zgpbf-72754.pdf

This one is just as an example but others are similar.

The maximum continuous drain current is over 100A.
The maximum pulsed drain current is 560A.
This seems like it should be plenty to charge a capacitor. But if you go down the page to Fig. 8, Maximum Safe Operating area, I think is where we find the answer. Just for example, let's take our 60V pack. If the charge rate is too fast, we end up outside the safe operating area on the right side of the graph. Maximum power dissipation is 250W. At 60V and 10A, we'd blow that baby for sure even though the FET can handle 560A.

In the circuit I posted above, the current will be constant, and in a fairly extreme case is limited to 500mA, which would be a dissipation of 30W, well within the safe area.

I forgot to ask; what voltage and controller you are using?

OK maybe it's not the Capacitance of my board, but it has to be something I am doing wrong.

I am using Lyen's 18FET 4110FETs version 1 if that matters. I have a version 2, but I never got around to send it back to him to mod it for my uses with my current bike. I am not sure if the version 2 is any better at this.

I am running 74V Lipo (20s).

I am using 47K and 10K as the bottom resistor. I would definitely need a switch version as I don't want to plug and unplug each time. Judging from the 47K and the 10K I should get max of 14V on the gate if I go up to 90V. But my max is 82V normally at fully charged so it should be somewhere around 13.5V which is well within the maximum specs of the 4110 FETs.
 
mvly said:
I am using Lyen's 18FET 4110FETs version 1 if that matters. I have a version 2, but I never got around to send it back to him to mod it for my uses with my current bike. I am not sure if the version 2 is any better at this.

I am running 74V Lipo (20s).

I am using 47K and 10K as the bottom resistor.

OK, helps to know the setup. Those controllers have pretty hefty input capacitors and 74V is a lot.

I would suggest trying one of my circuit variations. If that is too much work, I'd at least suggest adding a zener and increase the capacitor/resistor values enough to get the turn on speed slower. With 10uF and a 1M top resistor, the turn on time should be slow enough to avoid destruction. 47uF might be better.
 
i'm not sure, but i think that i read that you can't use that circuit if there is a load attached to the switched side of the circuit. i planned to insert this inside the controller, and only power on the controller.
now i need to install some led light, which i want to have turned on all the time i ride the bike (controller turned on), so i don't forget to turn it on. it's a 10w load, or 0.12A for the 80v battery.
will this cicuit work flawlessly if a led light is connected in parallel to the controller feed? (to be precise: the dc/dc converter is connected - not the led directly ;).)
 
That should be no problem. It will charge the controller caps and power the lights at the same time.
The circuit simply provides a linear voltage ramp to the output regardless of the current. With a linear ramp, the current to the main capacitors will be constant, and well below what the FETs can take. An additional load will add to what the caps take, but should still be well within the safe area.

With my circuit, you could also have the controller 'key' wire permanently attached to the positive wire and use the circuit switch to turn everything on and off.
 
fechter said:
With my circuit, you could also have the controller 'key' wire permanently attached to the positive wire and use the circuit switch to turn everything on and off.
which circuit are you referring to? jeremy's first circuit also allows permanent connection, and just putting a switch in the gate supply. or am i wrong?
 
I think it would be fine with either circuit. The drain of the controller logic is very small compared to charging the caps. When the circuit is off, it cuts the power to everything on the output side.

I still worry about using Jeremy's circuit with a high voltage, big amp setup that it could pop the FETs on precharge. I'd feel better about it if I could test the turn-on time and peak current it allows during precharge.
 
I finally got around to breadboarding the circuit drawn on page 7. Absolutely beautiful linear ramp during precharge.
If I look at the gate voltage, it jumps up to the threshold (around 3v) and holds there until charging is done, then moves up to 12v. Charging time was very close to calculated value. For a capacitive load, the current stays constant until precharging is done. Knowing the voltage and value of capacitance in the controller, the charging current can be fixed by the value of the capacitor and R1. The values shown should be pretty conservative and charge in about 2 seconds.
 
I think I will need to try your circuit. I bought some 4110 and tried with the 'old' circuit with one fet only. Worked nice on the bench (no load other then the controller in idle mode).
Then I paralleled 3 of them. Worked once, then a fet burned again. No idea what goes wrong. Maybe paralleling them is the issue.
 
Yes, it's 1M ohm. The bias current on the gate is kept low so the capacitor doesn't need to be as large. The same values will work over a wide range of pack voltages (about 30v and up). This also minimizes heating on the resistor and zener diode.
The gate is clamped at 12v by the zener. The 1M resistor is enough to bring the gate up to 12v.
 
Arlo1 said:
I think you could make a few mild changes to this to turn a contacter on.

OK, here you go!

This arrangement has the contactor on the negative side so we can use N channel FETs. The FETs don't need to be super high current, but it wouldn't hurt. It would be possible to use P-channel FETs and design the circuit to switch on the positive side, but I don't see any real advantage to this.

If the contactor coil is rated for the pack voltage, it can look like the drawing. In most cases, the contactor coil is rated for a lower voltage, so either a big resistor in series with the coil or a DC-DC converter can be used between the circuit and the coil. The main DC-DC coverter for a vehicle could be driven off Q2. If using a DC-DC, make sure Q2 is rated for the current it takes. Both Q1 and Q2 should have some kind of heat sink, but they should only need it during the two seconds during precharge.

Q1 does the precharging. Once the controller gets up to full pack voltage, the dV/dt drops to zero and allows Q2 to turn on. One advantage of this arrangement over the typical resistor precharge is any additional drain from the controller circuitry or a DC-DC converter won't prevent it from reaching full voltage.

If the controller is shorted, the fuse will blow, preventing the contactor from kicking in. The fuse should be around 2A plus whatever the DC-DC or coil takes.

In practice, turning on the switch starts precharge, which will take about 2 seconds. After that, the contactor automatically engages.
Automatic Precharge for contactor2.jpg
 
fechter said:
Arlo1 said:
I think you could make a few mild changes to this to turn a contacter on.

OK, here you go!

This arrangement has the contactor on the negative side so we can use N channel FETs. The FETs don't need to be super high current, but it wouldn't hurt. It would be possible to use P-channel FETs and design the circuit to switch on the positive side, but I don't see any real advantage to this.

If the contactor coil is rated for the pack voltage, it can look like the drawing. In most cases, the contactor coil is rated for a lower voltage, so either a big resistor in series with the coil or a DC-DC converter can be used between the circuit and the coil. The main DC-DC coverter for a vehicle could be driven off Q2. If using a DC-DC, make sure Q2 is rated for the current it takes. Both Q1 and Q2 should have some kind of heat sink, but they should only need it during the two seconds during precharge.

Q1 does the precharging. Once the controller gets up to full pack voltage, the dV/dt drops to zero and allows Q2 to turn on. One advantage of this arrangement over the typical resistor precharge is any additional drain from the controller circuitry or a DC-DC converter won't prevent it from reaching full voltage.

If the controller is shorted, the fuse will blow, preventing the contactor from kicking in. The fuse should be around 2A plus whatever the DC-DC or coil takes.

In practice, turning on the switch starts precharge, which will take about 2 seconds. After that, the contactor automatically engages.
I planed to do this and post a schematic but hey its likely you would have had to help me edit it so it would work. I will make sure I build this asap! Thank you very much!
 
Arlo1 said:
I planed to do this and post a schematic but hey its likely you would have had to help me edit it so it would work. I will make sure I build this asap! Thank you very much!

Awesome!
What's the coil voltage on your contactor and your pack voltage?
 
fechter said:
Arlo1 said:
I planed to do this and post a schematic but hey its likely you would have had to help me edit it so it would work. I will make sure I build this asap! Thank you very much!

Awesome!
What's the coil voltage on your contractor and your pack voltage?
I have about a million projects. Ok so dune buggy 50v (for now) controller limited, BMX 100v, YSR 84v heading for 170v, Drift trikes 84v, electric drag car for drag race hi 150v but we might get a soliton controller and they have built in contactors/precharge or maybe they just control it lol I have to double check but we are not there yet. Electric car conversion is TBD it might be pushed up if I can find parts cheep enough. Right now I only have 2 contactors 1 is a reversing contractor but I'm running them at 12v and they seem OK. I never asked randy what they were rated at.
 
Sounds like a small dc-dc converter (wall wart) might be good for driving the coil. A big resistor will get pretty hot.
 
fechter said:
Sounds like a small dc-dc converter (wall wart) might be good for driving the coil. A big resistor will get pretty hot.
Yeh I got a couple off ebay I tried a laptop power supply but it would not work from 84v If you find them dc/dc converters are pretty cheep on eaby.
 
Standby for correction....

I see the fuse is in the wrong position. It would never blow where it is.
 
OK, updated schematic below. With the fuse in this position, if there is a short in the controller or overload, the fuse will blow and prevent the contactor from closing.

Automatic Precharge for contactor2a.jpg
 
i've built v3 of the switch, and now have some questions:

.) is 2.5mm2 copper wire enough for 40A (@75v) max?
.) is it ok to put the switch just between gate and vcc from battery (over 1m resistor)? i have a 1-way switch (on/off) at home - or do i need to buy a 2-way one? why is off connected to GND?

this is what it looks like now. the thin black GND wire acts as kind of fuse, in case something goes wrong during testing. this will be changed during installation :)

inrush.jpg
 
2.5mm2 wire should be OK, but a little small. As long as they are short, there won't be much loss.

The reason for the ground connection on the switch is to speed up the turn-off. It will work without it, but when you turn the switch off, it might take 30 seconds before it really turns off. This might be OK in most cases. It might be bad in an emergency, like stuck throttle. It would also be bad if there was a heavy load (motor running) when you turn it off.
 
Back
Top