Active pre-charge/inrush control
- Thread starter boo
- Start date
hey guys i finally finished reading all 21 pages!
i was working on a pre-charge/anti-spark circuit for some time and it wasn't exactly perfect so i decided to look in the place i should have looked at in the first place, ES forum!
my setup is a 12s (50v) lipo that would be running at a MAX of 50 amps. however i also need to power a separate receiver with a 5v rail.
The circuit i was working on my own was using a separate resistor to first charge the caps and then the mosfet would turn on after 3s. i did the time delay for the mosfet using a 555 timer delay circuit. i tried using capacitors but they weren't working for some reason. this is the circuit.
![IMG_6133[1].JPG](https://endless-sphere.com/sphere/data/attachments/105/105452-912d2e4b3615c204d3352f18fc5809d7.jpg "IMG_6133[1].JPG")
the timer pin is pin3 connect to "source" since i need to connect the 2 negative rails for the gate to open (using a different positive rail for the gate). i could have turned on the GATE by adding another relay and put pin3 to the relay.
here is that circuit:
![IMG_6135[1].JPG](https://endless-sphere.com/sphere/data/attachments/105/105453-54b64a6b129791dae14c10d825ef1263.jpg "IMG_6135[1].JPG")
my circuit some times worked and other times it acted funny (gate was open even when no positive voltage was applied).
however after reading some of these posts it looks like i could have just blown the fet.
i want to use one of the circuits that you guys approve as i know very little about this stuff. Fechter posted this schematic a while back but im not sure if anybody tried it and it worked or if it was suppose to work in theory. here is the schematic:
![IMG_6136[1].JPG](https://endless-sphere.com/sphere/data/attachments/105/105454-2ff7450612c63342efc0b7f323599a7e.jpg "IMG_6136[1].JPG")
i have 2 issues:
1)why do i need a 12v zener if i am using a 9v battery? the voltage wont go past 9v. (in my case 5v)
2)i don't have a on-on switch so can i just put a 1M resistor between gate and source permanently? i think izeman did this in his switch and it worked fine:
![IMG_6128[1].PNG](https://endless-sphere.com/sphere/data/attachments/105/105455-074a9628ab931bcbc2c31af560c99381.jpg "IMG_6128[1].PNG")
if i didnt explain it good, sorry, let me know and ill do my best. thanks for the help.
i was working on a pre-charge/anti-spark circuit for some time and it wasn't exactly perfect so i decided to look in the place i should have looked at in the first place, ES forum!
my setup is a 12s (50v) lipo that would be running at a MAX of 50 amps. however i also need to power a separate receiver with a 5v rail.
The circuit i was working on my own was using a separate resistor to first charge the caps and then the mosfet would turn on after 3s. i did the time delay for the mosfet using a 555 timer delay circuit. i tried using capacitors but they weren't working for some reason. this is the circuit.
the timer pin is pin3 connect to "source" since i need to connect the 2 negative rails for the gate to open (using a different positive rail for the gate). i could have turned on the GATE by adding another relay and put pin3 to the relay.
here is that circuit:
my circuit some times worked and other times it acted funny (gate was open even when no positive voltage was applied).
however after reading some of these posts it looks like i could have just blown the fet.
i want to use one of the circuits that you guys approve as i know very little about this stuff. Fechter posted this schematic a while back but im not sure if anybody tried it and it worked or if it was suppose to work in theory. here is the schematic:
i have 2 issues:
1)why do i need a 12v zener if i am using a 9v battery? the voltage wont go past 9v. (in my case 5v)
2)i don't have a on-on switch so can i just put a 1M resistor between gate and source permanently? i think izeman did this in his switch and it worked fine:
if i didnt explain it good, sorry, let me know and ill do my best. thanks for the help.
BShady said:i have 2 issues:
1)why do i need a 12v zener if i am using a 9v battery? the voltage wont go past 9v. (in my case 5v)
2)i don't have a on-on switch so can i just put a 1M resistor between gate and source permanently? i think izeman did this in his switch and it worked fine:
if i didnt explain it good, sorry, let me know and ill do my best. thanks for the help.
The 12v zener is for protection. Cheap insurance. With the "Methods' version" you probably don't need it, but you do with the other version. The second version works without a switch.
ok i worked on this switch a lot over the weekend and ordered some 12v zeners for protection. the cap on the gate also slowed the gate turn on time for 1-2s like you said.
in the mean time i was testing out the mosfets with just using a voltage divider from the battery, to power the gate of the mosfet (which its drain was connected to a controller and the source to the battery). i used a 11k and 5.1k resistor for the divider and it gave me a output of 13.5v. i saw that as i gave juice to the motor (very littke) the gate voltage fluctuated from 14v-11v.
does the gate current change as i increase the current to the controller/motor?
i also need to power a 5v receiver. Until now i was using a separate 5v battery, but is there a way i can just run it off the main battery?
in the mean time i was testing out the mosfets with just using a voltage divider from the battery, to power the gate of the mosfet (which its drain was connected to a controller and the source to the battery). i used a 11k and 5.1k resistor for the divider and it gave me a output of 13.5v. i saw that as i gave juice to the motor (very littke) the gate voltage fluctuated from 14v-11v.
does the gate current change as i increase the current to the controller/motor?
i also need to power a 5v receiver. Until now i was using a separate 5v battery, but is there a way i can just run it off the main battery?
izeman
1 GW
imho it should NOT.BShady said:
yes. search for ac/dc converters. most work at lower dc input voltage as well.
izeman
1 GW
no. i meant ac/dc. what ppl call wallwarts. those are cheap and the only difference to a "regular" dc/dc converter is the rectifier at the 110/200v side that converts ac to dc. then dc is converter from 110/220 to 5v. and a lot of those work at voltages down to 40-60v as well. you just have to try them.BShady said:
izeman
1 GW
why? because most of us have several of them at home. ALL usb chargers are 5v. they can deliver up to 2a. so why buy a new converter if you have some spares at home already? also it's difficult to find 60v+ dc/dc converters. so if you have a 80v pack it's more appropriate to use an ac/dc. if you have a 48/60v- pack you may find cheap dc/dc converters.
for FETs a seperate converter would be overkill. they need mA to switch. so a little resistor is all you need to bring the battery's pack voltage down to 15v.
for FETs a seperate converter would be overkill. they need mA to switch. so a little resistor is all you need to bring the battery's pack voltage down to 15v.
Cheap dc-dc converters are available up to 60v (or more like 53v depending on output voltage) input. Search eBay for "LM2596HV". Beyond that the wall wart types are generally better.
it could be im doing something wrong but i tried the wall-wart charge and i didnt get any voltage on the output? :?
this is the wallwart i used: http://www.ebay.com/itm/5-Micro-USB...509549?hash=item5afa7c57ad:g:w-sAAOSwMr1XPKuo
does that wallwart work?
this is the buck converter i was looking into:
http://www.ebay.com/itm/8-50V-12V-2...507668?hash=item20f2dcc354:g:LN4AAOSwnDxUlNlE
this is the wallwart i used: http://www.ebay.com/itm/5-Micro-USB...509549?hash=item5afa7c57ad:g:w-sAAOSwMr1XPKuo
does that wallwart work?
this is the buck converter i was looking into:
http://www.ebay.com/itm/8-50V-12V-2...507668?hash=item20f2dcc354:g:LN4AAOSwnDxUlNlE
izeman
1 GW
No way to say if a ac/dc works by looking at it . You need to try every individual one and see if it works. If this is the only one you have, i think the dc/dc you linked should be a good choice if you don't need more than 3A. And now back to topic...
. You need to try every individual one and see if it works. If this is the only one you have, i think the dc/dc you linked should be a good choice if you don't need more than 3A. And now back to topic...back on topic:
the 1M between gate and positive & the 1M between the gate and negative was giving me a gate voltage of 23v (which is still in spec). i dont have the 12v zener diodes yet so to keep the gate voltage at a lower level i replaced the 2, 1M resistors with a voltage divider that gave me 15v (11k and 5k).
issues arouse: 1)the gate voltage goes straight to 15v no delay from the caps. i tried it with a 1uf cap same results.
2) the gate voltage even when the switch is off is 30mV? is that normal?
this is the schematic i am using: im using the version izeman used 3b
It wasnt posting normally sorry
the 1M between gate and positive & the 1M between the gate and negative was giving me a gate voltage of 23v (which is still in spec). i dont have the 12v zener diodes yet so to keep the gate voltage at a lower level i replaced the 2, 1M resistors with a voltage divider that gave me 15v (11k and 5k).
issues arouse: 1)the gate voltage goes straight to 15v no delay from the caps. i tried it with a 1uf cap same results.
2) the gate voltage even when the switch is off is 30mV? is that normal?
this is the schematic i am using: im using the version izeman used 3b
It wasnt posting normally sorry
izeman
1 GW
your post is missing the picture
BShady said:
The divider should be OK as long as you make sure the gate doesn't go over 15v or so.
30mV is not a problem, but it should be zero. Things don't start to happen until the gate reaches the threshold voltage, which is generally around 2v.
The delay is not really long so it might just be the way your meter responds. I used an oscilloscope to look at the gate voltage. It should go up to the threshold and hold there for about a second, then go up to 12-15v. It's pretty fast.
i got a 12v zener diode from a old board and put it in ciruit.
it still sparks when i first turn it on.
whats also wierd is that even after the spark, the controller turns on but when i run the motors the voltage on the gate goes up and down. therefore the motor only spins for a second and then it stops. Also the mosfet gets really hot. Each time i pull the trigger this happens.
i dont know if it matters but i am using this mosfet just to test the circuit as i dont want to blow the good mosfets.
http://pdf1.alldatasheet.com/datasheet-pdf/view/213420/TOSHIBA/2SK3563_06.html
the motor is only running at 2 amps.
it still sparks when i first turn it on.
whats also wierd is that even after the spark, the controller turns on but when i run the motors the voltage on the gate goes up and down. therefore the motor only spins for a second and then it stops. Also the mosfet gets really hot. Each time i pull the trigger this happens.
i dont know if it matters but i am using this mosfet just to test the circuit as i dont want to blow the good mosfets.
http://pdf1.alldatasheet.com/datasheet-pdf/view/213420/TOSHIBA/2SK3563_06.html
the motor is only running at 2 amps.
Maybe try posting a picture of your circuit. It should not be sparking when you hook it up.
The FET you are testing with is not really suitable for running the controller as it has an on resistance of over 1 ohm. Since the resistance is so high, the controller voltage will vary widely with load, which will cause the gate voltage to change. With proper FETs, the on resistance should be a few milliohms so the controller voltage won't sag with load, thereby causing the gate voltage to change. Still, even with the 2SK3563, you should be able to precharge the controller and it shouldn't be sparking.
The FET you are testing with is not really suitable for running the controller as it has an on resistance of over 1 ohm. Since the resistance is so high, the controller voltage will vary widely with load, which will cause the gate voltage to change. With proper FETs, the on resistance should be a few milliohms so the controller voltage won't sag with load, thereby causing the gate voltage to change. Still, even with the 2SK3563, you should be able to precharge the controller and it shouldn't be sparking.
the picture of my breadboard isnt clear. the wires make it hard to follow. I can only tell that it sparks when i leave the switch "on" (gate shorted to negative) and connect the battery. but if i have the switch "off" and then connect the battery and turn the switch "on" i dont see a spark? make sense?
here is the pic anyway:
these is the schematic I'm following:
here is the pic anyway:
these is the schematic I'm following:
Yes, a bit hard to trace out the breadboard from a picture. I only see one side of the caps connected, that would definitely be an issue. There should not be a spark.
What is your 'load'?
What is your 'load'?
Damn I have been trying to setup a pre-charge for the 470v leaf controller I built. I have a contactor on each side.... What I want is to use TO247 or TO264 igbts to pre-charge before the second contactor turns on. But when the first contactor turns on there is a measurable ring across the second contactor as high as 830v even though its not on and there should be no current flowing anywhere. I am going to order a few 1200v rated igbts to try this with.
I have tried 600v rated before scoping to find out the ring was so high.
I need to do this as I have 2 DC/DC supplies to turn on as well and I think if I use a pre-charge I can hook the DC/DCs to the pre-charge as well.
I have tried 600v rated before scoping to find out the ring was so high.
I need to do this as I have 2 DC/DC supplies to turn on as well and I think if I use a pre-charge I can hook the DC/DCs to the pre-charge as well.
izeman
1 GW
the voltages you're playing with frighten me. A LOT. REALLY deadly forces are present. i even handle my 80v system with great careArlo1 said:
Similar threads
