high current switch or rather a contactor

[notger]

hi

I'm searching for a quite compact small high current(70A constant/120A peak) switch for quite long time now.

it schould have two positions like on/on/off or on/off/on for precharging the controller capacitor

what i have found unitl now is http://www.conrad.at/ce/de/product/...-IEin-IIIEin-II-32-VDC-200-A?ref=searchDetail
but this one is giant in it's measurements, so nothing for my purpose

and this one http://www.conrad.at/ce/de/product/...t-19353218-IP65-tastend-1-St?ref=searchDetail
would be ideal with it's 120A peak but it'S concidered to take just 20A constant at 220AC

do you think it would take 70A constant at 36 DC without to big of an electr. resistance?

this one http://www.conrad.at/ce/de/product/...19-IP65-rastend0rastend-1-St?ref=searchDetail
would even have the two positions on/off/on but is just 12A constant at 220AC

I anyway use 2X4mm² for + and 2X4mm² for - so i could also use two of these switches, than it would be 24or 40 official and -200 Apeak? what do you think

I 'm not fully satisfied with this choice, so does anyone of you know better smaller and more efficiant switches for high current and best with on/on/off.

another idea is to use two of these contactors http://www.conrad.at/ce/de/product/...esser-80-A-Max-75-VDC-1120-W?ref=searchDetail

does anyone of you have experience with contactors and driving shock, cause im just
driving around in the woods and there quite extreme

thanks in advance

gernot

 

[schwibsi]

Why don't you use an antispark mosfet as described in the German forum and use any cheap switch you like.
A switch that can take this much current is going to be big and expensive.
I'd recommend a IRF4368, 2 resitors, 1 capacitor, 1 diode, heat shrink, soldering iron and you're good to go.

 

[notger]

if i look at the small tiny tootsies of the fet i doubt that it will transport the current without big heat production and resistance?
but i have no experience with it?

what do you think

 

[schwibsi]

Well, for a educated guess, it would help to look at a datasheet.

This is the fet I was referring to.
http://www.irf.com/product-info/datasheets/data/irfp4368pbf.pdf

Inside the TO-247 it can take 195A.

The typical RDSon (resistance when fully switched on, which is, what you're interested in) is just 1.46mOhm.

Every connector or soldered connection you could make gets warmer than that fet.

 

[Matt Gruber]

Moroso sells a 175 amp 36v switch peak 1000 a.
no. 74101 about $40
check auto racing suppliers

 

[Teh Stork]

Well, for a educated guess, it would help to look at a datasheet.

This is the fet I was referring to.
http://www.irf.com/product-info/datasheets/data/irfp4368pbf.pdf

Inside the TO-247 it can take 195A.

The typical RDSon (resistance when fully switched on, which is, what you're interested in) is just 1.46mOhm.

Every connector or soldered connection you could make gets warmer than that fet.

Well, Anderson Powerpoles come in at 0,6mOhm. 4mm bullets are around 0,3mOhm. Soldered connections are even lower.

To stay on topic tho, MOSFETS using the attatching wire as heatsink is more than adequate for relay usage. But remember that current can only be blocked in one direction. While relays do work, they are archaic compared to MOSFETS.

 

[d8veh]

if i look at the small tiny tootsies of the fet i doubt that it will transport the current without big heat production and resistance?
but i have no experience with it?

what do you think

Use multiple FETs to reduce the current through the legs. I've got these on several of my bikes:

 

[friendly1uk]

The fet method is very tidy, but if you really want a contactor then a 25A one would probably do it. You see, a contactor is usually 3 phase, and that 25A will be per pole. Summing up to 75 amps. Normally they don't do full duty, but your not doing full voltage. I can see the contacts running a little cooler if anything. You would add auxiliary contact blocks to ensure the resistor dropped in before the main contacts. Maybe 8 cubic inchs. Massive but straight forward for some folk.

 

[Drunkskunk]

if i look at the small tiny tootsies of the fet i doubt that it will transport the current without big heat production and resistance?
but i have no experience with it?

what do you think

Use multiple FETs to reduce the current through the legs. I've got these on several of my bikes:

Do you remember the title of the thread that pic came from? I seem to remember there where some improved versions of that scematic.

 

[friendly1uk]

The drawing just shares the load, but a few r/c networks could have a progressive switching approach working. Least that was my first thoughts, but why. It works.

 

[danielrlee]

I've been doing similar research recently and came to the conclusion that the solution posted by Dave is the one to go for. I'm still yet to do I myself though.

I think you'll find the info you're after in this post by teklektik on my thread from a couple of weeks ago.

http://endless-sphere.com/forums/viewtopic.php?p=848677#p848677

 

[d8veh]

Do you remember the title of the thread that pic came from? I seem to remember there where some improved versions of that scematic.

The one above is very simple and it works. There was a lot of intellectual masturbation that came up with a theoretical improved version if you have access to electronics materials, parts and equipment. The one above can easily be made by anybody that can solder. you don't even need a PCB, though it's easier to mount it on a bit of copper clad with a groove Dremmeled into it to separate the source from the drain.

http://endless-sphere.com/forums/viewtopic.php?f=3&t=40142&p=782947&hilit=soft+start+switch#p782947

 

[Matt Gruber]

will the FET's work with 3.7 volts?
i'm making a scooter go slow in the house for my dad, he broke his hip,
thanks!

 

[Teh Stork]

The one above is very simple and it works. There was a lot of intellectual masturbation that came up with a theoretical improved version if you have access to electronics materials, parts and equipment.

http://endless-sphere.com/forums/viewtopic.php?f=3&t=40142&p=782947&hilit=soft+start+switch#p782947

W/o reading that topic. Swapping that 10k resistor with a fixed ~12V zener + a 1M resistor is an obvious and big improvement. This protects agains overvoltage on fet gates. The 1M resistor is used to turn off the fets once the pack is disconnected.

 

[dnmun]

instead of all this you can just turn off the battery by turning off the circuit current in the BMS with a small DIP switch.

 

[Drunkskunk]

The one above is very simple and it works. There was a lot of intellectual masturbation that came up with a theoretical improved version if you have access to electronics materials, parts and equipment. The one above can easily be made by anybody that can solder. you don't even need a PCB, though it's easier to mount it on a bit of copper clad with a groove Dremmeled into it to separate the source from the drain.

http://endless-sphere.com/forums/viewtopic.php?f=3&t=40142&p=782947&hilit=soft+start+switch#p782947

Cool, thanks. Was a lot of good info in that old post. JH's design should work just fine.

 

[dnmun]

The one above is very simple and it works. There was a lot of intellectual masturbation that came up with a theoretical improved version if you have access to electronics materials, parts and equipment. The one above can easily be made by anybody that can solder. you don't even need a PCB, though it's easier to mount it on a bit of copper clad with a groove Dremmeled into it to separate the source from the drain.

http://endless-sphere.com/forums/viewtopic.php?f=3&t=40142&p=782947&hilit=soft+start+switch#p782947

Cool, thanks. Was a lot of good info in that old post. JH's design should work just fine.

there is no zener to protect the gate. i think the design was shown inadequate and updated in that thread. just that it is not needed if you turn off the battery before connecting to the controller.