Need a snubber or DC circuit breaker

[bwilson4web]

Hi,

The NHW11 traction battery is actually two banks 18 and 20 modules in series. The orange safety plug opens the interconnect between the two as does a separate relay. So I intend to preserve this safety feature in both the ReInVolt and NHW11 pack. To simplify the language, I'm going to adopt the names:

• NHW11 pack - means the pack using the older style battery modules, the one currently in the car.
• NHW20 pack - this is the ReInvolt pack with the newer style batteries.

My plan is to carry all four conductors between the two battery packs but via my custom, battery controller. I know there is a risk of unbalancing the NHW20 pack but I'll just have to be careful. The main thing is when I disconnect the NHW11 pack, the NHW20 pack will be 'normal'.

Now I've been looking at the IBGT vs MOSFETs and am starting to lean towards MOSFETs as the switch. The diodes on IBGT suggests that if I reverse the current flow, the diode may just keep it latched up. Perhaps I'm misreading the datasheets but I get the impression the MOSFETs are bi-directional and this is my planned operational scenario.

My assumption is the MOSFETs don't really care which way the current flows and need only a modest gate control voltage that I can provide via an audio isolation transformer into an RC filter. I'm also making the assumption the MOSFETs are without a gate bias, OFF. This means loss of the AC gate control signal should make them switch off.

The RC time-constant and gate control frequency can make the transition from ON to OFF fairly quick. By driving the AC gate control signal from an ignition switch, 12 V source, I can be assured that turning the key off and/or loss of the 12 V battery will 'safe' the system. . . . At least that is my understanding.

What I'll do is work up the MOSFET switch design and test it with some cheap Radio Shack parts. If it works as expected, I'll get some power MOSFETs and build out a brassboard for 'smoke' testing.

BTW, since I plan to use these as ON/OFF switches, there will be a very low duty cycle. My expectation is little 'heat sinking' will be needed.

Thanks,
Bob Wilson

[JasonG]

I may be wrong but,

If you drive any transistor with AC the out put wil match the input, you will get a modulated signal.

Yes, MOSFETS and IGBTs are off with no V applied. However their internal draw is so low you need to place a "pull down" resistor from gate to ground. For your application 100K should be low draw. Without this the applied 12V gate signal will take some time to slowly decay before it shuts the output fully off.

I dont think MOSFETS can handle reverse polarity, but if they can it would be cool. I havent worked with them before so I can't say. Testing with 5V will prevent most components from frying if something is wired wrong. Old computer power supplies have 5 and 12V leads and can be found in the trash for my favorite price $0.00 !

[bwilson4web]

Originally Posted by JasonG

I may be wrong but,

If you drive any transistor with AC the out put wil match the input, you will get a modulated signal.

A current control device, the output is proportional to the base current but when used as a switch, we 'bang the stops' and either no current, it is OFF, or to saturation, ON.

Originally Posted by JasonG

. . .
Yes, MOSFETS and IGBTs are off with no V applied. However their internal draw is so low you need to place a "pull down" resistor from gate to ground. For your application 100K should be low draw. Without this the applied 12V gate signal will take some time to slowly decay before it shuts the output fully off.

I dont think MOSFETS can handle reverse polarity, but if they can it would be cool. I havent worked with them before so I can't say. Testing with 5V will prevent most components from frying if something is wired wrong. Old computer power supplies have 5 and 12V leads and can be found in the trash for my favorite price $0.00 !

Will do and soon.

Bob Wilson