WheelchairDriver

I don't suppose you know what MOSFETs they are, or have pics of the other side of the board showing them??

The side with the writing is folded on to the board so not visible without bending them up. I did compare an 80 to a 100 years ago and they were the same. So its only a firmware difference.

Could be a firmware difference, or a different shunt in the current measuring circuit.

Its firmware or the programming tools wouldnt have different limits on the Amp settings? Or maybe both... Theres a feeble 50A one too.

I have a few 50 amp modules, will open one up and compare with some help from my Brother or Carer.

Steve

Do you realise how feeble that is? My model helicopter weighs 8lb and has a 70A controller which isnt enough, and that actually pulls about 85A at max pitch (no current limiting). Just about adequate to get grandma to the dining room...

Yup, which is why I never use them. Probably be ok for a kids indoor chair but nothing else. Will ComparethePowerModule.com

Steve

To increase efficiency and allow lighter wiring and smaller motors/controllers most of the serious model helis are using lipo and 10 or 12S now. So are at 40 to 50 volts. As we should be!

Convert rear light to LED

Red LEDs in the red, white ones in the turn lights... I have ordered some red ones for those too. Because I have no need of turn signals. They will light up red!

So Projector lamps for the front. And this F55 one for the rear (because I like it) but turned to LED. I may not fit it. Depends on the seat... Might use red projector LEDs on the rear instead. Right now its taking 3 watts at 16 volts... So the 3kwh battery would run it for 44 and a half days.

Burgerman wrote:To increase efficiency and allow lighter wiring and smaller motors/controllers most of the serious model helis are using lipo and 10 or 12S now. So are at 40 to 50 volts. As we should be!

Interesting trade-off there for helis/quads between battery weight and motor/wiring weight, can't be one-approach fits all...

The reason I was asking what MOSFETs they use is to think on whether they could be replaced with a newer type with a lower Rds(on) (on resistance) and then tweak the current limit. I would expect that current limiting uses a sense resistor that could be changed. The firmware will most likely measure voltage across the sense resistor and once over the limit will roll-back to a lower voltage for a period of time. Changing the limit by changing the sense resistor will retain the same normal v rollback ratio without needing a firmware change.

50 Amp PM insides...



Steve

Lift up a power mosfet and see what its numbers are. Irving wants to know!

PG 50 Amp PM 7 Photos inside a 50 amp PM.

Steve

Trans MOSFET N-CH 55V 85A 3-Pin(3+Tab) TO-220AB

Does it help?

Where is the sense resistor?

but none show the MOSFET type.

There are 8 MOSFETs in the H-bridge, arranged in 4 pairs. At any one time only 2 of the 4 pairs is on, each MOSFET carrying 1/2 the current, so if limited to 100A each would be carrying 50A. A typical MOSFET from as little as 4y ago will have an on resistance of around 20milliohm, so heat losses = I^2 * R = 50 * 50 * .02 = 50W, WORST CASE at stall per MOSFET, which is fine with a decent heatsink. It'll get hot (case temp ~100degC) but wont die.

Replace that with a current top-of-the-range HexFET. On resistance 6.5mOhm, so for the equivalent heat loss it can handle 100A, or putting it another way a 100A controller could, in theory, be upgraded to 150A with a reduced heat loss of 36W per MOSFET and a case temperature of around 60degC (assuming its just a sense resistor that needs to be changed as well).

And the price? £14 inc p&p

That's a lot cheaper than a RoboTeq! In fact the Roboteq's oft quoted 32 MOSFETs (per channel?) is arguably overkill using current technology.

Current technology

http://www.eurotech.co.uk/dynamic/detai ... +RECTIFIER

What about one of these?

Burgerman wrote:Trans MOSFET N-CH 55V 85A 3-Pin(3+Tab) TO-220AB

Does it help?

Where is the sense resistor?

Not really, could be any one of 100s of devices...

our posts x-ed

Burgerman wrote:http://www.eurotech.co.uk/dynamic/detail.php?part1=AUIRF2805&man=INTERNATIONAL+RECTIFIER

What about one of these?

Actually those are pretty good devices, apart from they are no longer manufactured, not dissimilar to the ones I was suggesting. 55v a bit low for my liking on the regen spikes, but if that's what's fitted then just changing the sense resistor would work to 125 or 130A. I'd not take it to 150A

Burgerman wrote:
Trans MOSFET N-CH 55V 85A 3-Pin(3+Tab) TO-220AB

Does it help?

Where is the sense resistor?

No, thats what it said on the 50A photo one!

Where would I find the MOSFET type? on the top side of it?

Steve

Burgerman wrote:

Burgerman wrote:
Trans MOSFET N-CH 55V 85A 3-Pin(3+Tab) TO-220AB

Does it help?

Where is the sense resistor?

No, thats what it said on the 50A photo one!

Must have missed that, I'll go look again. However the device you linked to is still manufactured (just no longer stocked by my usual supplier) but its a recent (11/2014) product, unlikely to be found in a >5y old controller..

steves1977uk wrote:Where would I find the MOSFET type? on the top side of it?

Steve

Yes on the side opposite the tab.


Incidentally these devices are limited to 75A by the TO220AB package. The silicon is good to 175A. So with only 2 parallel devices 150A is the maximum you can upgrade it to.

OK I give up, I've looked again and can't see it..

As far as current sensing, I'd hazard a guess the two devices in the middle on the lower row are either resistive shunts or hall-effect sensors

No. Those just go to the relay. And that would do what the roboteq does and sense battery current?

Burgerman wrote:No. Those just go to the relay. And that would do what the roboteq does and sense battery current?

yes, I can see that now. Might not be sense resistors but just some carefully sized PCB tracks..

still can't see where that info was written, must get new glasses

http://www.irf.com/product-info/datasheets/data/irf1010n.pdf

Steve

steves1977uk wrote:http://www.irf.com/product-info/datasheets/data/irf1010n.pdf

Steve

Are those the actual devices?

OK, slightly better than I expected 11millohm at junction temp of 20degC, so about 17milliohm at working temp. Still nowhere near the best available

It's just a general specshet on the IRF1010N. If these were replaced, what would be a better replacement?

Steve