Paul & Sabrina's cheap DIY 144v motor controller

[snakub]

Yes R3 was a mistake and I now have it wired just the way you said with the 4.7k's in r10 and r4 again and its giving me the error light. Now no voltage at all is getting to pin 23 so I will try just a bit more resistance on r10 again.

[MPaulHolmes]

I'm confused. Do you mean R8 was a mistake? It's just ohms law at work here. If R8 is 2k, and R10 is 4.7k, it will work, assuming no solder bridges. R4 is totally irrelevant except for acting as a low pass filter. R4 can be anything you want, and it won't affect the voltage at the microcontroller throttle pin. If you measure 0v at throttle input to the microcontroller, then there is a solder bridge to ground somewhere before that pin. It's not that no voltage is getting to the pin because of resistor sizes.

[snakub]

Excuse me r8 is what I meant to say. But if R10 shorts to ground and it is of lesser resistance than R4 and R4 is in series with pin 23 of the atmel than no current will flow to pin 23 it will just short to ground through R10 at least that's the way it looks on the schematic and I am not seeing any solder bridges.

[MPaulHolmes]

The values of R10 and R8 are important, because the formula for the voltage at the microcontroller is:

5v*R10/(R8 + R10)

R4 plays no role whatsoever in the voltage at the pin to the microcontroller. That was really confusing to me for a long time. The reason is, as best as I understand it, inside the microcontroller on an input pin is really high resistance. So, a 1 Ohm resistor or 4.7kOhm resistor on an input pin has almost no voltage drop. I have no idea what's going on inside there, but that's the affect.

So, with R8 = 2k and R10 = 4.7k, the voltage at the throttle pin isn't about 3.5v?

[snakub]

I just soldered a wire from the R4 lead to pin 23 and now I am getting 3.5 volts! YAY But the yellow LED is still on is that normal(not blinking)? and what of the behavior when trying to run a small twelve volt motor? When the current sensor senses real amperage then the throttle will behave correctly? Did Jack Bauer use a starter motor to simulate current needed to test your control board?

[The Toecutter]

Originally Posted by MPaulHolmes

The reason is, as best as I understand it, inside the microcontroller on an input pin is really high resistance.

So the Atmel is supposed to have a very high resistance. I was thinking that something may have been wrong with it.

[MPaulHolmes]

Hurray snakub!!!!! The yellow LED should stay on. That means everything is functioning normally!!!!!! HIp HIP HUrray!

Ya, it's supposed to have a high resistance. I think it keeps the current draw way down that way on the inputs, and I guess with how tiny and low power stuff is these days, the chip doesn't need much current to know what the voltage is on an input. So, they just use what is needed, which I think is super super super super... super tiny current. I would like to see an equivalent circuit for a microcontroller input pin some day, because I honestly have no idea what's going on in there.

[The Toecutter]

Now that I know what to expect with one, I'm going to purchase and assemble one. I think it is interesting that this can control virtually any power setup. Perhaps I should try IGBTs instead of MOSFETs. Assembling this appears to be very straightforward as long as the right parts are used.

On a side note, is there a reason that you chose the current sensor that you did? Perhaps there is a cheaper substitution.

[MPaulHolmes]

The LEM is the Rolls Royce of current sensors. I was going to use a Tamura, which is almost half the price, and probably works fine, but Fran was talking to me and he said he uses the LEM all the time, which was $27.50 each, so I went with his suggestion. However, as he sometimes says, "my projects are not cost sensitive". hehe.

If you do IGBTs, you should talk to Adam and Jack Bauer. They are the experts.

[The Toecutter]

Originally Posted by MPaulHolmes

However, as he sometimes says, "my projects are not cost sensitive". hehe.

They should be! The idea behind the open source controller is its affordability, right? Of course, reliability is also a factor... that being said, there probably are cheaper substitutes that will last the life of the controller, however long that may be(at least 2,000 miles so far in some examples).

When I build it, the control board will definitely see what happens when it is abused in a racing environment with 500A at the power board for ~3 minutes straight(that's all the PbA battery will last!). You'll get another data point there. Later on will be an upgrade to 160AH LiFePO4, to test it further...

If you do IGBTs, you should talk to Adam and Jack Bauer. They are the experts.

Thanks.

Snakub knows a bit about them with regard to controller applications as well. The amount of current I want won't be too demanding; his home built controller can output 1600A to the motor, but that was with no over current protection added(hence his need for your open source board)! I'm sure he will set the limit to the max 800A though, which is still plenty for a Netgain 9"(~180 lb-ft).