OK, so let's go through a bit of math here.
AWG-14 wire has a resistance of about 2.525 Ohms per 1,000 feet.

You say you measured the voltage across 4" of AWG-14 wire.

So, 2.525 Ohms / (1,000ft * (12inperfoot/4in)) = 2.525/3000 = 0.8417milliOhms.

I=E/R, or Current = Voltage/Resistance.

9.7mV / 0.8417mOhms = 11.525 Amperes of ripple current.

That's quite a bit of ripple current. You're going to have a good bit of losses from expanding/contracting the magnetic field.

If I'm understanding your scope correctly, you are measuring the peak-to-peak voltage from the lower white line to the upper brown line.

If you now move the upper brown line to 4.85mV, and then move the lower white reference line to 0v, you should get the average voltage.

Take that average voltage, and divide it by 0.8417mOhms (0.0008417) and then you will know what your average current through the motor is.

 

Why does this sound like a pod race?

Gee, thanks - now I just used up another 11 minutes of my life watching that clip.

Back in the spring of 1977, my entire Marine F-4 Phantom fighter squadron went to see the premier of "Star Wars: Episode IV, A New Hope" at the base theater (MCAS Beaufort, South Carolina).

At that time, I was a tall, thin Corporal with hair and mustache of lengths at the bleeding edge of regulations. We returned to the squadron area, and my commanding officer, a Lieutenant Colonel, spun around, pointed at me and yelled, "YOU'RE WOOKIE!"

I did the only thing an enlisted Marine could properly do, by responding "Aye aye, Sir!" which means that the command was understood and would be obeyed.

I received a promotion to Sgt before I hit the end of my active service.

But 33 years later, the nickname's still here.

 

Robb,
Found this relatively new 2A optocoupler gate driver from Fairchild:
http://www.fairchildsemi.com/ds/FO/FOD3180.pdf

Digikey stocks them, here:
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=FOD3180S-ND
Those are an 8-pin SMD package.

Mouser stocks the 8-pin DIP package version:
http://www.mouser.com/ProductDetail...=sGAEpiMZZMt82OzCyDsLFO/4x5xxRv3rGIb9PHVtV/A=

I think these will get rid of a lot of headaches; high-speed optocoupler and gate driver all in one 8-pin package. The driver board can be made small enough to fit right in with your MOSFETs, so that eliminates the additional problem of wire length between the driver and the MOSFET itself. They operate on from 1v to 20v, and can drive MOSFET gates at up to 250kHz.

 

Ok. Thats cool. I get some on order.
Thanks.

 

looking at the spec sheet , I think i need about 200 ohn resistor on the input to get 25ma of drive?

Not sure about the output circuit other than I need 10v. Can you help with the design between pin 7 output and the gate of the MOSFET.
And if you don't mind, an explanation of the values. Parts are on order. I ordered 10 so if you need a couple to play with, no problem.

 

looking at the spec sheet , I think i need about 200 ohn resistor on the input to get 25ma of drive?

I'm looking at the datasheet for the FOD3180.
Recommended current for the emitter is 10mA to 16mA.
Vf @10mA may be from 1.2v to 1.8v; typical is 1.43v.
You are using 5v for the PIC's Vdd.

Let's look at the typical Vf (1.43v) first.
5v-1.43v = 3.57v.
3.57v/13mA = 274 Ohms.
The closest standard value is 270 Ohms.

Let's look at the minimum Vf value.
(5v-1.2v)/270 Ohms = 3.8v/270 = 14.074mA - that's still within specs.
Max Vf:
(5v-1.8v)/270 Ohms = 3.2v/270 = 11.85mA - looks like 270 Ohms is what you want to use for the emitter side.

Not sure about the output circuit other than I need 10v.

10v is the minimum; below that and the driver will shut down. However, you're using 12v batteries, so you are above that.

Can you help with the design between pin 7 output and the gate of the MOSFET.

Actually, you'll want to join pins 6 and 7 together, and use a nice wide trace to a resistor right before the signal connects to the MOSFET gate. Having a wide trace will act as a heat sink for the IC. The value of resistance needed will depend on the length of the trace from the driver IC to the MOSFET, the total gate charge, and parasitics involved.

This is one of those things where you may need to experiment a bit. You could start with a resistance between perhaps 4.3 and 7.5 Ohms (with a 12v supply, 12v/6 Ohms =2A maximum current); a thick-film surface mount resistor (perhaps size 1206 or 2012) is preferable, as the inductance will be far lower than an axial-leaded resistor. It will also take up less space on the board. Adding resistance to the gate charge/discharge path seems counterproductive, but it's necessary to "snub" the tendency to "ring", or oscillate at high frequency due to the L (inductance) of the trace from the supply to the gate, and the C of the gate, when the gate is rapidly charged or discharged. You want a very fast charge/discharge with no ringing. However, in order to eliminate the ringing, you have to snub it using resistance.

You also need a 0.1uF cap directly across pins 5 and 8 of the driver IC, with the leads as short as possible.

And if you don't mind, an explanation of the values. Parts are on order. I ordered 10 so if you need a couple to play with, no problem.

That would be nice.
Gee, I just remembered - I was meaning to send you some connectors for your board. Seems like I should get off my keister and do that.

I'll have to go back and look at what you're using to put it all together.

 

Ok. Here is what i came up with. Not sure about the surface mount resistor (never worked with them) nor do I have any. If the circuit looks good I will start trying to do a board layout. Did some test and have decided to go to 24 volts on the motors only. I will probally try to get the 12 volts from the 24 volt supply since the 12 and 24 need a common ground.
Parts should be in this weekend and Ill forward a couple to you. Going on vacation sometime next week or so for 4 or 5 weeks. Won't get much done for a while.

 

Robb,
That looks good, except:
1) Don't forget that the GND line needs to be connected to the PCB ground as well.
2) I don't know why you have +24v coming on the board and then going right back out to the relay. You could have a wye splice from +24v go straight to the relays, and then the +24v input to the board to provide a current path through the reverse-EMF diodes.

I mean, the way you have part 2 wired will work; but it's another board connection that doesn't really have to be there.

 

Missed the 24 volt thing. Here is a corrected drawing. This should be a good general purpose circuit where people can leave out the relay if they want and also switch different voltages.