Need help wiring brushless controller

[doanwannapickle]

Excellent. I can see I'm going to have a few questions.
I'll start with the easiest for me. Could R1-R2 be replaced with a pot to make that tunable?
I'll get the components on order and try to puzzle out by myself as much as I can.

[CitizenOfDreams]

I'll start with the easiest for me. Could R1-R2 be replaced with a pot to make that tunable?

Absolutely. You can also replace R5 with a pot to adjust the 2.5V level shift.

[doanwannapickle]

Absolutely. You can also replace R5 with a pot to adjust the 2.5V level shift.

I suspected the 2.5V was probably tunable but I wasn't sure which resistor(s) would do the trick. What value pots would be appropriate?
What is .tran 10?
I did a quick sketch. Could you give it a once over and see if I'm remotely on the right track? The way I've got the transistor seems odd to me.

[CitizenOfDreams]

Your 2.5..4.8 voltage will come from pin 7 (opamp output), not pin 8 (that's opamp positive power rail).

tran10 on my diagram is the 10V power supply.

The "transistor" (TL431 voltage reference) on your sketch is wired correctly.

If you simply replace R1+R2 with a pot, the adjustment would be way too coarse. It is best to use R1+pot+R2. I am too lazy to solve quadratic equations and calculate the resistors value properly, but R1=3.6k, Rpot=200R and R2=1k will work good enough (the divisor will adjust from 4.0 to 4.8).

Same thing with R5: replace it with something like 900R resistor plus 200R pot.

[doanwannapickle]

Makes sense that the output voltage comes from pin 7 but now I'm a bit foggy on V1 and V2.
If V1 is the 10V power supply, is V2 the 0-10V input from the G540, the one Gecko labels VFD Out?
I'll try to find time to update my sketch later today.

[doanwannapickle]

Try number 2. I think maybe I've got it this time.

[CitizenOfDreams]

Sorry, I misspoke. V1 on my diagram is the 0..10 voltage from your Gecko. V2 is the 10V power supply.

So Gecko pin 8 goes to R1, Gecko pin 9 goes to opamp's pin 8.

[doanwannapickle]

Like so:







BTW, I told my wife you were penning this circuit I've been stuck on. She was duly impressed. Kudos.

[CitizenOfDreams]

It's not every day that you get a chance to impress someone's wife with you knowledge of circuits. :-)

Your diagram appears to be correct now. Is it time to do some soldering? A "smoke test" as we call it?

[doanwannapickle]

I got the order in with Mouser. Eight bucks shipping for five bucks worth of components.
And then I forgot the pots. Sigh.
I ordered some from eBay: https://www.ebay.com/itm/10PCS-3296W...CTOr:rk:5:pf:0
They will take a while to get here but I should be able to try out the circuit and see if I'm in the ballpark once the Mouser stuff gets here.

Speaking of which, the TL431 came in two flavors. The were cheap so I got one of each.
https://www.mouser.com/ProductDetail/511-TL431AIZ-AP
https://www.mouser.com/ProductDetail/595-TL431CLPRE3

Do you have a preference or is either fine?

[CitizenOfDreams]

Speaking of which, the TL431 came in two flavors. The were cheap so I got one of each.
https://www.mouser.com/ProductDetail/511-TL431AIZ-AP
https://www.mouser.com/ProductDetail/595-TL431CLPRE3

Do you have a preference or is either fine?

Both ST and TI are well-known manufacturers. The additional letters in the part number may mean things like temperature range, pin plating, packaging and so on - which are not relevant for this application. So either one will work just fine for you.

[doanwannapickle]

One is described as "Voltage References 2.5-36V Programmable Adjustable" and the other as "Voltage References Adjustable Shunt ".
I have no idea what either description means.

[CitizenOfDreams]

Just different ways to describe the same thing.

A voltage reference is a circuit that outputs predetermined stable voltage (that other voltages are compared to).
A shunt regulator is a common variety of such a circuit. In our case, R4 and TL431 make a shunt regulator.

[doanwannapickle]

Work has been relentless lately but I finally got a bit of time to get back to this project.



It's a bit crude at the moment (understatement) but the results were good. 2.7v out with 0v in and 4.77v out with 10v in. Once I get a couple of trim pots it should tune in dead on.

I plan to attach it to the G540 later today but I don't foresee anything unexpected there. I'll wait to get it all soldered to a board with the trim pots before I actually attach it to the controller.

[CitizenOfDreams]

I'm happy to see my circuit in flesh. :-)

Did you check where the extra 0.2 volts come from? TL431 should be accurate within 2% at room temperature, and the opamp input offset should not be that large either. What is the tolerance of your resistors?

[doanwannapickle]

Merry Christmas.

The closest resistor value I had for R1 was 3.3k and the closest for R4 was 510R. Then there's the breadboard I couldn't have paid more than a dollar for and the fact that it's all held together with alligator clips. No telling what the actual resistance values are. I was pleased it landed as close as it did considering the way it's cobbled together.

I was thinking the trim pots would take care of all that once it's soldered to a board. Am I incorrect?

I'm guessing the 510R is the main culprit. Might it be prudent to add a third trim pot around R4? Or find a pair of resistors to get that 500R spot on?

[CitizenOfDreams]

I'm guessing the 510R is the main culprit. Might it be prudent to add a third trim pot around R4? Or find a pair of resistors to get that 500R spot on?

R4 is the only resistor in this circuit whose exact value does not matter (to a certain point). TL431 will regulate the current through that resistor so that it would always have exactly 2.5V on the cathode.

[doanwannapickle]

Well, I connected things up to the G540 yesterday. It was not the drama-free event I was anticipating. The voltage fluctuated wildly around 1/2 volt.
Disconnected it and checked it at the Gecko. It was fluctuating wildly so your circuit was doing exactly what it was supposed to be doing. Mirroring the Gecko.
Spun my wheels for a couple of hours trying to figure out what's going on to no avail.
I'm inclined to think it's something with Mach3 rather than the G540.
The fun just never ends.

[doanwannapickle]

Just love it when a solution turns out to be trivial. The PWM base frequency had somehow gotten changed back to 5 Hz rather than the 50 Hz that the Gecko likes.
Everything is smooth and stable now.

[CitizenOfDreams]

Good to know you got it working! Just in time for the New Year, too.