[mikie]

Hi,

Can someone explain to me how the G540 knows to micro-step.
I have read both manuals on the Gecko site and it only states that "Four 10-microstep motor drives", which implies that the motor needs to be capable of micro-stepping.

I am trying to setup my X-axis which is a belt drive unit in Mach 3.
Here's my maths:

Pulley 75mm diameter
Motor 200 steps/rev

75 X 2 pie => 1 Step is 0.5890mm or 43.11 SPI

So in Mach 3 I can set 43.11, but that gives me bad resolution.

What do I need to do next to get this working.

PS: Why do i need to set imperial setting in Mach 3 when my global setup states that I am working in mm? I have checked my travel on a vernier and it outputs close to Mach 3 DRO output.

Thanks in advance
/Michael

[ger21]

When using the Geckos, it's 2000 steps/rev. (200 x 10 microsteps).

Also, circumference would be 75 x pi, (2 x pi x r, or pi x d)

[mikie]

Yes your right my mistake with the formula.

I forgot to mention i'm running 4:1 reduction on the belt drive

Okay so what I have is:
1 step = (2 X 3.142 X 75)/200 = 471.2385 / 200 = 2.3561925 mm/step

or

1 step = 25.4 / 2.3561925 = 10.78 SPI

Therefore given your statement, the G540 is automatically using micro-stepping and i should set:

107.8 X 4 = 431.20 steps in Mach 3 for the X-axis? (which is incidentally what I had above)

Other questions:

2. The G540 manual does not state that a cap is required on the motor supply, unlike the G201 manual. Does this imply that one is not required? I have an un-regulated 50Vdc supply)

3. I have wired my 8 wire stepper in series. I just read that half the current will be used, so given my rates output is 2 Amps, I should use a 1K resistor for these motors. (Just checking with this question)

4. Series V's parallel : I just read the Gecko step motor basics pdf and it explains the voltage/current relationship, but can someone please explain application advantages for this. I have a gantry router which is for personal use. I run a 50Vdc supply.

/M

[guru_florida]

The geckos microstep in 10 divisions so we get 0.18 degrees/rev for each step with standard 1.8degree steppers. Nice. What's even better, is the geckos will gradually switch to full stepping mode when the steppers are going fast. This is the best of both worlds. We microstep to get better resolution at the cost of torque output. We full step for greater torque. Unlike the cheap motor drivers geckos do both modes automatically. Our signals to the motors dont change though, we send 2000 pulses/rev no matter what.

Use parallel wiring for the greatest torque from the steppers. Serial wiring will save you power but at the cost of torque. Your decision, if you dont need the extra torque you could use serial wiring. With serial you can double the voltage to get more power out of them but you still wont get the power of parallel wiring due. Do you have 6 or 8 wire motors?

I dont really get your math. With a 4:1 reduction using a pulley, then you have 200*10*4 = 8000 pulses/rev at the big pulley. So do you have a leadscrew the pulley is driving? What is the threads/inch? (or mm/rev?)

[mikie]

Thanks for the reply. Clarified it beautifully.

The pulley is driving a belt which connects to the gantry. (No lead-screw on my x-axis)

I run an 8 wire stepper, so after doing the math with the serial and parallel setup and knowing that the G540 only accepts 50Vdc, which is what i am running on my transformer output i will stick with my serial setup.

Here is a picture of the router:

[guru_florida]

Now I understand your math now that I have seen your photo. I see where the Pi is coming from. Good work!

...A picture is worth a thousand equations....

[ger21]

I run an 8 wire stepper, so after doing the math with the serial and parallel setup and knowing that the G540 only accepts 50Vdc, which is what i am running on my transformer output i will stick with my serial setup.

Here is a picture of the router:

You want parallel, not serial. Parallel runs much faster than serial at any given voltage, and only needs half the serial voltage for max performance. Are the motors rated 2 amps parallel, or serial?

[ger21]

Let's see if I get this right.

With your 4:1 reduction, 4 turns of your motor gives you 75 x pi of movement, or 235.619mm/ Divided by 4, you get 58.9048 mm per rev. With 2000 steps per rev (200 x 10 microsteps, you get 33.953 steps/mm.

[mikie]

Gerry,

Here is the spec sheet with limited info:
Sorry about the clarity, but this is best I could do with my scanner.

That is quite poor resolution really. Mmm might have to get rid of the belt and re-fit with screws.

[ger21]

That is quite poor resolution really. Mmm might have to get rid of the belt and re-fit with screws.

You could also cut the pulley diameter in half and double the resolution.

With steppers, there's a dilemma that you can't get around. The faster you want to go, the lower the resolution you'll have to live with. Right now, you have about .03mm /step. If you're machine can hold that tolerance, that's not bad for cutting wood.

That sheet doesn't tell you if it's rated serial or parallel. You can check the resistance with a meter and figure out which it is.

[ger21]

Forgot to say to double check my numbers and make sure they're right.

[mikie]

Thanks gerry,

I did check the numbers, they are right.
as mentioned, the sheet is very limited on detail.

/M

[Crevice Reamer]

Use parallel wiring for the greatest torque from the steppers. Serial wiring will save you power but at the cost of torque. Your decision, if you dont need the extra torque you could use serial wiring. With serial you can double the voltage to get more power out of them but you still wont get the power of parallel wiring due. Do you have 6 or 8 wire motors?

Sorry guru, but some of this is confusing. Torque = force. Power = force times RPM.

Both Parallel AND series wirings of the same motor have the same holding torque. That is what the motor rating is--HOLDING TORQUE. In our real world though, we need POWER. That is torque along with motor movement.

Series wiring gives full torque at rest but it quickly drops off as the motor moves. Parallel wiring gives full torque at rest, but it drops off much more slowly and the motor can run at much higher RPMs than Series wiring.

So you WERE right when you said Parallel will give more power.

Series wiring has 4 times the inductance of Parallel and so needs double the Voltage and 1/2 the Amperage,

Parallel wiring needs lower Voltage but uses higher Amperage.

CR.

[mikie]

No wonder i was confused when i tried to read the gecko application notes.
Thanks for all the feedback guys.

Ger,
I could also go another 4:1 reduction, which would give me 16:1 reduction.
So the resolution would then be:

14.726215563702155805293640859123 (16:1 reduction)

and

135.81221810508401985611414474455 steps/mm

so thiswould then be:
0.006642 mm/step which i can live with, even with the slower gantry speeds.

[guru_florida]

Sorry guru, but some of this is confusing. Torque = force. Power = force times RPM.

You're right, it was confusing. When I said "serial saves you power" I meant the electrical kind, not mechanical. All other things equal, wiring in series would double the resistance therefor lower the current require draw...thus saving you electrical power...but at the cost of mechanical power.

I wired my 6-wire steppers in serial at first and like most people doubled the voltage to compensate. However, I was later schooled about inductance and you in conclusion you can't equal the performance of parallel so I rewired them. I wired "end to center" and I have a dangling other end wire (not literally speaking). Wish I knew about this stuff before buying these unipolars, but oh well. They still perform well.

[Crevice Reamer]

You're right, it was confusing. When I said "serial saves you power" I meant the electrical kind, not mechanical. All other things equal, wiring in series would double the resistance therefor lower the current require draw...thus saving you electrical power...but at the cost of mechanical power.

I wired my 6-wire steppers in serial at first and like most people doubled the voltage to compensate. However, I was later schooled about inductance and you in conclusion you can't equal the performance of parallel so I rewired them. I wired "end to center" and I have a dangling other end wire (not literally speaking). Wish I knew about this stuff before buying these unipolars, but oh well. They still perform well.

So you wired them "half coil." That's ALMOST as good as parallel.

CR.

[guru_florida]

So you wired them "half coil." That's ALMOST as good as parallel.
CR.

Exactly.

...And yesterday I accidentilly typed G0 instead of G1 and fed my tool straight into the part at 8in/sec...It wasn't pretty! I don't think I need any more power than they have now (about 100oz-in) when that little display tore off a vice that was very securely held on the table.

[mikie]

Good to know that i'm not the only guy that makes great mistakes.

/M