Gerald,

Will do. Is an hour long enough?

Gerald,

I prefer half-coil connection. That costs me about 30% of the full-coil torque, but it gives me much higher speed. Assuming that the motor is the PK296A2A-SG7.2, the gear box could never deliver all of the torque that the motor generates but the gear box can deliver the speed (up to 500 RPM, which would be about 40 inches per second - which would be impossible, given the mass of the various axes).


My work is almost 100% cutting cabinet sized parts from MDF. I can use all the speed that a motor can give me. On the other hand, if I were doing 3D or work that required slow speeds (1 to 2 inches per second), wiring the motor full-coil would be more practical.

Thanks Mike, my reasoning is that the very slow CNC work (like 3D) will not require the high torque that full coil will provide. I will pursue the half coil configuration, and a supply voltage that gets the motors nicely "hot" but not quite toast. . . . :)

Thanks Doug, an hour should do.

Gerald,

I hope you don't mind but I stopped the test after 30 minutes. I set the motor for about 10 RPM and forgot about it for the first 15 minutes. After that point, I kept checking the motor and the temperature didn't change between 15 minutes and 30 minutes. It was warm but I could have easily kept my hand on it forever. Nowhere near as warm as a fresh mug of coffee which is too hot for me to keep my hand on. Maybe as warm as a ceramic mug of coffee that was poured 5 to 10 minutes ago.

If you need me to go for a full hour or run it at a slower RPM or any other test, please let me know.

Thanks a lot Doug! If the temperature stabilised within 30 minutes, then the test was long enough. What was the approx ambient temperature in your workshop at the time of the test?

It was in the mid 80s in my garage. Sorry but I don't know what that is in Celsius.

85F is about 29C :)

Doug,

Have to give you a hard time on that last unacceptable answer. ;)

Your a MechMate builder of international means. :)

http://www.onlineconversion.com/temperature.htm

Greg,

I know. I know. Pretty lame excuse on my part but I figured that Gerald had the conversion equation in his head.

PS. Thanks Heath.

Doug,
I still prefer F Measurements for hot weather but like C measurements when talking about cold weather. :confused:

Connected a motor/gecko/supply together tonight and left it switched on, non-moving, unipolar/half-coil/39V supply for 2 hours. . . .

Well, if I did not measure the slight rise in temp, I would have sworn it did not heat up at all. The increase in temp was only 5.2ºC [9.6ºF]

Now I must find a simple way to feed steps to the gecko to dupe its current reduction mode . . . . .

Another Data Point
 

Conducted this test tonight,

Motor - PK296A1A-SG7.2 (unipolar/half-coil 7.7mH, 88.8Volt "capable")
PS - 70 V
Gecko Drivers - G203V (current limit resistors 12.8k for 1.5Amp?)
Condition - No motor moving.
Time - 2 hours
Ambient Temperature - 90 deg F

Increase in Temperature,
X motor - 6 deg F
A motor - 6 deg F
Y motor - 1 deg F
Z motor - 7 deg F

All four G203V's showed no increase. Actually, the temperature decreased as the ambient temperature decreased.

Conducted a second test on the Z motor. I operated the Z motor up and down at 200 ipm for 5 minutes. The increase in temperature was an additional 6 deg F for a total (1 hour, 5 min) of 12 deg F.

Did some testing on my setup:
4 x Geckos G 202's ...with 100k resistors...between 4 - 5 Amp
4 x MotionKing 34HS9801
42 volt AC transformer
Set X-axis (+slave) on SLOWWWWWW rev's
Set Y and Z on a nice Fast pace.
Stuck some tape on keyboard and waited

Observed the following after about 30 minutes:
the Y + Z motors were hotter than the X- motors. All four nice and hot...could keep my hand on it without any problem.
What I did notice.....not measured.. is that the Gecko's on the X-motors were or felt hotter than the Y + Z Gecko's.

Run a job in Mach3?

This test is at my home workbench - no Mach3 there.....yet

Greg, even though your motor is the A1A version, your measured results agree with mine.....While the geckos are idle in current reduction mode, the heat rise is minimal (below 10 F).

I have a feeling we can set these Oriental Motors quite a lot higher than the cheaper motors (our money must have bought us something extra?). Greg, your options are limited by the A1A version - you have about the best you can do with the voltage. But, is your current set resistor on the gecko at 13kiloOhm, verified by measurement?

Kobus, you bordered on an interesting test . . . . . :)
If you had set all 4 motors for a low speed, then you should have blown your main power supply! The calc for power supply size assumes that only some motors are drawing peak power, not all of them.

Well, glad to hear that the MotionKings are staying cool enough. That passes their first test. Watching your tests with those "cheap" motors keenly!

Gerald,

I'll check resistance tonight.

Watching this thread with interest. May already be "upgrading" to A2A version.

Greg, nobody is yet saying that the A2A is performing better than your A1A. The jury is still out. :) (Suspect there is not much difference except that the A2A gives you the option of going bipolar with the G203)

Greg,

I have a PK299-02, which has inductance similar to the PK296A1A-SGxx motor. In half-coil mode, the PK299-02 is very impressive. The problem with that motor is that when connected full-coil, the Gecko G20x drives cannot handle enough voltage to take full advantage of the stepper motor. I would not feel handicapped using the PK296A1A-SGxx motor, when wired half-coil. In theory, the A2A motor would perform a faster, but I doubt that you would be able to tell much difference - if both motors were connected to the proper voltage power supply.

Spent all morning running a PK296A2A-SG7.2 motor, with thermocouple attached, at 25mm [1"] per minute equivalent router speed, with no load attached, at increasing voltages until the temperature stabilised.

At 39V power supply, the case rose 23ºC [41ºF] above the ambient temp.
At 45V, the temp increase was 27ºC [49ºF]

And finally, with a 50V power supply (which is about 30% higher than recommended by the Mariss induction-based formula) the temp settled at 31ºC [56ºF] above the room temp. On a hot 40ºC [104ºF] day, the motor should be at 71ºC [160ºF] . . . . . . . too hot to hold, but still well below the most conservative allowable temp for a stepper.

At this stage of testing, it seems this motor (wired unipolar) will be okay on a 50V supply . . . . will test at other speeds as well . . . . . it is a long weekend here. . . . .

Tested the bare motor at an equivalent router speed of 320 inch per minute. It got a LOT hotter. When the temp reached 50ºC [90ºF] above ambient, I trimmed the power supply voltage back to hold the temp steady at that point. The temp steadied (at 50ºC [90ºF] above ambient) when the power was reduced down to 36 Volts.

This is all very confusing as the conventional wisdom is that a motor gets the hottest at very low speeds. Maybe the G203 and G201/2 behave differently to each other - the conventional wisdom comes from the G201/202 days. Need to repeat this test with a G202 . . . . .

This result is worrying, because 320 inch per minute could be a common speed for a lot of work.

Gerald,
I have been running my table today since 6am (7 hrs straight) average speed of 200ipm and motors are just warm to hot.....Not to hot to hold, but like a fresh cup of coffee.
FYI. I don't have any measuring devices here for a real measurement.
Shop temperature is 89F....and climbing to 96F later on.

Sean

Sean, your motors are wired bipolar(series), the voltage calc says they are capable of 78V and you are supplying them 56V, so they should definitely run on the cool side. In my unipolar(half-coil) case, the calc says 39V, and I am experimenting with voltages between 36 and 50V.

Will do more testing tonight, particularly with a G202 drive that has served us so well.

I've got the a2a motors, which I intend to wire bipolar. I'm thinking of getting a dual output toroid xfmr which will give me either 25 or 50 volt output depending on how it's wired. So...70V dc would be the likely output which, if I understand this correctly, is a little on the high side for when motors are operated at low speeds. Here's the Q....

What if I feed the toroid with a variac? Would I maintain required isolation?

Chuck

Chuck, I think you will be okay, but I am not a licensed electrician.

If the Variac feeds the toroidal transformer, the circuit will work. I do exactly that when I'm testing various motors. NEVER use a variac in place of a transformer. Also, make sure that the variac only affects the toroidal transformer and not any other auxiliary power supplies.

Seventy volts is a little high for an A2A wired bipolar series, but within the permissible range (77 VDC is the computed maximum voltage).

Thanks Mike and Gerald,

I'll start off with the variac and see what my optimal voltage is. Using the variac on a permanent basis is not that appealing.

Could I trim the output dc voltage, either with series diodes or using half a bridge rect.? Possibly a circuit that could be selected as needed?

Chuck

Chuck, I am about to run a machine for a couple of months with a variac before its toroid (Our hottest month is February). I don't think it is dangerous if one respects the output as mains supply with a distinct live/hot and neutral, and there is a RCD breaker feeding it.

Trimming the DC voltage side is unheard of in my book, but then my book on this is very thin.

Hi all,

I’m having trouble deciding/working the power supply needed.
Could someone be kind enough to check my calculations?

Motor: PK296A2-SG7.2
Connection: Unipolar
Current: 3 amps
Inductance: 3.5 mH /Phase

For Volts:
32 x SQRT 3.5 = 60VDC
Therefore need a toroidal transformer
60/1.4 = 42.857 VAC

So use either
40VAC x 1.4 = 56VDC or 45VAC x 1.4 = 63VDC

Motor amps
= 4 off x 3 amps = 12 amps
= 0.67 x 12 = 8.04 Amps

For VA
8.04 x 56 = 450.24 watts, or 8.04 x 63 = 506.52 watts

Plan to go with the 56VDC option
thus looking at using a toroidal of 45 + 45V, 11.1A in parallel, 500VA

Cheers,

The voltage appears to be too high - check the inductance rating of the motor

I think this is where the problem is. Where did you get the 3.5mH figure?

Gerald,
Sorry you posted while I was writing.

This is where i get lost.

if you look in the

"2000-2001 ORIENTAL MOTOR GENERAL CATALOG" page B-281

for a PK296A2A-SG7.2 motor it lists a current of 3A and a inductance of 3.5mH.

would attach the pdf if i knew how :(

going to emial oriental, but redo the calcs to 39 volts

I agree with Gerald and Heath. When I went to http://www.orientalmotor.com and entered PK296A2A-SG7.2 for the motor and then clicked on "More Specifications", the inductance for that motor, Unipolar, is 1.5mH per phase.

The formula 32 * SQRT(1.5) = 39.14V. Keep in mind that 39.14V is the MAXIMUM voltage. I would use 35V or less with that motor. A 25VAC transformer would give about 25V X 1.4 = 35VDC.

Figuring the VA of the transformer is 3A X 4 X 66% = ~ 8A and 8A X 25VAC = 200VA (but I would use at least 250VA or even 500VA - the prices are nearly the same).

If I were building the power supply, I would use a 25 + 25 toroidal transformer rated at 500VA to get about 35VDC. If possible, I would use a 20 + 20 volt toroidal transformer to get about 28VDC so that the motors wouldn't get so hot. (I have used an 18 + 18 transformer @ 250VA with great success with 4 each PK296B2A-SG3.6 motors.)

thanks Mike,

Have you ever looked at using a switch mode type?

cheers,

Jason,

A switching power supply is normally used in applications that require constant current at a fixed voltage. Stepper motors require varying current and can handle a range of voltages, so a switching power supply is not optimum for stepper motors. In fact, if you use a switching power supply, you should still use a large 10,000 uF (or larger) capacitor to help the power supply deal with the changing current requirements.

(Gerald will probably get a kick out of this, since, I basically defended the switching power supplies several years ago on the Shopbot forum. He had more experience than I and his wisdom proved to be correct.)

Jason, you are indeed correct, that is what the 2000-2001 catalogue states for that motor.
http://www.orientalmotor.com/product..._B267-B288.PDF <---obsolete, don't save

However, the current catalogue has it at 1.5mH and our experience is that this is the figure to be used for the motors currently being sold.

Here are Oriental's current catalogues:
http://www.orientalmotor.com/products/CatalogPdfs.htm

The 2000-2001 version is in the Archive section.

I defended the unregulated transformer/rectifier/capacitor because I could understand it and because I had just blown the ShopBot-supplied switch-mode because it became clogged with sawdust. It was a short experience and obstinacy that told me this stupid thing with the 3 components is going to work better out here than that thing with a million components and a teeny cooling fan :)

Sometimes the simplest solution to a problem is the best one. In my line of work, we only use switch mode power supplies where high efficiency is needed or space or weight is at a premium. I am also with Gerald on the linear supplies for a cnc table. :)

Hi everybody!

In first place, this forum is pretty usefull and kind. I've been reading this forum for over a year learning about cnc routers. And it is time to build my first router!

I have a million questions to be done, and I will start by asking about my choice of Avel Y236801 500VA 25V +25 V Toroidal Transformer and PMDX-135. I tend towards the half-coil configuration and PK296A2A-SG7.2 motors.

Gerald and Mike thank you on behalf of humanity for his brilliant creation and wealth of knowledge.

My greatest respect,

Germán