[JVWilliams]

I've seen a few posts regarding the Wantai (Changzhou Wantai Electrical Appliance Co. Ltd.) DQ860MA stepper drivers, asking about wiring and such. They're part of some of the stepper motor kits Wantai sells on Ebay, and don't come with data sheets.

Some information I came across that is entirely in German can be found here:
http://www.joker-technik.de/media//D...3effffffef.pdf

I'm now the proud owner of three of these things, as part of a kit I bought from Wantai for a 3-axis mill build I'm working on. I've done a bit of research, and hopefully what I found will be helpful to others.

The DQ860MA may be rebranded/OEM devices manufactured by Leadshine Technology Ltd., also out of China (Leadshine Technology - Home). The PC board in the driver says "m860a", which isn't a model sold anywhere, from what I can find, though the numbering scheme is consistent with Leadshine's. Also, the DQ860MA seems to be sold through a number of outlets and under various brand names and even different model numbers.

Leadshine manufactures and sells a range of stepper drivers, including some with basically the same specs and appearance. The closest is the M880A:

http://www.leadshine.com/UploadFile/Down/M880Am.pdf

One difference between this datasheet and the information Wantai provides for this item is the value of the input resistor: the datasheet says 270 Ohm, and Wantai says both 270 Ohm and 200 Ohm. In fact, at least for my drivers, the input uses 220 Ohm resistors.

The Leadshine datasheet points out that there are jumpers that allow additional pulse settings on the internal driver circuit board. After opening the driver up, I can see they are really there, though I would guess most users probably won't need to change these, I'd imagine, for use with Mach3 or LinuxCNC.

The datasheet also provides good information on thermal considerations and mounting of the drivers, and information on protection functions.

Leadshine is a company that appears to have solid American connections, and indeed, looking at the driver's internals, they use quality parts by ST Microelectronics, Sharp, Fairchild Semiconductor and others. There's also at least one unmarked chip that looks like it's probably an ASIC. The printed circuit board is professionally made (2-layer, I believe) and coated, and the part placement and soldering is obviously done by machine.

On the other hand, the "JH" electrolytic capacitors used throughout (Jianghai, a Chinese brand) have a mixed track record, especially if exposed to excessive heat. In fairness, they're used in a lot of consumer electronics made elsewhere, not just in China (e.g. APC UPS's), and many brands of electrolytics have had problems over the last decade.

The input optocouplers are two types. The optocoupler on the "enable" input is a Sharp PC817. This is a relatively slow optocoupler (of the same variety as used in the Wantai break-out-board for all signal paths), and the "enable" input is obviously not designed for rapid input changes.

ftp://ftp.elektroda.net/pub/Karty%20...we/pc817xx.pdf

On the "pulse" and "direction" inputs, by contrast, the driver uses a Fairchild Semiconductor HCPL-2531 dual optocoupler device:

http://www.fairchildsemi.com/ds/6N/6N136.pdf

These parts are much faster, and allow the drivers to work with Wantai's specified pulse widths of 2.5 microseconds or longer (the M880A datasheet actually says 1.5 microsecond, which the optocouplers, at least, should typically support).

Wantai's bread-out-board, I suspect, and the slow optocouplers on it, is the cause of some difficulties users have had with Wantai kits--type 817's are simple devices that aren't designed to be fast, and the slow rise and fall times of the Everlight EL817's on the output of all signal paths would probably lead to smeared pulses and minimum pulse widths of 5-10 microseconds or longer. The good news is the output optocouplers on the BOB can be bypassed, as the DQ860MA already provides isolation from the motors with its input optocouplers. If the BOB optos are bypassed, though, it will result in inverted pulses going from the BOB to the driver which need to be taken into account.

Alternatively, a different break-out board that uses faster components would work, too, as long as power supply and signal voltage level recommendations are observed.

In summary, the drivers appear, overall, to be of good quality. The only potential issues I see, at least after a quick look-over, are the caps and whether enough thermal paste is used between the driver transistors and the heat sink.

I wasn't able to see the make and model of the driver transistors, nor did I map out the circuit schematic overall.

[MooseCat]

I'd be interested to know how those drivers work out for you. Wantai has a mixed reputation.

Leadshine make a m860, datasheet is here:
http://www.leadshine.com/UploadFile/Down/M860m.pdf

[JVWilliams]

Thanks, Moosecat. I saw the spec sheet for the m860m, but the current output settings diverged from those of the dq860ma more than the m880a.

I'll be hooking these things up for experimentation in the next couple of days.

[JVWilliams]

I've hooked up all three drivers to their stepper motors, and after solving a number of issues, have gotten them running.

By the way, Longs Motors (Google Maps has them located about a block away from Wantai in Changzhou City) sells the DM860A, with identical specs and case.

Some issues I had to address, that may trip up other people as well:

1. Not all parallel cables are the same. Some are crossover cables that reverse signal lines. That one took me a while to solve.
2. The "Enable" signal is ACTIVE LOW - which means the "Enable" input has to be off for the motors to run. The data on the Wantai website only says one of the "Enable" lines is typically left unconnected -- correct as far as it goes if you want to have the motors enabled, but the information is not very clear. That one took me a while to solve, too.
3. As I posted above, the minimum pulse width is a real issue for the Wantai break out boards. I was getting missed steps at less than 15 microsecond pulse widths. A new (hopefully better) bread-out board from CNC4PC is on the way -- yes, I was too lazy to do the soldering work on the Wantai board.
4. The microstep settings for the driver listed in the m880a datasheet I posted aren't quite right. The ones on the Wantai website are right. If Leadshine manufactures this piece for Wantai and Longs, then they have some custom programming.

WATCH THE CURRENT SETTINGS! Wantai definitely doesn't ship the drivers set up to work with any steppers that are included.

I am running these drivers set to 3.0 amps RMS (just a little more than the 2.8 amps RMS one would expect from the hybrid stepper's rating of 2.0 amps per phase, configured in bipolar parallel) with microstepping. The drivers came set to the maximum 5.6 amps RMS/7.8 amps peak.

[JVWilliams]

Just swapped in the CNC4PC model C1G break-out board. I dialed the pulse width all the way down to 1.5 microsecond. The steppers responded as they should -- no missed steps

Now, I need to swap out the steppers. I'll admit -- I bought this as a kit on Ebay, without doing due diligence, and the steppers have very high impedance (15 mH). By my calculations, that means I'd be clipping the microstepping motor-drive waveform at the +/- 48 Vdc power supply limits right around 150 RPM (using 3.0 amps). I pushed it a little, to 175 RPM, but definite problems with missed steps beyond that.

Wantai has another NEMA 34 stepper with 3.5 mH/phase and only a little loss in holding torque, and they might do an exchange for me. If the shipping ends up being a lot, though (the replacements will come from China), I might sell these steppers (basically brand new, plus now they're tested working) and buy a set of 620oz-in Nema 23's from Rex Machine/Modelshipcnc.com (1.6 mH). Still more than enough estimated torque for my mill build...

I'm building a separate unregulated power supply using a 20-amp, 48vct toroidal transformer, so I have substantial headroom for driving higher current/lower inductance steppers.

[microdot]

JV,
Thanks for all this. I recently bought a PS, DQ860MA Drive and NEMA34 stepper from Wantai. I'm going to be controlling them with an Arduino to make a 2ax rotary positioner for welding.

I have a question about setting the dip switches on the DQ860 to set the micro stepping. I have it set for 2000 pulses/rev (off, on on, off) which is 10x the 200 setps/rev of the 1.8deg motor. This seems to work however in my sketch for the arduino i have to have it pulse the step output pin 8000 times in order to get one complete revolution of the motor shaft. Other lower pulse/rev setting result in the motor freezing at higher speeds (lower delay between pulses).

My motor is spec'd at 860 oz/in (WT860STH) so I'm suspecting the inertia of the shaft and windings is part of the problem.

How do you have the dip switches configured on you drives?

Thanks