@Jim_Fong thanks! I will!
Mpja sells power supplies at reasonable prices.
@Dale_Royer Thank you for clarifying that for me! Not that Iād be running total full power to just one motor except maybe for a few rapids
@George_Allen I think 48V may be too high for TB6600 drives? I run mine at 36V. Realize that while you run stepper motor drives the voltage in the system will be higher than the voltage you input. Because of magic. Well thereās a scientific reason for it, but Iām not in the mood to explain it right now so just realize that voltage rises in induction PWM circuits. Also you should measure your actual current draw. It is probably less than youāre thinking it is. Thatās also because of induction, or inductive reluctance.
Over volting stepper motor drives is another excellent way to burn them out. Your Geckos can handle 48V though. I think Geckos are rated up to 80V? To do that Mariss builds his drives with parts that can handle 160V. Because Mariss knows a lot about stepper drives. He knows about the voltage rise and what have you.
Yes, I think Iām going to use my meanwell PSU For the tb6600. Yes, the gecko is rated up to 80v, but I only have one of those. Iāll do some analysis. I think you were right about the cause for the problem on my other tb6600. The terminals werenāt entirely screwed in like I thought it was and the gauge of wire was too thick for the terminal and when I adjusted it it might have come disconnected while the motor was energized.
@George_Allen Yeah it just takes a bad connection and a momentary interruption to toast stepper motor drives. You donāt really need to use wire any thicker than the wire that comes out of your stepper motors. Thatās usually about 18 gauge or smaller. Iāve seen people use ethernet cable for stepper motors. That might be a bit chintzy but it seems to work for folks. I stripped a bunch of IEC power cords and salvaged the wire out of them for my motors. I had a pile of those power cords Iād collected over the years. More than Iād ever use. I still have a little basket full of those things. But my collection is more manageable today.
Hereās a picture of me making up my 4 wire stepper cables http://i.imgur.com/KKL8nGG.jpg
Yeah your power supply will be plenty. Everyone oversizes them by adding the stepper currents together. It does not work like that.
@Paul_Shaw better too big than too little but most people are not drawing the current they think they are.
long form replyā¦
āIf Iām running my drivers at 3A RMS, 4.2A peak, is that too much for my stepper motors? They are supposed to be rated for 3A (57bygh115-003 Nema 23ās 425 Oz/In (3Nm)) on a 36V 350Watt power supplyā
I think you (and others) are looking at it wrong.
Firstly, you arenāt ārunningā your drivers at 3A RMS - they are capable of handling that level of current.
Your driver can DELIVER 4.2A peak - in other words, between 0A and 4.2A it wonāt burn out. It cant sustain 4.2A, but will sustain 3A for as long as it remains within the temperature tolerances TI have tested it at. A driver without sufficient heatsink will overheat and fail before a driver with sufficient heatsink, as it canāt dissipate the excess heat. āFailā doesnāt necessarily mean āmagic smokeā - they are designed to shut down before then to preserve themselves. magic smoke is usually a function of too much voltage, as they nearly all have thermal shutoffs.
Your stepper motors will DRAW 3A running at 36V, necessitating a 350W 36V power supply to deliver that amount of current.
Your drivers arenāt pushing - theyāre supplying a need (and the first taste is free 
). If the stepper requires more power than the PSU can supply, they will just not get what they could, in theory, take and will thus not be able to operate at their full potential. They will operate at their maximum for the amount of volts and current you can supply
Where things get burny and broken is where thereās a mismatch: you have a 48V supply to a driver rated for 36V. You have a supply rated to 50A when your driver can handle only 4.2A at most, and for a short time.
In those cases, the driver will break down through too high a voltage or overheat quickly and shut down. Your stepper motor will take all the power it can get and if the work you are asking it to do lifts the current over 4.2A, the driver canāt cope and it overheats and shuts down, or magic smoke occurs
The rating of a stepper is a known specification at a specific set of values - in your case, at 36V it will draw a max of 3A while delivering 3NM of force. But, the windings and circuitry in the stepper can probably take a whole lot more volts and amps before you get internal arcing and it melts.
So, from the perspective of what goes where, you size your steppers to give you the motive power you need to move and hold your gantries to do the work you need doing.
If you have a lightweight diode laser on a lightweight aluminium frame, you donāt need a lot of torque or holding power. Thereās no pushback from the tool (the laser) and the inertia of the gantries is low. A NEMA17 with <1 NM will probably be more than adequate, unless you need higher speed than they can deliver. So you would specify a higher rating of NEMA17, or go to the next size up - a NEMA23.
If your machine is a water jet cutter with a heavy steel gantry on rack and pinion drive system, you may need 10NM to reach the speed the water jet tops out at - for example, your waterjet can cut 25mm stainless steel plate at 100mm/sec, it has a gantry weighing 200kg with an x-Axis weighing 30kg, with associated pipes and cables adding another 15kg of drag. but you are cutting mostly 50mm plate, which means your cutter needs to operate at a maximum of 25mm/sec you donāt need to move your gantry as fast, which means you could specify a smaller mechanical setup.
Obviously you want your water cutter to be able to operate as fast as possible cutting whatever you may throw at it, so you spec it to be able to handle 200mm/sec. Better to have too much power than not enough.
Your steppers can hold back torque of 427oz/in when operating at 36V and will consume 3A of DC while doing so. If you could only supply 2A of 36V DC, they may only be able to hold back 350oz/in of torque. If you could supply 3A but at 24V, you might only be able to hold 300oz/in. If your drive system is coarse and the turning moment lever arm is 2in, then it will only hold back 208.5oz in real force applied. If your pulley is only .5", it could hold 854oz of force. If you have an 8-turn/in screw, you can hold 3416oz of force.
Itās all relative and every component in your power train has an effect.
You have built your machine with significant (for aDIY CNC router) power. 3NM is, given gearing % increases in torque, going to have no problem moving an alloy gantry on your average home CNC machine. but will struggle with a 50kg steel gantry with significant friction. The steppers on the commercial 8x4 table I occasionally use are 4.8NM units at 50V. My Ox CNC at home uses the same spec as yours - 3NM at 36V, but my supply is 24V. It still moves frighteningly fast - faster than I would ever need in actual cutting. As Iām using a 3GT belt system, the speed is turned way down in software as it has enough power to stretch a belt when accelerating from a stop.
As long as your PSU can deliver enough current at the right voltage, and your drivers can take the current you require and the voltage you are supplying, you should be fine. Pay attention to heat - put heatsinks on the driver chips, even if the manufacturer says they donāt need them. I use a TinyG on my Ox and have the heatsinks on the back-side of the board, not the top of the DRV8825 chip, because the variant of chip used on the TinyG has the heat pad on the base, mated with solder to the heavy copper base layer of the PCB. If you use discrete stepper drivers (like the Longshines, Gecko, etc.) they will already have their cooling requirements met, but you might want to ventilate your electrical cabinet with a couple of big PC case fans.
Sorry to ramble on so much on not-necessarily-related subjects, but there are so many contributing factors when putting bits together to make a device.
Feel free to ask for further clarification or to point out if/where I blundered - it wouldnāt be the first time 
1 Like
@Paul_Frederick different models of TB6600 are rated at different maximums. The TB660HG, for example, is rated at 5A peak/4.5A nominal at 50V. It can dissipate 40W
Itās a Zip25 package with significant area for a heatsink, not an SMD chip.
The TB6600FG, which is an SMD package, can take 4A at 42V and can dissipate 4.2W max - only a tenth of that of the HG
Heat dissipation matters. A lot.
That is a great explanation and teaching, thank you! Yes, my drivers are dq543maās with the built in heat sinks and Iāve got a case fan blowing on them at all timesā¦ bit of overkill, probably, but canāt hurt keeping them a little cooler
You can never have too much money, too much power or too much cooling 
And yes, probably way more stepper and driver than you NEED, but for your next machine may be just right :
I am on the equator and my workshop is 35C before I switch anything on. Airflow is imperative to keep things running along.
Tim Allen said it bestā¦
missing/deleted image from Google+
@Robert_Hogstedt Yeah I have a little box fan blowing over my motor drives. Better too cool than too hot when it comes to electronics.
@Mike_Thornbury being as Iāve made a number of stepper motor drives with several of them using ZIP IC packages like the SLA7026M and the TB6560AHQ I am familiar with the TDP of them too. I even had to make my own library packages for the schematic capture software that I used to design the boards I made. Because back then those oddball packages had no available libraries for them. So I had to make my own footprints and everything. I think I drew this in the late 90s? It was a long time ago now. http://i.imgur.com/a9LoMOB.png
