Gecko 214 Wiring Diagram To Acorn

[tuffduck]

Thanks for the input I will look at these.

Tblough - something for you to read and explain, thanks. If it is Part of the advertising that it is supported, when updates are made by 3rd Party companies up to date testing may be required.

[CNCMaryland]

Gecko makes many drives. I use a G540 without any issues.

[Richards]

I can see that this has been a frustrating journey. Sometimes things don't go as expected. When that happens, a DIYer has to put on his "engineer's cap" and figure things out. For instance, stepper motors must be commanded to move using step inputs to their stepper drivers. If a stepper motor makes random moves, the "engineer" gets out his oscilloscope and looks at the signals going to the inputs of his drives. If there are no pulses, then he knows that the problem is inside the stepper driver. If there are pulses, then he disconnects the drive from the Acorn and installs a resistor in place of the drive. If he sees pulses on the resistor circuit, he knows that the step pulses are somehow coming from the Acorn. If there are no pulses, then he know that something else is generating those pulses, so he looks for sources of electrical noise.

Finding the source of unwanted step pulses can be a tedious procedure, but if the procedure is methodically followed, the source of unwanted pulses will be found. Once the source is found, then the "engineer" will install whatever needs to be installed to eliminate those unwanted pulses.

Not everyone has a 'scope, but everyone has friends. Certainly one of those friends has a friend with a 'scope who would be willing to help.

It would be impossible for Centroid to pretest every possible stepper driver. Fortunately stepper drivers are similar and mostly expect the same kind of inputs. GeckoDrive drivers are an exception. The G201 style drives expect the pulse generator to sink current. Those drives use +5V as their common. The G203v drives expect the pulse generator to source current. Those drives us GND as their common. The DIYer has to know what type of signal his Geckos require. I've used G201/G212, a G540, several G203v, and some G251 drives from Gecko. I've used some very old drivers from Oriental Motor. I've used Leadshine style drivers. Everything worked for me right out of the box. That doesn't mean that everything would work for you right out of the box. Stuff happens. I live in a desert, so I always take special precautions to eliminate the possibility of static even to the extent that I keep the hair on my mostly bald head short ;>.

Even with all the precautions that I take, I once fried an Acorn board because of a freak accident (which I wrote about on the forum). In that case, I opted to use Centroid's board replacement service and got a replacement Acorn for about $200.

I'll admit that my background differs from most builders. I've designed and built a lot of process controllers when I had to design things from the chips to the end product. I used a 'scope when bread-boarding the chips. I used the 'scope after designing the circuit boards and prototyping finished boards. I used the 'scope after assembling the board(s) into the process controller. I also used the 'scope when I first received an Acorn. It became evident that a 'scope wasn't needed. The Acorn is very well designed and very robust; but, even though it is well designed and robust, it is only a small part of a CNC controller. It is the central item, but everything added to the system has to work together properly. That's when being a DIYer can be challenging.

[tuffduck]

Hi Mike,
Thank you. I have never used an oscilloscope before. I have no idea even where to buy one, the type I need or even where to start.

[drdennis]

@tuffduck
The kind of a problem you have really requires EE expertise to address. An EE would likely use a scope as one of the tools to diagnose the problem, but the scope would be just that -- one of the tools. A scope in hands of someone that has no experience using it is of very little use. Too many variables, too may gotchas (e.g., a probe ground clip is connected to earth, so vaporizing a wire in high-power application (when not careful) is not uncommon), and too many knobs. So no, acquiring a scope is not a good option for you.
As MIke suggested, you really have only two options:
1. find an EE to help you with the problem
2. alternately and painfully, you can try having us walk you through the requisite debug, step by step (as I have tried).
#1 is definitely a better option, if possible, but if your location prevents that, then #2 is your only choice, painfully slow and requiring a lot of patience and discipline.
Cheers

[Richards]

I think that Dr. Dennis made things look a little bleak; but, as he said, a scope is a tool.

When looking at a step signal, a scope is easy to use. Attach an isolated ground plug on the scope's AC plug to isolate GND from Earth. Attach the scope's GND input to DC ground. Attach a signal input to the step signal. Adjust the input voltage range so that you can see both the top and the bottom of the signal that you're watching. Set the sweep time so that you can see several stepper pulse at one time. Use the Acorn's MDI tool to run a stepper at a low to moderate speed for several seconds. That's about all you have to do. If the steps are evenly spaced and are uniform in size and shape, you're getting a good step pulse from the Acorn. If you see extra signals or distorted pulses, it's time to dig deeper. In any case, find a friend who has had a little experience with a 'scope and watch over his shoulder for a few minutes. Everything seems difficult until you know what to do. Everything becomes easy after a little practice.

I don't necessarily recommend buying a scope for a one-time project. It would be cheaper (perhaps free) to ask someone in your area to help. If you're building something out-of-the-ordinary, a scope could easily be as necessary as a multi-meter. If using a scope seems intimidating, watch a few YouTube oscilloscope tutorials. See if a scope makes sense for you. Once you've used a scope, you'll use it every time you have an unexpected problem. It's just a tool, but a very useful tool once you understand the basics.

In reviewing your problems with stepper motors and stepper drivers, using a scope for five minutes would have shown whether you had an Acorn problem, a stepper driver problem, or an electrical noise problem.

[drdennis]

@Mike
I generally agree with what you said. Except...
If my boss asked me to work on this problem, with the information I have collected thus far, I would not commit to have a solution in five minutes.
I would say: "Give me five minutes to scope it, and I will tell you more. Something like this could take an hour, or multiple days if we are not lucky".

Indeed I made things look a bit bleak. For a reason. We are not looking at an obvious problem like miswired circuit. The description of this problem is that one motor at a time works fine, but things go south when multiple motors are turned on. I am an EE with more decades of experience than I want to admit and I love troubleshooting difficult problems. Still I would call this a potentially very hard problem. Anything from fried drivers (easy) to crosstalk or dirty power (hard).
I could be lucky and see marginal signals (good enough for one but not for more motors) on the scope, and take it from there. But I would not be surprised if upon casual inspection those waveforms looked "good". One problem I can think of is a cold solder joint in a power supply, or in the Acorn output stage. May work fine with one motor, but with more motors running at the same time it may occasionally (the extraneous movement is slow) produce an extra pulse (maybe every time the driver pulses coincide). Now in order to "see" that, you have to take a multi-second trace and examine it looking for ~100ns glitches -- so a digital oscilloscope (likely one better than what you find in most DIY shops) may be required (plus expertise setting everything up).
You may need to examine your DC voltage for glitches or other junk. Again, a scope can help, if you know what to do.

I would not want to discourage anyone from learning how to use a new tool. On the contrary -- and I will help if I can.
I am just being pragmatic, based on my experience with junior EEs and interns. A problem like this would likely be beyond their reach.

@tuffduck
Q1 Have you tried replacing the power supply that powers the motors (drives)?
Q2 Does the extraneous motion you observed require positive torque? Do we know that the extraneous motion is not due to losing the holding torque? (I don't know if your axes can be back-driven.)
Q3 Have you tried running the three motors on the bench (unloaded)?

Cheers

[Richards]

Dr. Dennis, An electrical engineer knows that there is a cause to every effect. An electrical engineer looks for the cause when he has found the effect. An electrical engineer uses the necessary tool to show the cause. Nothing happens electrically unless it is caused. My mentor had a way to simplify every problem. When I told him I had a hard problem to solve, he would ask basic questions about voltage, current, wave forms. He would ask what was causing what. Until I could tell him what was causing what, we didn't go to the next step.

A stepper motor moves only when commanded to move, unless the motor is defective. If it moves when removed from its driver, it is defective, otherwise it is being commanded to move. If it does not move by itself but moves when it is connected to the driver and no steps are being fed to the driver, then the driver is defective. If it moves properly when the driver is receiving steps from the Acorn and improperly when connected to the Acorn and the Acorn is not sending steps, then noise of some sort is causing the problem. That analysis took less than five minutes to write. It would have taken less than five minutes to scope out. A detective looks only at the clues at the scene. He leaves speculation to people who cannot see what he sees. This is a simple cause and effect situation. Start with the effect and then find the cause, step by step.

[martyscncgarage]

Gentlemen,
I certainly appreciate your expertise and knowledge!
I think this is over the original posters experience and it sounds like he makes money with his machine and just needs it to work.
He's already sent the drives back to Geckodrive.

There is another user on this forum successfully using 3 Geckodrive GR214V drives on his milling machine.
I have tested two versions of the drive on the bench with Acorn with no issues.

Usually it is mismatched components, power supplies, or not setting the current limiting correctly on the drives for the stepper motors (including the way they are wired). I believe he did not have specifications on two of steppers and they seemed to me to be sub par. I would have replaced the motors from a reputable supplier, properly sized to his machine, properly chosen for the given power supply and I would suggest against using cheap no name brand switching power supplies. I think a single good unregulated power supply such as those build and sold by www.antekinc.com would be better.
I have followed Geckodrive's white paper on Stepper Motor Basics https://www.geckodrive.com/support/step ... asics.html , as soon as I did, my stepper based control systems became immensely better. I have spoken with Maris Fremanis the founder of Geckodrive and he truly is a very knowledgeable Engineer.

I have never experience an issue where one motor and drive work, but adding 2 more make the whole system unstable. We can rule out Acorn for the most part because his other drives run fine. As previously mentioned, I suspect the stepper motors themselves.

Just my .02 on the subject.
I think we should just let this one die off as TuffDuck has moved on.
Marty