Repeated loss of Gecko 202 Drive...HELP

[jdw]

Its deer season here but soon it will be open season on my SB if I can't get this settled!

Agek control box
PRT w/standard motors
latest control software

Intermittent loss of drive...today I was watching when it lost X2...when I reached the box the drive power light was on, drive was cool, mounting plate was cool.

I could spin X2 moter by hand...power off/on and no longer able to spin motor.

This situation was repeatedly occurring on another drive which I removed. It was hooked to Y.

I get 74 volt DC on drive when idle, 71 when air cutting a file. Did not do this for more than a few minutes. I have a good Fluke meter.

I just received the control box back from SB. Thier verbal report was everything checked out fine. However, when returned there was no diagnostic report/incident report...nothing. I find that strange. Thier conclusion was drive was overheating. To address that I;
1 Cut out the back of the control box so the AL mounting plate was exposed.
2 Now operate with covers off.
3 Removed front plastic of box and now blowing a house fan over the entire unit.
4 Remounted all drives with heat sink compound.
5 Checked and tightened all connections. Replaced suspicious wago connectors.

I'm left with with what Mike Richards posted recently regarding a capaitor on the distribution side of the power supply. But if this is a big problem why don't we here more about it?

I've spoke with a few of you off grid and I do appreciate all your input. I'm now hoping the collective effort may have some new insights.

I did not expect ShopBot to work on my Agek box for free. I was very greatful for their rapid response and quick turn around, but after $200 in shipping I've gained nothing, have no report of what was tested so no baseline to work from. I'm wondering what the policy/relationship is between ShopBot and the now defunct BotRods, maker of Agek.

I'm all very frustrated ears?

And all this is happening in the middle of a large job!

TIA,
Jon

[Brady Watson]

Jon,
My suggestion would be to buy 1 or 2 spare Gecko 202s and keep them on the shelf as spares. In my experience, what you are going through is not all that uncommon with Geckos. I've had 2 of them do the exact same thing that you are describing, and I have even sent one of them back to Gecko for evaluation/repair & they couldn't find anything wrong with them. So...since they are so cheap, buy a couple spares. Otherwise it's a case of 10 cents holding up a dollar when you are relying on them to make you money.

-Brady

[jdw]

Hey Brady...thanks for your input. I'm all about having spare parts..in fact I saved my original control box in the event I needed a back-up. However, is there no way to be sure that it is just the driver?? Or is that the nature of the beast we're dealing with? That would mean out of 5 drives on this new board, 2 are bad...seems like fairly high odds. Anyone know which is the more prudent way to purchase these drives...direct from Gecko or from Shopbot? How about the new 203V?? Any reason to go with those drives instead of the 202? Thanks for the help!
Jon

[richards]

Stepper motors, Gecko stepper drives and switching power supplies should never be mentioned in the same sentence. DO NOT expect a switching power supply to work with a stepper driver. The load that a switching power supply supplies, by definition, is constant. The load required by a stepper driver and a stepper motor is not constant. A switching power supply tries to self-regulate both its voltage and its current, but the time lapse built into a switching power supply makes it TOTALLY UNSUITABLE as a power source for a stepper motor UNLESS a LARGE FILTER CAPACITOR is added.

Personally, I'm getting a little frustrated when I continue to hear horror stories about Gecko stepper drivers. Mariss has told us how to figure how much capacitance should be used with his stepper drivers. Those who ignore his specifications are going to have trouble!

Regardless of the power supply that you use, if you use a Gecko stepper driver, add a filter capacitor to the power supply. Don't be stingy with the capacitors. I use a minimum of 15,000uf and have NEVER had a problem with more than twenty-five gecko stepper drivers. BUT, I have simulated the exact problem that has been discussed on this forum by eliminating the filter capacitor and trying to run the stepper drives/stepper motors directly from a switching power supply.

Bottom line: If you want to use Gecko stepper drivers, do yourself a favor and memorize the white paper that Mariss publishes "Stepper Motor Basics" that is available for download at www.geckodrive.com (http://www.geckodrive.com). Then, buy or build a simple linear power supply that consists of a toroidal transformer, a bridge rectifier and a large filter capacitor. Nothing could be simpler. What Mariss specifies works. It is inexpensive. It is reliable. It keeps the G201 the G202 and the G203 from shutting down unexpectedly.

There is no need to re-invent stepper technology. There is no need to imply that Mariss is putting out an inferior product. Follow his recommendations and enjoy the fruit of his labor. I have personally logged more than four-thousand hours running stepper and servo motors using his stepper drives, linear power supplies, and both Mach 3 and Shopbot SB3 software without a single problem. On the other hand, I have re-created some of the problems that have been discussed on this forum by ignoring Mariss's recomendations. Do what you want, but don't blame Mariss if something doesn't work - unless you follow ALL of his specifications to the letter.

Edited: Jon, you posted while I was composing. I have used both the G202 and the G203 stepper drives. I will use the G203 on all new projects because it incorporates some niffty electronics that are not available on the G202; however, when properly used, both the G202 and the G203v will drive the same motor to the same performance level using the same power supply and the same pulse generator. Of cource, the G202 requires a pulse source with a common 5VDC and the G203v requires a pulse source with a common ground, Other than that, you can use either stepper driver with complete confidence.

[richards]

I've already received a few emails asking me to clarify what I wrote about switching power supplies NOT being suitable for stepper motors. So, let's look at how some power supply fundamentals:

LINEAR POWER SUPPLIES
First, the power that we get from the power company is AC, or alternating current. Depending on where we live, the voltage is 120VAC or about 220VAC to 240VAC. A stepper motor requires a DC volatge much less than 120V. So, given those facts, what do we do? The most simple solution is to use a transformer to convert the 120VAC (or 220VAC) power coming from the power company to a level that we need. Lets say that we need about 45VDC to 50VDC to drive our stepper motor. The first thing we do is to find a 35V transformer so that whatever voltage we feed the transformer is converted to 35VAC on the output of the transformer. Many of you might well be saying, "Hey, what's going on? We need 50VDC, not 35VDC." You have to understand that almost all voltage meters read the voltage using ROOT MEAN SQUARE, not peak to peak. So, if you're one of the elite and are privy to the mysterious electrical formulas that we geeks find entertaining, you'll know that to convert RMS voltages to peak-to-peak voltages requires that you multiply the RMS voltage by 1.41 to get the peak-to-peak voltage. That means that 35VAC RMS is really about 50VAC peak-to-peak. (You can trust me on this. I'm not trying to pull the wool over your eyes.)

Second, the power coming out of the transformer is still A.C. or alternating current. We need D.C. or direct current. A 35VAC transtormer's output is +50V and then -50V fifty or sixty times a second, depending on whether your power company gives you 60hZ or 50hZ power. However, if you use a simple $3.00 bridge rectifier, which is simply four diodes oriented correctly, you will end up with either +50V OR -50V at 2X the normal line frequency. We'll assume that you want +50V and that you've connected the bridge rectifier to give you +50V.

Third, the output of the transformer and the bridge rectifier is still not pure D.C. voltage. On an oscilloscope, it looks like a line of hills instead of a constant +50V plane. To fix that, we just add a large filter capacitor. The capacitor acts like a storage reservoir that gets voltage ranging from 0V to 50V and outputs a stream of voltage that is about 50V. Well, using the analogy of a reservoir, it is easy to see that if the reservoir is large enough and if the steams feeding the reservoir carry enough water, then the steam exiting the reservoir will be able to carry a constant flow of water. However, if the streams feeding the resevoir carry too little water, or if the resevoir is too small, then the output of the stream exiting the resevoir will fluctuate. That is exactly the principle used by a filter capacitor. The capacitor is the resevoir. The transformer feeds electricity into the capacitor. Whatever is connected to the output of the capacitor draws electricity from the capacitor. If things are matched up properly, then whatever is drawing electricity from the capacitor gets all the electricity it needs at the level (voltage) that it needs.

SWITCHING POWER SUPPLIES
A switching power supply also converts standard voltage power to a desired voltage level. How it does that is different from the way that a linear power supply handles the job. Not all switching power supplies work exactly the same, but many of them do two things different from a linear power supply. 1). They boost the voltage to a much higher level. 2). They multiply the frequency to a much higher level.

By boosting the voltage to a higher level and increasing the frequency to a higher level, the filter capacitors on a switching power supply can be much smaller than the filter capacitors on a linear power supply. You don't need to know why, just trust me that it does work. (For the skeptical, just do a Google search on 'switching power supply theory' and then spend a few hours reading about switching power supplies.) The main thing that you need to know is that a switching power supply normally boosts frequency from 50hZ or 60hZ to 15,000hZ (or higher) so that we don't hear the 'whine' or 'squeal' that is normally associated with switching power supplies. The second thing that you need to know about a switching power supply is that the power supply incorporates feed-back sensors to know how well it is doing its job. On most switching power supplies, if the voltage or the current starts to droop, the power supply increases the frequency so that the expected voltage is delivered, and then uses its electrical magic to deliver the expected voltage or current. Well, that is the problem. During the transition period, the switching power supply might deliver a lot of garbage on the power line as it tries to stabilize the output voltage (current). Don't take my word for it. You can hook up almost any oscilloscope to the outputs of a switching power supply and watch the garbage for yourself. Now, if the garbage level gets out of control, a sophisticated device like a Gecko G202/G203v stepper driver goes into self-protection mode and shuts itself off.

That's what happens. There is no mystery. There is no failure. The device does exactly what it is supposed to do to protect itself. To keep the garbage from shutting down the Gecko, add a large filter capacitor as a resevoir and the stepper driver will be able to do its job without thinking that the sky is failing and that it needs to shut down and run for cover.

Edited: For the purests, I'm well aware of the fact that a bridge rectifier drops a little voltage across the diodes - meaning that a 35VAC power supply may produce a voltage closer to 45.35265VDC than to 50VDC. The fortunate thing is that a stepper motor really doesn't care very much. As long as the voltage is not too HIGH, the stepper motor is happy and will do its job - albeit, just a little bit slower than it would if the voltage were exactly right.

[Brady Watson]

I've had 2 out of 6 new drives flake out. I mean it's not like we are talking a lot of money here...but then again, it makes no sense to waste it either. Replace with the same thing - G202. There are some things with the G203 that are undesireable, and they are the wrong common polarity for your application.

I've never used a switching power supply with a Gecko. (I take that back, I have a 4G with original PRT switching PS and it is set to 48v...and I've never had a problem with it or it's Geckos) I use toroid power supplies with capacitors made for the application. Either way, Geckos do sometimes flake out & it doesn't hurt to have a few spares on hand. You can buy spare Geckos directly from Geckodrive or ShopBot 6 of one and half dozen of the other...Geckos either work or they don't in my experience, the 1st week of operation. I've never had the condition where they work & just stop all of a sudden. I have also heard from others who use Geckos and they agree that they either work or they don't. Not to take anything away from GeckoDrive, they have very good customer service. You may wish to send your drives back after you receive your spares and see if they can find any problems with them.

-B

[jdw]

Mike...Thanks for your input. If I've said something that implies I consider anything from Gecko or Shopbot inferior, I sincerely apologise...it was not my intent. My intent is to make money with my SB which currently employs Gecko drives. I am not a tinkerer, wouldn't mind being one but for now I do not have the time nor money to be one. I simply want this thing to work properly.

I followed Mike's advise and read Mariss's white paper. Most of it is over my head. What I could gather that is prudent to my situation is the possible lack of capacitance somewhere. I don't understand the difference between the local capacitor attached directly to power input of the drive and the capacitor Mariss talks about when discussing regulated, unregulated, or switching power supply.

How do I even tell what type of power supply I have? My huge disadvantage is my ignorance on the subject. I look at my control box and I can't even name the parts..On a good day I can tell the difference from a capacitor and a rectifier...zener diode, forget it! Combined with my ignorance is no schematic of the Agek box...I might have sufficient capacitance, but I have no way to confirm.

If anyone else has an Agek box and is able to shed some light here I would be greatful.

I'm still open to this whole situation being something simple such as my input computer or some "electrical noise" I'm totally unaware of. However, how would any of these situations cause one drive to go into reset mode?

I did install my original control box with latest software and firmware for said box....I forgot how good I have it when the Agek is working properly!!! And I repeatedly have false proxy switch sensing stopping the machine cold in the middle of cutting. I suspect this is software related and not part of some larger problem that also affects the Agek box. Never had that problem with 3.24 software.

As always thanks for any and all input!

Jon

[richards]

Jon,
Sorry, but nothing was directed at you (or at any other specific person). It was just a shot over the bow, so to speak, against the idea that a Gecko driven stepper is inferior to any other stepper motor/stepper drive configuration. When properly used, a Gecko driven stepper motor gives results equal to or better than any other stepper drive/stepper motor normally available. (That will have to be the text of another post, perhaps tomorrow, when I have time to defend that statement.)

To tell what kind of power supply you have, just follow the leads back from the Gecko's #1 and #2 terminals. If they are connected to a rectagular metal box, chances are that you have a switching power supply. On the other hand, if you can clearly see a transformer (either round toroidal shape or traditional), a bridge rectifier (usually looks like squarish component about 1-1/8 inch square) and a large can-type capacitor, then most likely you have a linear power supply. Even if you have a switching power supply, a large filter capacitor will keep the 'garbage' from affecting the stepper driver. So, if you don't see a large filter capacitor between the stepper drivers and the power supply, just add one.

Mariss also specified that a G201 stepper driver should have a smaller 470uF capacitor screwed directly into terminals #1 and #2. The G202 already has that capacitor built into the drive. The G203v was designed to not require that capacitor, but adding it can be thought of as a safety measure - not required, but still helpful.

One of the problems with stepper drivers and variable frequency drives for spindles is that power is amplified with large transistors (or transitor-like devices). Those large transistor-like devices can only handle a finite amount of heat. In a stepper driver, forces are working that can cause heat to rise to the danger level in a very short amount of time. To keep the transitor-like devices from overheating, the drive has electronics built in that immediately shuts off the drive. In a well designed system, false triggering never happens.

What I'm trying to say, is that the end-user should never experience a problem with a stepper drive shutting down unexpectedly. If a drive does shut down, it should be only because an error that is fatal to the stepper motor or to the drive has occured (such as wire insulation rubbing away - or other mechanical problem). In that kind of situation, if an error occurs, you would turn off your machine and keep it turned off until you had discovered the short circuit. But, in your case, when the machine shuts down randomly, for no apparent reason, then the only absolute result is frustration.

[richards]

Brady,

I missed your post. Sorry.

You and I have a point of disagreement. Maybe your sample size is much larger than mine. Maybe I've just been lucky. I've even got four of the infamous G202 drives with the blue connectors - which were known to be unreliable. (The connector had too little spring in its contacts resulting in intermittent failures if the connector was removed and then reinserted. That problem was caused by a supplier changing materials. Mariss switched suppliers and send replacement connectors to anyone who had - or thought they had - a problem.)

One thing that I have done is to torture those drives in every way known to man including high voltage, high temperature, high current. No matter what I've done, the drives continued to work as expected.

I prefer the G203v drive. That drive has green, yellow and red LEDs to show its state at a glance. Because of its circuitry, it runs much cooler than the G202 or the G201. It uses a different reduced current mode that gives a motor much higher holding torque (when in standby) than the G202. However, it was not designed to work reliably with a motor that has high impedance, such as the PK296A1A-SG3.6 when wired Bipolar Series. But that same drive and that same motor gives spectacular results when the motor is wired half-coil. (As a side note, if anyone wants to use the G203v with the PK296A1A-SG3.6 motor and wants to use Bipolar Series wiring, a resistor change on the drive will make the drive compatible with that impedance range. I don't like to mess with surface mount components, so that's not an option that I would consider.) Also, as you've also stated, step and direction signals are Active High with the G203v drive and Active Low with the G202 drive, so the drives are not interchangeable without modification of the step/direction polarity.

Please realize that I believe what you're saying, but I just can't duplicate the problem - no matter how hard I try. In telephone conversations with Mariss, I know that he is very concerned when anyone has a problem with any of his products. He goes to extensive lengths to find the source of any problem. When someone finds a reproducible problem, Mariss freely and clearly admits his fault - and then does something to keep that kind of error from happening again. Last March I spent a very enjoyable thirty minutes at his home in the L.A. area visiting with his wife and one of his employees. When I asked them what his greatest frustration was, they basically said non-reproducible problems. He's the only person in the electronics field that I know that has a "stuff happens" policy. If someone ruins a drive for any reason - and openly and freely admits what they did to destroy the drive - he will replace the drive at no cost to the customer once.

[br928]

Jon,

I have had a Agek on my PRT since they first came out (pre 4G). I have had a G202 stop working on one of my X motors once in all that time. That was after a 3 hour 3D carving where the x was running constantly. I still do not know if it overheated or stopped for another reason. With the heatsink that the gecko is mounted on (with the Agek) I doubt if it overheated. I changed the drive out with a spare and have not had the problem since. My suggestion would be to replace the suspect drives with new G202's. Double check all wire connections with the gecko terminals and Agek terminals for loose or poor terminations. This could cause unexpected problems.

Like other have said, the G203's are not compatible. Make sure the resistor that sets the motor current is in place. I run mine with the auto current reduction (internal dip switch setting) although some botters might disagree. My motors stay cool like they used to with the old contol box. I have not noticed any negative results by running this way.

I hope this helps. I have be pleased with my Agek.

[Brady Watson]

Mike,
I am aware of the blue/black connector issues and loose connections on the G202s. It was not a connector issue. The G203 has internal software shifting, which in it's generic form, performs worse than a G202 using the SB controller. It chokes where the control shifts from microstepping to full stepping (somewhere in the transition). Aside from this, the G203 is the opposite common, so it's not plug & play. I know Gecko's policies & they are generous. I am in no way flaming Gecko - their products make me a living. Just advising what I would do, as I am a cut the non-sense kinda guy that likes to just get to work rather than monkey around with the electronics (although I am fully capable, it isn't my bliss) I can only speak from my experience...your results may vary.

-B

[br928]

Jon,

How long have you been running the Agek? What is the history of the problem?

[richards]

Brady,
I think that you've nailed the point of concern when you said, "I am a cut the non-sense kinda guy that likes to just get to work rather than monkey around with the electronics . . . " That seems to be the real point to Jon's posts and the posts of others who have similar problems. Your suggestion to keep a spare drive on hand is simple, practical and realistic. Keeping a machine in production is more important than worrying about a $125 stepper driver; but, at some point someone has to identify the problem and then find a solution to that problem. When one person has an intermittent problem, everyone feels empathy and hopes that a solution is quickly found. When two or more separate people have very similar intermittent problems, the situation starts to become acute. Somehow, someone involved in design has got to duplicate the problem and then find a long term solution.

The morphing problem that you wrote about seems to depend on the stepper motor. I mostly run Oriental Motor PK296B2A-SG3.6 motors with my G203v stepper drivers. The motors are wired to the drivers using the half-coil wiring method. They don't exhibit a morphing problem. Unfortunately, I don't own a PK296A1A-SG3.6 motor, so I can't thoroughly test that motor. I've borrowed one of the A1A models when I heard about the inductance problem - so that I could verify that the problem was a general problem. With the steppers that I use, the harmonic imbalance region of the motor is in exactly the same region whether I'm running G202 drivers or G203 drivers. (All of the stepper motors that I own have harmonic problems at certain speeds. That can be a problem with any CNC router running multiple axes because, invariably, one axis will end up running at the problem speed. Most torque/speed charts clearly show the range of problem speeds.)

Anyway, I've copied three posts from the Gecko forum on Yahoo groups where jeetendra asks a question about morphing, Les answers the question and Mariss verifies that the answer is correct as long the number of microsteps is changed from 8 to 10:


----------------------------

Re: Stepper servo update 10/15/07

Correct. Small point: it's 10 microsteps / step, not 8.

Mariss


--- In geckodrive@yahoogroups.com (mailto:geckodrive@yahoogroups.com), Leslie Newell <lesnewell@.. (mailto:lesnewell@.).> wrote:
>
> All that happens is that the sinusoidal microstepping waveform is
> squared up to increase the on time and therefore improve high speed
> torque. At low speeds each of the 8 microsteps is a different voltage.
> At high speed they are all the same. This is all internal and does not
> affect the number of pulses needed to move a given distance.
>
> Les
>
> jeetendra_g10 wrote:
> > when the 203v morphs from microstepping to full stepping,how are the
> > step pulses counted?
> > do the 203v ignore 8 microsteps at high speed,or does the g203v
> > actually move 1.8deg per step?

[bcammack]

I worked as a bench technician at L.H. Research, the company that created the first commercially viable high-current switching power supply back in the mid-70s, so I know a little about them.

Switching power supplies utilize a constant frequency around 20,000 cycles per second (20 kilohertz) This keeps the resonance of the transformers above the range of normal hearing. Voltage is regulated via a normal linear regulation circuit just like a linear power supply. The trick is that the amount of energy supplied is controlled by Pulse Width Modulation (PWM). If the power transistors feeding the transformers are just barely flipping on at 20kHz, then very little power is being passed to the linear regulators. If power transistors are on for 80% of the time during a 1/20,000th of a second, then quite a bit more power is passed down to the regulator circuit.

There isn't that much of a trick to a linear power supply and a switcher. The idea is that the high the frequency of the electricity being passed to the transformer stepping down the voltage, the smaller the transformer can be to handle the same power. (way, way, smaller!)

They get into the PWM tricks because, if your rectifying the AC and then chopping it back up at 20kHz, you may as well massage it even more to further improve efficiency by not making the linear regulation portion throw off a lot of power as thermal energy when there are low demands on the supply. It is a sort of pre-regulation.

Bottom line, the trick is performed via pulse width modulation, not by modulating the frequency of the switcher.

You can smooth out darned near any switcher if you hang enough of those big electrolytic capacitors across the output. Just be sure to limit the inrush current to them at power-up with a power resistor so you the supply doesn't go straight into current-limit when you power it up.

[dirk]

Jon
Sorry your having problems. Your problem drivers are going into reset. I know of only 3 causes; Faulty wiring or bad motor, overheating, or a faulty driver. I’m assuming the same drivers are resetting. I don’t think heat is an issue. The Agek was designed to be very flexible. Any driver can be used for any channel (Axis). I would try swapping the motors (Wago connectors) to a different driver and switch the driver to the corresponding Axis. If the problem persist on the same driver more than likely it’s the driver, or the wiring harness between the driver and Wago connector mounted on board. If the problem moves to another driver than it’s the wiring or motor beyond control panel. If it turns out to be the driver I would get a spare and return to Gecko to test. They may come back that the drivers are ok since this is an intermittent problem.

The Agek is no longer being produced. This was a project I took on entirely on my own and one that I’m most proud of. Financially it ended up costing me, but the rewards of seeing something I came up with materialize into what is now a mainline system gives me Goosebumps.

Anyways if any of those with Ageks are having problems drop me a line and I’ll try to help.

Dirk

[jdw]

Hello All...its nice to break for lunch and see how many folks are willing to lend a hand...Thankyou!

Stan, I too purchased from Dirk prior to the G4 being available...Sept/Oct 2006. From the beginning I had intermittent problems, however I was not using the SB all that much. I'm now in the middle of a millwork project that utilizes $4K of spanish cedar glued-up into semi circles so any lost axis results in a huge set back. And I would like to make the bot much more of my business..so I want to get this figured out.

I ordered two drives today and they will be here tomorrow. I will send the 2 suspect drives back to Gecko for analysis.

Dirk...thank you for contributing. I too have been very pleased with the overall performance of my Agek controll box...especially now after reverting back to the original box...it simply is unacceptable! I have not swapped the drive/motor that you mention, I will do that. As you point out, this will isolate to a large degree.

Dirk, do you have a schematic of the box? This would be helpful not only now but in the future when I want to fully take advantage of what you have provided.

Stan or Dirk, can you tell me more about what Stan mentioned..."Make sure the resistor that sets the motor current is in place. I run mine with the auto current reduction (internal dip switch setting) although some botters might disagree." Where is this resistor and is the internal dip switch in the Gecko drive??

Thanks again for all the help!

Jon

[dirk]

Jon
I don't have a schematic readily available. All of my files are archived on a removable drive that is in storage in another state. The control circuits used are virtually identical to the current 4G. Your power supply is a 300 watt 50 volt torroid transformer which recitifies to 70 volts. The capacitance is more than enough at 12,000 uf for the supply.
The resistor mentioned is direcly connected to the Gecko Drives terminal block, If I remember its about 15k which provides approx 1,75 amps as opposed to the 2 amp setting on the 4g. If this resistor was not connected properly the motors would be getting a full 7 amps, however they would be noticeably noisy and run hot and more than likely pop a fuse.

Dirk

[richards]

My appologies to everyone for the tone of my posts yesterday. When I re-read everything this afternoon, I was more than a little embarrassed. For the last several weeks my wife has undergone several tests for cancer. This morning she had a large benign tumor surgically removed. Until the surgeon reported that the tumor looked normal and that there was no indication of cancer, I hadn't realized how much stress had built up during the last few weeks. And, I especially hadn't realized that I'd allowed my stress to show up in my posts.

Dirk has given an important clue in the possible cause of the problems being reported. If a 15k resistor is being used and if the stepper motors that Shopbot sells are electrically equivilent to the PK296A1A-SG3.6 motor in Oriental Motor's catalog, then the motor is being over-driven. The PK296A1A-SG3.6 motor is rated 1A when wired Bipolar Series and 1.5A when wired half-coil. The formula that Mariss uses for the Gecko G202 is: 47 X I / (7 - I). So, 47 X 1A / (7 - 1A) = 7.8K. An 8.2k resistor is the closest standard resistor. If the motor were run at 175% of normal current, problems could eventually develop.

The current limit resistor is connected to terminals 11 and 12 on the G202. To measure the resistance, remove one end of the resistor from the G202 and then measure the resistance.

The "reduced current mode" jumper is located on the option block. If the jumper connects pins 2 & 6, then the reduced current function is turned on and the G202 will automatically reduce the current going to the stepper motor about one-second after the motor stops receiving step pulses. If the jumper connects pins 1 & 5, then the current going to the stepper motor will stay at 100% - and both the drive and the stepper motor will run much hotter. The G202 can handle stepper motors drawing as much as 2A when the jumper is in position 1 & 5, so either jumper setting could be used with a PK296A1A-SGxx motor.

Edited: The portion of the G202 manual that describes reduced current mode is a little confusing. Rather than trying to explain it, here is the entire text:

(2) REDUCED CURRENT RANGE: In addition to the normal current range (1A to 7A), the G202 can also operate over a reduced current range (0.3A to 2A). This range is used for motor phase currents of less than 1A. Auto current reduction is not available for this reduced current range.

Auto current reduction and reduced current range are two different things; however, stating that reduced current range is for motors that draw between 0.3A and 2A and then saying that this range is for motor phase currents of less than 1A has me scratching my head. (Personally, I have the jumper set to allow auto current reduction on all of my stepper drivers.)

[harold_weber]

Hey Dirk, since you are offering to help AGEK owners I have a couple of questions:

1. You say you have designed for a current of 1.75 amps. In the Gecko "Step Motor Basics" paper, the formula for power supply capacitance then results in a required capacitance of 80000*1.75/70 or 2000uf. Since I have 6 steppers, I should have 12000uf, right? Looking on my AGEK board, looks like the two capacitors are labeled 4700uf. You just told Jon Coote he has 12000uf. Why do I have 9400uf and should I be buying larger capacitors?

2. I think I have the PK296A1A steppers and I think they are wired as bipolar so they should only be getting 1 amp, not 1.75 amp. What prompted you to design your system to raise the current by 75% over the stepper motor manufacturer's recommendations? You say that the ShopBot 4G board is running a design current of 2 amps so I guess this is a question for them as well if they run a design current of 2 amps.

3. Finally, will you please provide the name of a source where I can buy some spare connectors that plug onto the AGEK board and feed step and direction signals to the Gecko drives?

[jdw]

To those who may glean something from this, an update;

1. I replaced the Gecko drive that was reseting.
2. I replaced my resistors on all Gecko drives to better reflect Gecko's recommendation for my motors, 8.2k.

Gave it a go and after 4 hours the same driver went.

I now have switched x1 & x2 motors keeping all cables in same place...gave it a go and a different driver went but same mnotor...so the obvious culprit is the motor.

What happens with the motor that would cause the drive to trip? The motor never got hot.

Thanks for any input.

Jon

[Brady Watson]

Loose wiring and or bad connections at the terminals. Take the wires that go to that motor out of the Wago connector and re-seat them at BOTH ends. Write on the Wago what colors go where to save posting later on when you forget...

It is doubtful (but not completely impossible) that there is something wrong with the motor.

-B

[jdw]

Thanks Brady...I suspected this first as well. However, while running it would not short regardless of how much I jiggled, pulled, and generally disturbed the connection. I wonder if a previous owner could have played with connections inside the motor? Whats inside? Is it unwise to open it up?

Thanks,

Jon

[richards]

Jon,
Don't open the motor! A stepper motor is magnetized after it is assembled. Taking it apart could very easily make it act like an unmagnetized motor - which is to say, that it would become just another piece of metal.

[jdw]

Well I thought I had this narrowed down. To recap, I put in a new Gecko on x2 which was the drive consistently resetting...then that new one tripped. I switched x motors and the motor which was causing x2 driver to trip, was now in x1 position, tripped x1 driver. Replaced suspect motor with spare one on hand, ran 4 hours without incident, took a break for lunch, started back up and with in 15 minutes x2 driver tripped...this is the new driver.

In total, I've had three drives go into reset, with three different motors.

Whats that tell me???

Does anyone have a 4G box for sale or rent? SB says it could take awhile to have one ready.

[richards]

Jon,
You've narrowed the possible causes down, even though the cause may still seem inconclusive. First, by stating that the G202 goes into power-down mode (meaning that it can only be reset by removing all power and then restoring all power to the driver), shows that the G202 is getting a high current load from the motor OR from something between the motor and the G202. Second, by replacing both the motor and the G202, you've probably eliminated both the motor and the G202 as being the cause of the problem.

Where does that leave you? As far as I know, there are only two physical items between the motor and the G202: 1. Cabling and 2. Controller card.

You've already said that you have really wiggled the connections without being able to cause the G202 to shutdown. Personally, I use DIN terminal blocks on all of my Gecko/stepper motor connections (unless I'm only testing a motor and then I connect the motor's leads directly to the Gecko stepper driver). If you're not familiar with DIN terminal blocks, just visit www.factorymation.com (http://www.factorymation.com) or www.automationdirect.com (http://www.automationdirect.com) and do a search on terminal blocks. You'll be able to see some photos of a terminal block. They're bulky and ugly, but they eliminate loose connections. I also use high quality shielded 18-gauge stranded cable to connect a stepper to a Gecko. Smaller gauge wire can also be used (down to 22-gauge). (I'm fortunate to have a supplier that stocks high quality 18-gauge wire at an affordable price. The larger the wire, the less the current drop.)

If you hard-wire the stepper using terminal blocks, then the only part of the interface left is the controller board. I've only closely looked at a 4g controller card one time, so I'm no expert on that card. However, it never hurts to look at a controller with a magnifying glass. Carefully look for any loose wire strands that may have gotten wedged between circuit traces. Look for discolored traces or other signs of high current problems. Make sure that all connections are tight (not just snug).

Finding the cause of an intermittent problem is one of life's most difficult and tedious tasks. Approaching the problem systematically (like you are doing), and following all possible paths - one at a time - usually brings success.

[richards]

Jon,
Since I last posted, I looked again at a photo another Shopbotter sent to me showing the motor's spec plate for the 3.6:1 geared motors used by Shopbot. That photo showed that the current rating for the motor was 1A and that the resistance as 4.xx ohms. That alarmed me because, NORMALLY, Oriental Motor lists the UNIPOLAR data on the motor, not the Bipolar Series data. If the spec plate is corrent, then, the Bipolar Series data for that motor would have a maximum of 0.701A of current. If the motor is really only able to use 0.7A, then even rating the motor at 1A and then running that motor with current reduction turned OFF, would cause the motor to over-heat.

Because the Oriental Motor part number is proprietary to Shopbot, I have no way of knowing EXACTLY what the numbers mean. It would be really nice if someone from Shopbot that is privy to their specs for that motor would post the complete specs for the motor. If they did that, maybe a few of us could compare notes to see if someone might have specified the wrong resistor value for the G202 stepper driver.

[richards]

Here's a photo that I received (which I cropped and rotated for clarity) of the 3.6:1 geared motor that clearly shows the motor's ratings.


6632

As I posted above, ALL of the Oriental Motor PK series stepper motors that I have (all eighteen of them) list the UNIPOLAR current rating of the motor on their nameplate. The UNIPOLAR rating is 1.4X greater than the BIPOLAR SERIES current rating of the motor. If the A6488-9412KTG motor is really rated 1A UNIPOLAR and 0.7A BIPOLAR SERIES, then the correct current limiting resistor for terminals 11 and 12 on the G202 stepper driver would be 47 X 0.7 / (7 - 0.7) = 5.2K (or the closest standard resistor to that value).

EDITED: I just received an email clarifying the motor's number. The photo is of a 7.2:1 motor. I was told that the correct number for the 3.6:1 motor is NOT 6488 BUT 6497.

[richards]

I'm a little embarrased to admit that I jumped to conclusions on the current rating for the A648809412KTG motor in the photo above. When I contacted the owner of the motor and asked him to measure the resistance between the Black wire and the Yellow wire, he said that there was no yellow wire comming out of the motor. When he peaked under the strain-relief, he could only see four wires exiting the motor.

With only four wires, there would only be one possible current rating - Bipolar Series, which also matches the PK296A1A-SGxx motor's Bipolar Series current rating.

Sorry about any confusion that I may have caused.