[Chris Duncan]

Bought this machine in Jan 2013. It's been sitting most of the time since then. It's got about 20 hours of use on it.

Recently doing a project in 2024-T3 using a 1/2" 4 flute cutter with about .010" depth of cut at about 3000 rpm. Taking a slow feed because a lot of time is already invested in this part on the manual lathe. Cuts fine for about 10 minutes then the motor slows to half speed. Then after turning it off and on it cuts fine for another 20 minutes.

The next day it goes 15 minutes again and then slows and totally stops although the table keeps moving. It gouges the part but luckily it wasn't at full depth.
Run the motor with no load and it does the same slowing to 1/2 the set speed even using the manual motor control on the cabinet face. Open the cabinet to look at the VFD and it doesn't act up again after that.

Suspecting the Chinese motor, it just doesn't sound right when it slows down. Guessing the only other thing it could be is the VFD but it doesn't look Chinese.

Sort of disappointed as this machine has seen very little use and it already has a major problem. Guess we'll see what Tormach says on Monday, maybe there's some known issues.

[tmarks11]

I don't know where the Emerson Commander SK VFD is built, but it is a top-notch device.

From your description this sounds like a thermal issue (motor or bearings are overheating, or VFD is assuming the motor is overheating), or VFD overheating. VFD overheat (based upon the symptoms) is most likely the problem.

The drive thermal protection scheme (see Pr 5.18 on page 80) reduces the switching frequency automatically when necessary to prevent the drive from overheating. It is possible to disable this feature by setting this bit parameter to On(1). If the feature is disabled the drive will trip immediately on O.ht1 when the IGBT temperature gets too high.

Drive over-heat based on thermal model (see Pr 7.35 on page 108). The drive will attempt to stop the motor before tripping. If the motor does not stop in 10 seconds the drive trips immediately. Thermal model looks at DC Bus ripple as well as output current. This is to protect the DC Bus stage from overheating.

If this trip occurred, the VFD should be displaying "O.ht1" or "hot" on the display after it stops the motor.

Two possible causes: the cooling fan on the VFD has failed, or the heat sink is clogged by dust. You should be able to hear or feel the cooling fan when the VFD is running (although the Tormach controller case fan might make hearing it difficult). I had this problem with my Miller TIG welder...

If you can tell the cooling fan on the VFD is running, than manually start the spindle (using the panel controls), and run it at 3000 rpm until the problems shows up. Monitor speed and load from the VFD display (more on that below) during the run, and look for the "O.ht1" or "hot" error (VFD overheat) or "OVL.d" or "It.AC" (current overload trip from thermal binding causing current spike, or from VFD deciding that the motor might be too hot based upon current level and time it has been operating) when the motor gets stopped by the VFD.

If the fan is running, and you get "O.ht1" or "hot" then it is possible that the heatsink on the IGBT (transistors) is clogged with dust that his been sucked in by the VFD cooling fan. Take an air gun and gently (small controlled bursts) spray air in through the vent on the VFD. Don't just let loose; with 100 psi air, you can rip the fan blades off a small fan (and also overspeed the fan). If you have bottled air (like is used to clean computers), this is a safer way to do it... but I always just use my air gun to clean my computers (yes, it might spit out moisture, so you might let it sit and dry a bit afterwards).

If not the above, then it is a thermal binding problem (motor or spindle bearing), or the VFD shutting down because it thinks the motor might be getting too hot based upon current level and run time (which could also be associated with thermal binding):

When the estimated motor temperature reaches 100% the drive takes some action depending on the setting of Pr 4.16. If Pr 4.16 is OFF(0), the drive trips
when the threshold is reached. If Pr 4.16 is On(1), the current limit is reduced to (K - 0.05) x 100% when the temperature is 100%. The current limit is
set back to the user defined level when the temperature (Pr 4.19) falls below 95%.

Do you have a load meter on your 770? If not, you can get the Emerson SK to display % load on its display (Tormach's "load" meter is a simple 0-10v meter that displays the output of what the VFD calculates as the percent load). This is very easy: Press and hold the M (mode) key for two seconds. This toggles the display from SPEED to LOAD. The load will either display as % load (indicated by "Ld", I think this is set by default) or current (indicated by "A"). This does not make any permanent changes to any parameters in your VFD.

Then you can watch the display next time it starts slowing down and see what the load on the VFD does. If the problem is thermal binding (either motor or spindle bearing), you should see the load value displayed rise sharply.

Please post what the VFD display trip message is ("O.ht", "hot", "OVL.d", "It.AC" or something else), and we can help you troubleshoot further. BTW, you can download the Emerson Commander SK manual from Emerson's website.

[Chris Duncan]

Thanks for the informative post, gives me some things to try.

Yes the VFD looks like a quality piece, at least the outward appearance.

I did run a infrared temp gun over everything including the VFD right after the slowing on the 2nd day and didn't see anything much above what you would expect, like nothing over 80 deg F.

No smoke, no burned smell. Do hear a very slight amount of metal screeching from the motor at times, or something in the rotational assembly. But not necessarily coinciding with the slowing.

Not really much dust anywhere in the case. The case has a pretty good air filter system and the machine hasn't really seen enough use to even warrant a cleaning.

No load meter, although eventually want to get one. I will try using the VFD like you say. I had the case door open and ran some cuts on a piece of mild steel scrap and it ran fine. Had to put a piece of cardboard over the lower case opening to keep chips from going in. Now thinking about a window or little door to access the VFD. If you could get load readings right off the VFD why would you need a load meter?

Will post VFD messages. Sort of wondering now if it stores trouble code history.

When the motor slows it sounds like binding at first, like too big a cut, then it sounds like it's being suddenly electrically switched to 1/2 speed. Like the binding is random but the 1/2 speed switching is precise.

[Chris Duncan]

From the Tormach article "Small CNC Buyer's Guide -Engineering and Machine Design"

""Tormach mills use a premium grade 3 phase induction motor and a sensorless vector VFD. Our VFD supplier is Control Techniques, a division of Emerson. Emerson is a world class supplier, manufacturing VFDs from 0.37kW to 1.9 MW (0.5 hp to 2500 hp).""

[tmarks11]

I did run a infrared temp gun over everything including the VFD right after the slowing on the 2nd day and didn't see anything much above what you would expect, like nothing over 80 deg F. ...Sort of wondering now if it stores trouble code history.

The VFD has a temp sensor mounted right to the heatsink, and will shutdown long before any temperatures are approached that you would notice or could measure external to the case.

I didn't see anything in the manual about calling up stored error codes. If it does stored error codes, then I suspect that you would only be able to retrieve them with a computer running their diagnostic software hooked up to the VFD. Some of the higher level parameters are only accessible from an external computer, IIRC.

It is possible that an thermally induced mechanical failure is causing higher load on the motor (which means it runs slower than the supplied frequency, which is called "slip"). At 50% slip, the current will be much higher than full-load operating current, which could result in the VFD shutting down on the over-current error. Hard to guess unless we see the shutdown error code.

But the noise makes me think you are barking up the right tree, and the motor is going bad. Although "occasional" screeching" generally makes me think of some problem with the fan on the motor. Best case is to find a quality Leeson, Marathon or Baldor three phase motor with the same face plate as a replacement (always worth replacing chinese motors with a quality brand when it is time to buy a new motor. Of course, all of those companies now have manufacturing facilities in China IIRC).

[Chris Duncan]

One of the places the Tormach is not really dust protected is the top of the motor. The motor drive case is just open on the top and it is a dusty shop. I have blown the dust out of the top of the motor on occasion and you'd think the bearings and such would be sealed. It's just an awfully short amount of run time for the motor to already be going bad, if it is it's a lemon.

Yes a better motor if it's not too much more dollars.

[Chris Duncan]

Contacted Tormach. They don't seem to have an info to add for this problem, no known issues. The motor is about $250 and the VFD about $350

Ran the thing for about an hour today cutting mild steel, couldn't get it to act up. But the Mach3 wizard for a circle is glitching but that's another problem.

[R.DesJardin]

I would check all the wiring connections with the power off of course. Use a small screw driver and also unplug and replug the connections where applicable, be careful. Takes maybe 5-10min to do this. I do it maybe once a year just as a maintenance thing.

[MichaelHenry]

I finally got around to the Series III upgrade on my Series I mill and found that several of the wires popped out of their little on-board terminal strips with very minor handling.

[bobeson]

I experienced intermittent speed-control problems that sound much like, although not identical, to what the OP is describing. The fault for my machine turned out to be the ribbon cable which connects the bulkhead-mounted 25-pin connector from the control computer to the machine control board. The insulation-displacement connector on the ribbon cable appears to be at fault, presumably from an intermittent contact. You might try wiggling that particular cable around a bit when the fault returns.

[Chris Duncan]

Duplicated the problem again tonight. Was running 500 rpm no load, just checking the run-out on a drill bit. The motor stops completely. Quickly looked at the VFD (case is open with cardboard shield on the bottom 3/4) and it was reading 0 rpm with no fault codes or anything. And Mach 3 was still reading 500 rpm with the yellow run light on. Spun the spindle by hand (after turning the key off) and it spins like butter, no binding at all.

So starting to suspect a wiring connection. Because when I first got the machine there was more than one loose connection. I wiggled the serial cable, I'll try the ribbon cable next.

Also I think the squeaking noises I've been hearing is just the drive belt.

[tmarks11]

well that is just annoying.

I would agree that it is probably a loose wire. The BOB delivers a 0-5VDC signal to the VFD to control speed, coming from pins J1-1 and J1-2 on the BOB, going to T1 and T2 on the VFD. If the wire was loose, it would be providing a 0V signal (i.e.. commanded 0 rpm). I would remove, inspect (corrosion?), and retighten those terminals.

Question:
Was your VFD reading "Fr 0.0" or "rd 0.0" after it stopped? If it says "rd 0.0" than the BOB is not sending the run command (28vDC from J1-2 to J1-3). The manual describes this in Table 5.1 and 5.2.

[Chris Duncan]

Question:
Was your VFD reading "Fr 0.0" or "rd 0.0" after it stopped? If it says "rd 0.0" than the BOB is not sending the run command (28vDC from J1-2 to J1-3). The manual describes this in Table 5.1 and 5.2.

It was reading "Fr 0.0"

[popspipes]

I had to reinstall Mach 3 a couple times for various reasons, you say it has been sitting for quite sometime, I would try a reinstall just to eliminate the possibility.
I have had no problems with Pathpilot in this regard.

[tmarks11]

It was reading "Fr 0.0"

Then I would suspect the 0-5V analog speed signal is the fault.

I would remove, inspect (corrosion?), and retighten the wires from pins J1-1 and J1-2 on the BOB, and on T1 and T2 on the VFD.

Also could be the ribbon cable from the inside of the parallel port connector up to the BOB, as was previously mentioned.

[Chris Duncan]

FINALLY FOUND IT. Went out to the garage to check something with Mach 3. It was cold (under 50 deg F) and didn't want to fire up the heater just to play with Mach 3. Was trying to run a wizard without a bit, just cutting air, and the motor did not spin. Opened the case and pushed on the bottom center ribbon cable where it connects to the break out board, had to push pretty hard, flexing the board a bit, and the motor would start up. Release pressure the motor would sometimes stop. Pull on the ribbon cable, motor stops. Held the board from flexing and pushed the cable, motor did not run. Let the board flex, motor runs. Pushed on the green 6 pin screw connector just to the right of the ribbon connection on the BOB, motor runs. This connector contains most of the power wires going to the board, so not VFD signal but power for the VFD signal maybe.

Tightened the screws on the 6 pin connector. Now motor runs all the time. I think the cold was a factor, and going even colder helped diagnosis. The first time it acted up was about 55 to 60 deg F, and below 50 it was doing it all the time.

The connector screws are the optional type, straight blade or phillips. But I think the phillips may be some oddball type because a standard phillips doesn't fit very tight, and slips before you can really get a lot of torque. Because I had already tightened these screws one time. Used a straight blade and it allowed to get them a little tighter. 4 of them were already tight but 2 of them got about a 1/4 to 1/3 turn tighter. Just can't stress enough that all connections need to be checked when you get these machines new.

[pickled]

Great news! It is very rewarding when you track down an oddball intermittent! After reading this I am going to run out in the morning and check all my connections just to make sure that I don't get bitten by the bug too. Congrats on prevailing.