[torinwalker]

There is something wrong with my vertical slide - when my machine turns on, the hydraulic pump to the vertical slide turns on and never turns off. It continues to draw air until my industrial compressor runs out of air. The only way I've been able to use the lathe is by disconnecting the hydraulic pump.

I figured it was normal and that my 10.5SCFM compressor couldn't keep up, but then I read in the manual the pump should only engage when the slide is being commanded to move. Otherwise, the pump should turn off again.

As further evidence, I was running a job recently when suddenly I got a "Vertical Slide in Motion" error, yet re-engaging the motor did not cause any movement (but the error eventually went away.)

Can anyone discern why the pump is constantly active? I suppose it could be a limit switch that isn't tripping... I imagine the hydraulic pump lacks the finesse of fine movement, so two course switches ought to govern it's feed and retraction. The trouble is, I can only test the motion for about 20 seconds before the pump runs out of air, and neither an M23 nor M25 seem to move the slide in either direction. I wonder if I'm even looking at the problem in the right way.

Is anyone familiar with the installation of a vertical slide on a CHNC machine? And/or, would anyone happen to have a PDF of the vertical slide installation instructions??


Torin...

[txcncman]

I would think the pump is controlled by a contactor. Either the contactor is stuck in the closed position or it is getting a signal to stay closed. Find the contactor and begin your trouble shooting there.

[torinwalker]

The pump is an air-driven hydraulic pump. The air solenoid that actuates the pump IS in the continuously on position because the signal from the terminal block is always on. When I disconnect the signal wire, the motor turns off, so it's not a stuck contactor.

The thing I'm trying to figure out is, why is the signal always on?

The vertical slide has a limit switch at either end of its travel. M23 sends the slide in to cut which opens one switch and eventually closes the otherat the end of its adjustable travel. This switch should tell the software that the M23 is complete and turn off the pump. Likewise, during an M25, the slide should retract which opens the lower (extension) limit switch, and closes the upper (retraction) limit switch at the end of its travel. The odd thing is, the upper limit switch shouldn't be adjustable since it is at the end of its travel.

It is this retraction limit switch I suspect isn't being hit which may be causing the motor to continue to run (until the control detects it has completed the move). It would be nice to hear from someone who knows the vertical slide mechanism to provide insight into this switch function.

Once again, the odd thing is, the control won't move the slide either way. One would think that issuing an M23 or M25 would cause the slide to at least move in one direction, but it's not moving either way.


Torin...

[txcncman]

Then next trouble shooting step would be remove and test the limit switches for open and close.

[Al_The_Man]

What controller is on the lathe?
I don't have personal knowledge of this slide but it sounds to me like the pump is going to run until a L.S is made so what happens if you trip the limit by manually?
If a move is commanded, I assume it runs the pump which would also pick up a solenoid for the extend or retract?
If so is either solenoid picking up? If neither the pump will run continuously.
IF a solenoid is picking up, and the slide is not moving, then it points to a mechanical seize of some kind?
Do you have schematics of any kind?
Not sure if this helps, but it is hard to troubleshoot from a distance.
Al.

[torinwalker]

This is getting more and more interesting.

It's a Siemens 810T.

So, I checked the following:

Home switch, hall sensor: Trips when vertical slide is retracted - registered in PLC->IW location #4, Bit #2 = 1 (true) when retracted, zero when extended. Check!

Limit switch (adjustable), mechanical switch (trips when vertical slide reaches depth limit - registered by control as PLC->IW #4, Bit #1 = 0 (false) when retracted, true when extended (and limit reached). Check!

When an M23 is commanded, the solenoid that admits air to the air motor hydraulic pump actuates, turning the pump on. A second solenoid is supposed to switch the flow of the hydraulic motor, sending the vertical slide inward.

When an M25 is commanded, the second solenoid reverses direction, retracting the vertical slide to the home position where the control recognizes the home switch and turns off solenoid #1 to deactivate the motor.

I verified the switches are working correctly. They are showing the correct bits, according to the maintenance manual.

I verified that the slide itself extends and retracts by disconnecting terminal block #18 (the directional valve), causing the vertical slide to move inward until it reaches the limit switch. Reattaching the directional valve causes the vertical slide to retract.

So... home and limit switches work, and are shown in the correct memory locations, AND they show the correct states when the vertical slide is retracted and extended.

Now... the motor runs continously regardless. That is, once the control turns on, BOTH 110V signals, #18 and #19 are continuously true regardless if an M23 or M25 is commanded.

I did notice however that when I issue an M23, I get a HOLD: No data transfer, and when I issue an M25, the program runs until I hit the limit (not home) switch. So, I am getting some cooperation by the control when I hit the limit switch, but the home switch doesn't appear to be recognized.

I am incredibly surprised that no one on CNCZone has ever installed or configured a vertical slide on one of these puppies. Am I really the only one here who's ever encountered these issues??

So, the issues boil down to these (i.e. my observations are

#1. The control issues a positive signal upon power up to terminal block #19 even without an M23, causing the hydraulic pump to start. This is the first problem - why is this signal even on? If the control recognizes the home signal, why isn't the signal off (to stop the motor)? From a fresh power-up, the signal from the terminal block should be OFF.

#2 When an M23 IS commanded, the directional valve signal should turn OFF, allowing the fluid from the pump to extend the vertical slide (what happened when I disconnected the signal.) This signal does NOT turn off as expected.

I wonder if the PLC must be configured to recognize the vertical slide when installed. This might explain why it's not reacting to the M23/M25 commands. Perhaps it thinks the hardware isn't activated? (Doesn't explain the bits flipping on my screen as I pressed the home and limit switches though.)

Unforunately, I don't have any instructions on what bits to set where to tell the control that I have a vertical slide installed. Somewhere, someone knows this information, and only Murphy will introduce us at his leisure.


Torin...

[Al_The_Man]

Bear with me, what type of control and does it have an internal PLC or is this outboard of the control?
It is very beneficial to have a PLC where you can see a ladder that displays the status of the rung components, this make trouble shooting a breeze if it is available?
Al.

[torinwalker]

I know little about this machine - it's very new to me - bought it in July. It appears to have a PLC because, as I said, I detected the bit changes from each of the home and limit switches in the diagnostics section under PLC->IW. I don't know what IW stands for.

As far as visualizing the ladder, it shows bit addresses one or two eight-bit words at a time, but one must know at what address to look to get the information. Therein lies one of my problems. The maintenance manual shows little information about what addresses hold what information. For the vertical slide, there are two bits referenced - one for the home, and one for the limit. As for the other six bits, I have no idea as I need to switch away from the diagnostics mode to execute any programs.

But I'll hang in there if you have an idea.


Torin...

[Al_The_Man]

IW, Input Word Maybe?
Still don't know what controller it is, I am getting the feeling it is PC based though?
Al.

[torinwalker]

It's a Siemens 810T.

[FanFan]

Hello, I am curious. When you say vertical slide, you are talking about a slide that goes up and down, correct. Not one that goes left and right or in and out? Also when you are trying to the move your vertical axis, are you getting a position readout on your crt display? I guess what I am trying to ask is there a servo motor system used to position that axis. I wonder if the machine is using any type of counterbalance or axis braking/clamping system. Just wondering.

FanFan

PS: here is a link to post that might help you find manuals for the Siemens 810T:

http://www.cnczone.com/forums/siemen...mentation.html

[torinwalker]

FanFan,

Hey thanks for the link. I found that one a while ago, followed it (it's broken), and managed to download a ton of the 810-related stuff. The more modern versions contain a much more comprehensive list of error codes, some of which were not referenced in my original Hardinge manuals.

The slide I refer to is a hydraulically-actuated tool holder for a cut-off tool. It mounts at the top of the spindle and comes down to cut off the workpiece. Mounted in the slide mechanism is two switches, a magnetic home switch, and a mechanical limit switch.

The problem I'm describing has to do with a signal emanating from the control. The control should normally show Zero volts on terminal #19 that connects to the vertical slide motor (an air solenoid that admits air to an air motor that powers the hydraulic pump). When I command an M23, the control drives that signal (a 110V relay) to power the air-soleoid, turning on the air motor/hydraulic pump. A second solenoid connected to #18 which is normally high (also a 110V signal) should then turn off (0V) which causes a directional valve to admit hydraulic fluid to the vertical slide's hydraulic cylinder, driving the tool down into the workpiece to part it off. When the slide reaches a limit, a long set screw depresses a button that trips the limit switch, which tells the control to stop. An M25 reverses this process, returning the slide to home.

When the magnetic home switch is tripped again, the control SHOULD drive #19 low, turning off the hydraulic pump.

Except, #19 is ALWAYS on, regardless. It's on when I turn on the control, and the air motor is such a pig, it depletes my large compressor in a matter of under 1 minute which then trips the CHNC's low-pressure alarm and the machine goes into emergency stop.

The only way I can use the machine is by disconnecting the air solenoid (signal #19) that drives the motor. Then the machine works just fine, except once a while back, a program I was running failed with a transient "vertical slide in motion" error. It eventually went away after mucking with the VS signals (in an effort to get the machine to think it had completed the slide movement.)

After diagnosing the limit switches, I noticed the magnetic home switch was very far away from the magnet - it was supposed to be .008 ~ .022 from the magnet, but it was more like .125. After adjusting it to spec and verifying the bit was turning on and off correctly in the control, the problem was still not solved. The only thing I have left to try is to power up the control from scratch and hope that the initial home switch being detected somehow changes the remainder of the runtime state. That was the one thing I didn't do yesterday evening - reboot the entire control.

It's anyone's guess at this point what the problem might be. I'm hoping to get some schematics for the AC signals coming into the valve compartment. I don't have schematics for that part of the lathe. Perhaps then I can trace the signals back to their source and find some stray chip shorting the signal into a permanent on position. Wouldn't it be nice if it were that simple?


Torin...

[Al_The_Man]

If #19 is always on, even at first power up, is it possible the output for that device is shorted?
If this is a controller output, it could be a solid state device or reed relay etc.
Can you see the status for this output in the diagnostics? If it shows off in the diagnostic, then it would point to a defective output.
Al.

[torinwalker]

I got home today after finally finding a complete set of schematics for my lathe, and while I didn't learn anything new related to my problem, it did however help me understand how the system is wired from control to interface board to relay panel. I must say, having the schematics makes me feel like I can now contend with anything this machine will throw at me, short of smoking chips, of course.

So, I went into the garage and started poking around in the back because I had an inkling the relays might have something to do with it. I went across all the relays, particularly the one I knew worked - the one driving the control and verified 24V logic in, 110V out from the output. I went down the list and verified the first nine relays are working, then interestingly, relays 10 (vertical slide direction), 11 (vertical slide motor), and the rest from 12 through 17 for miscellaneous features such as bar feed, air blast, chip conveyor, and open and close coolant guard door (motors, I guess) had all blown. They were all showing 110V output even though the controls signals were all 0V. I can't believe that many had failed, especially considering so many of them have probably never been used.

Anyway, feeling good about figuring this out, I rushed back to the CNCZone to confirm my suspicions - that failed relays are a usual suspect, and sure enough, Al, you did just that. Thanks.

Are failed relays really that common? I would think that solid state devices would be more reliable than their mechanical counterpart.


Torin...

[txcncman]

Are failed relays that common? What age of a machine are you working on?

[Al_The_Man]

Are failed relays really that common? I would think that solid state devices would be more reliable than their mechanical counterpart.
Torin...

Are these SSR output? They can fail like anything else I would not expect that many, were these in operation? or seldom used?
Many controllers use either SSR or reed relays, which are prone to failure.
The easiest to change of course are the type such as Opto22 plug in type, but these also have a fuse for each which usually saves them.
Do you have a number on them?
Al.

[torinwalker]

Yes, they're Crydom D1210S (or just D1210) which are 4 ~ 32VDC input and 28 ~ 140VAC output. They're Solid State Relays in a "hockey-puck" package (a small rectangular cube about 3/4" high, 1 1/2" wide and 2" tall, built-in heat sink, two mounting screws, and four screw-terminals for wiring.

Honestly, I'm very surprised that so many (7 out of 17) have failed. As I mentioned earlier, these relays drive a whole bunch of features that are unused, but they're still all wired to 110V regardless, and are therefore subjected to potentials throughout their life of ... 30+ years? Two have been replaced with newer ones, but the rest are original equipment.

I just ordered a lot of six for $49 from eBay. Normal retail price is $34 apiece from digikey. I only need two, but I may as well replace a few more in case I want to use the extra relays for something else, like a through-tool coolant pump, or a bar feeder.


Torin...

[Al_The_Man]

Those are usually pretty rugged, I guess there was no individual fusing?
The most cause of a SSR failing is those that switch AC, because if the inductive device they are switching does not pick up completely or some one pushed a solenoid over manually when it is already enerigised, the current soars.
This does not occur with DC inductive devices.
Al.

[torinwalker]

Hey, now that I have the schematics, I can just look it up instead of poking around my machine in the dark with a flashlight. No, there is no fusing.

Still, it doesn't explain why the ones that weren't in operation blew too. Perhaps they too were swapped to replace blown relays from active circuits?

After looking at my diagram, it appears that I misdiagnosed #13 because it's attached to the worklight and I know that one works. I'll have to go back over the list and double-check to see how many actually failed. Certainly #4 (spindle brake) failed because it is now swapped with #14 (bar feed), but I was really only concentrating specifically on #10 and #11 for the vertical slide.

At any rate, the problem is solved now and I can be happy knowing that every bit of my lathe works as it should. This problem has been around since day one - I even went out and bought a bigger industrial compressor to keep up only to find that it still wasn't enough. I'm just glad it's finally over.

How many people here do you suppose own CHNC I lathes, or any other type of specific lathe? Are we talking ten people? Twenty people per brand/model?

Perhaps I should conduct a poll!


Torin...

[Al_The_Man]

BTW, If nothing was connected to the unused ones, how did you check them for shorted?
Al.