Printable View
Can anyone think of a reason why opening the main collet closer for more than a few (like 10 or so) seconds would cause a hydraulic overload and trip the breaker?
It only happens when the main collet is open. The sub is fine. Shouldn't the hydraulic system go into bypass mode rather than blow a breaker and shut down?
Torin...
I just looked at my manual and it looks like some machines have a pressure switch that reads ''chuck open'' (mine does not) It could be that the switch is not telling the computer that the chuck is open and to release the hydraulic pressure on open. I would expect relay MCDF.S to dropout on reading open pressure. To test this, maybe open the chuck and pull the relay from it's socket and see if the pump stays running.
Great idea, Jim. I'll poke around back there and see if the switch is faulty, loose, etc. Will update here after I've had a chance to look at it.
Torin...
I cannot find MCDF.S in the DIAGS or LADDER. What address is this bit located at? I'd like to watch it in the diag screen as the collet opens.
Torin...
In my book, the address is X16.0. The O/L for the hyd pump is set to about 3.6 amps.
The MCDF.S is the label on the relay in the magnetics cabinet. Circled in red below
https://www.cnczone.com/forums/attac...d=389072&stc=1
BTW, if you need parts, I am in process (nearly complete) of upgrading my machine. I removed everything that says Fanuc on it with the exception of the main spindle motor, so I have a few extra parts. The machine was working fine, so all of the parts are believed to be good, and I didn't cut any cables.
Now I understand. Your machine is quite different than mine. I'm running a Conquest T42, with Fanuc 18-TB control. Your machine appears to be a Conquest 42.
The schematic in my machine shows three-phase power coming into a 3-phase contactor
Attachment 389078
then from that contactor to a protective relay (also set to 3.6A, which matches the value printed beside its symbol in the schematic) that measures the current flowing through the circuit.
Attachment 389080
This protective relay is driven by I/O line Y4.0 which comes from Slot 9, A0.
After measuring more carefully, I have a new development... after watching the machine for a while - powering it on, measuring current through one of hydraulic motor leads, and having my son repeatedly open and close the collet, I could see no discernible difference in current consumption from one state to the other. They both read 0.15V (which on my cheapo-ammeter could be anything, but at least the current consumption is identical)... and then suddenly the machine shut down in the closed position. All at the same time, my shop door was open to the cold air, bringing the shop down to 5 degrees Celsius. For the moment, I will additionally pursue this as temperature related (cold = higher viscosity = higher motor consumption?) and try to see if the problem goes away when the shop is at a higher temperature.
This also leads me to wonder if someone changed the hydraulic fluid with an improper viscosity. My filter cartridge is showing a little red button indicating it needs to be replaced. I've ordered a replacement from Hardinge, I may as well take this opportunity to give it a flush and an oil change too.
Torin...
Yes, mine is a Conquest 42 with an OT control, the machine controls are quite different. The hyd motor starter circuit is almost identical, but the circuitry that controls it is different. It could be that the O/L (protective relay) is set on the ragged edge of holding and a small additional current draw of holding the collet open might be just enough to push it over the edge. I think you need to get a good reading on the current draw on all 3 legs to really make sure the motor is not going over the rated nameplate current. Then if everything seems OK with the motor, increasing the current on the O/L might be the fix. I have seen those get weak over time.
Jim,
I kept editing my post while you were responding, but I think you and I have come to the same conclusion - the relay on the edge, possibly caused by temperature as I've edited below, but also as you've stated - the circuit could be saying 3.6A, but tripping earlier. I will do as you suggest and take a reading on all three legs and make sure that none of them is tripping the protective relay (and especially not over the nameplate rating.)
Thanks for listening, Jim, and thanks for your input. I will post again once I calibrate my ammeter and measure all three legs. Now having looked at this, I suspect it's not as an elusive problem as I first imagined.
Torin...
Another interesting development:
I was fed up with the clamp meter results and decided to use some old Micronta DMMs inline with each phase instead. Each meter was set in current-reading mode (up to 10A) and inserted after the protective relay in series with motor leads labelled 17L1, 17L2, and 17L3. The results were unusual. I am gettting 3.9A ~ 4.0A on phases L1 and L3, but only ~1.2A on phase L2. To rule out the possibility of a faulty meter, I swapped L1 and L2 only to see the 1.2A follow to the other meter.
Attachment 389178
So now... does this seem right? Could one of the phases in the motor be bad? Could that same bad phase, not pulling its weight, cause the other two phases to jump from the nameplate FLA of 3.8 up to 3.9A and 4.0A?
Now I can understand why the protective relay is tripping out - the two outer phases are exceeding whatever current setting on the protective relay, but why would this middle phase be drawing such low current? Current leak? Bad motor?
Could I have measured this incorrectly? I think I need to measure the windings to make sure there aren't any shorts, etc.
You said you have parts - what does your hydraulic pack motor look like? Does it run on a Leeson 1HP?
Torin...
Sounds like you have a bad motor. If one leg is low, the other two will try to make up the difference because the motor is still trying to run at synchronous speed. Check the phase to phase voltage between all 3 legs. If the voltage is pretty close phase to phase and one leg is low current, then that would mean an open winding or a bad connection in the terminal box on the motor.
My motor is a Nachi 0.75 KW. Rated at 3.2 amps @ 220V per the name plate.
EDIT: The hyd motor is not one of the parts I removed.
I figured it out. I should have started with what you said, but I don't think I would have made the connection anyway until after I went through this process.Quote:
Originally Posted by Jim Dawson
Started by disconnecting the motor and measuring all the windings (dual voltage motor, 9 leads). First, every winding to ground - no faults, then measured resistance of each winding, and finally from each winding to every other winding (again, no faults.)
Attachment 389216
So this confirmed that the motor is good (whew!)
Next, I thought to test for a short/leak in the cable leading to the motor, but that turned out ok.
I then traced up through the circuit breakers, relays, etc. and noticed a short between L1 and L2. Was puzzled for a bit, but traced it to a 24VDC transformer across L1 and L2. So that makes sense now..
While out at Canadian Tire picking up closed-end crimp connectors to replace the ones I removed from the motor, I thought, perhaps, the three-lead AC cap on the motor itself was leaking. Nope. It checked out at 0.1uF between all three leads. Damn, thought I had it there.
So finally, I ran your voltage test and noticed 230, 215, and 208. Yes, it then occurred that 208 may be causing the droopy L2. I thought I had balanced out the 3ph RPC, but apparently I didn't do a good enough job, so I ran out with 30 minutes to spare before the electronics store closed and picked up a couple of 40 and 50 uF motor run caps. After plugging these in and a bit of trial and error, I arrived at a better balance between the phases UNDER LOAD (that's the lesson learned here - I didn't apply the hydraulic motor load when I originally balanced the RPC.)
Attachment 389218
So now with the meters in series with the hyd motor leads again, I fired it all up and voila - no more droopy L2. It's not perfectly balanced... still getting a little higher current on L1, but it's way better than before.
Attachment 389220
On Monday, I will pick up a few larger run caps and balance it out even more which will drop L3 and pick up the current on L1 and L2, and that should just about do it.
Torin...
AWESOME ! :cheers:
Here is how I did my RPC, adjustable on the fly. :) https://www.hobby-machinist.com/threads/another-rotary-phase-converter.63204/page-3#post-534110 (Scroll up a bit)
I like the switch idea, but I would want it done automatically somehow... for example using a voltage sensor on each leg that, when the voltage swings too low, closes a NO relay that engages another capacitor. Or possibly even better, a current monitor like those protective relays that measures current and when it exceeds, kicks in the extra capacitance (without tripping permanently mind you.)Quote:
Originally Posted by Jim Dawson
In my case, L1-L3 measurement is always rock steady at 230V, but I needed to add an additional 80uF and 100uF to boost L1-L2 and L2-L3 respectively. Having an extra 50uF on each of those pairs with an automatic switch would boost their voltages by another 5V or so in cases of heavier load. But for now, I've got way too many things on my shopping list to worry about than automating the RPC.
After picking up more (larger) caps from the surplus supply store, I managed to get L1 == 230V, L2 == 234, L3 == 234 while under load from the control with hydraulics running, which is within 2%. Haven't checked when spindle and motors ramp up, but I'm not worried because I don't make heavy cuts. I'm pretty happy this all worked out and nothing was broken.
Torin...
Hey Jim, can I bend your ear again?
For a while, I thought I had solved it... and in theory, I still have - but now when I have the machine running on all cylinders with the primary and secondary spindles running, and the collet open, the machine shuts down mid cycle again.
I decided to watch the phase balance as the machine is running, so I connected those three multimeters across T1-T2, T2-T3, and T1-T3 and watched as I stepped through my program. Obviously the voltage is going to jump around as more or less current is drawn by the system, but I'm seeing drops from 226V nominal to as low as 190 during "air cuts", then the voltage recovers. I've never watched the voltage before so I'm not sure what kinds of drops are considered normal.
Sure enough, when the motors are spinnin', and the axes are movin', the balance is out of whack again, and I think this is causing the hydraulics to over-current again.
Oh, and one last thing. Now that I'm watching the voltage, there is a definite drop in voltage while either collet is open. It is most pronouced when the main is open, and slightly less when the sub is open.
232V
245V When RPC on but machine is off
230V
222V
226V when main closed/sub closed
226V
219V
217V when main open
226V
219V
222V when sub open
226V
when both spindles are turning and the main and sub are both open, the voltage drops even lower and only takes a few seconds to trip the hydraulics. Sigh.
My RPC is rated for 20HP, and my motor is also rated for 20HP. My machine is rated for 15HP, but I am not driving the machine anywhere near full capacity. I'm just spinning empty spindles and doing air cuts. I should be well within the capacity of my RPC, shouldn't I?
Torin...
I would think that a 20HP RPC should be good for a lightly loaded 15 HP machine. The hydraulic pump actually adds to the RPC capacity. Based on your numbers, it looks like you are getting quite a bit of voltage drop on the single phase feed to the RPC. Are your wire sizes adequate between the shop breaker panel and the RPC. How about the shop main feed? I had a bad connection on the main feed to my breaker panel that was giving me fits until I found it and fixed it. I used an IR thermometer to locate the hot spot.
I'm guessing that #6 copper wire minimum would be needed if it's a reasonably short run, <100 feet. #4 copper might be better. And the same between the RPC and the motor.
A side note, I just about have my Conquest 42 back together after a complete controls retrofit. Ran the spindle for the first time today. Bench tuned the new servos yesterday. Getting close. :)
For what its worth. I have a Conquest 42 and ran it for a long time on a 10hp converter with no problems, only recently upgrading to a 15. Our voltages are similar.
The hydraulic load with the chuck open or closed should be the same. Its a double acting hydraulic cylinder after all. Do you have the correct oil in the tank? I think it calls for DTE22 for normal temp operation. A 22 weight oil.
I did some testing for you.
I have a Lesson motor but I cant read the HP of it
The recommended oil is actually DTE13M on my machine. A 32 weight oil.
machine sitting idle out of E-stop
amp draw at main disconnect
L1 4.9
L2 5.0
L3 2.6
amps after Hydraulic unit overload
L1 2.5
L2 2.8
L3 2.3
volts at hyd contactor for ref
L1 - L2 242
L1 - L3 238
L2 - L3 240
L1 - G 120
L2 - G 120
L3 - G 208
So a few things... and I apologize for how lengthy this is, but I've looked at a lot of stuff these past couple of days.
Jim, I took your advice and looked at the cabling. I have 30ft of #6 from the main panel (protected by 60A breaker) to my garage pony panel. From the pony panel, I have 30ft of #8 (protected by 50A breaker) to a buck transformer at the RPC/lathe.
In Ontario, we get "120/240V" which is actually closer to 125/250. It varies, but I often see it closer to 125. My fluke in the 115V wall outlet shows 123.3V as I type this. Using the voltage drop calculator on calculator.net, I plug in:
Segment 1
246.6V in, 6awg @ 30ft, 50A, 1.19V (0.49%) drop = 245.41V out
Segment 2
245.41V in, 8 awg @ 30ft, 50A = 1.88V (0.77%) = 243.53V out
And finally, my buck transformer now reduces this voltage to nearly exactly 230V then feeds into the RPC.
So at 0A load, I should see 246.V corrected to nearly exactly 230V. When I apply load up to 50A (also my breaker value, so I know I'm not exceeding this), I will get a maximum ~3.1V drop, which is ~227. Theoretically.
I should be running 6awg all the way to the lathe, but the 60/75/90 temp ratings for #8 are 40/50/55. I figured I could get away with 50, but I know I will never push the lathe beyond half it's rating for more than a few roughing passes (i.e. a few seconds.) All other operations run well below 5~10% (90% of operations are machining aluminum using .005/ipr and no more than 0.025 doc) One day, I will do this just to be even safer than safe.
So thank you Jim - it was a good idea for me to review that. All these calculations are theoretical though, so I still want to run over the pathways, make sure all the connections are tight, and measure reality against theory just to be sure. Would be nice if voltage drop is the issue.
On Thursday, I saw the A-C voltage somewhere around 226 ~ 229 depending on the state of the CNC and how I balanced the RPC. Tomorrow, I will start taking better records and produce a table of "balances". For each combination of capacitors, I will produce a table of voltages from no load, to steady-state (out of e-stop, hydraulics on), to running with the spindle performing cuts. I will then be able to see the amount of imbalance for each tuning, and settle on the one that gives the best balance for the hydraulics in the steady state.
BTW, I talked to my RPC manufacturer, and he gave me some interesting advice: He indicated I should find which of the legs run single-phase stuff (esp the control) and make sure my incoming lines (A and C) are connected to those two legs. Though my control is three-phase, the schematics show the whole 24V circuit and all the cooling fans on L1/L2. It turned out I had the generated leg driving L2, so I swapped them around properly. I don't think it made much of a difference though. The middle leg still sags a lot when the spindle ramps up... but as long as it's balanced for the hydraulics, I think I'm ok with that.
Next, I tweaked my RPC caps again, taking off a few and balancing for steady-state with the hydraulics running. I was able to run a program with the collets opened and closed long enough to get through the program.
Finally, I read this article
Voltage Balance & Three-Phase Motors - Phase Converter Information
And it freaked me out a little. It appears that a slight voltage imbalance can actually result in a large current imbalance, which is perhaps why I observed such a large current differential on my hydraulic motor (one was at 1.5A while the other two were pushing hard at 4.0 each, thus tripping the breaker.) I'm afraid that over time, I will ruin my hydraulic pump. So now I'm thinking I need a better RPC - perhaps one with dynamic balancing.
Now, of course, this article also leads to a "Phase Perfect" technology website that shows how their technology (purportedly the only *true* digital phase technology on the market) gives me exactly what I want - perfect balance regardless of load. I was sold... until I flipped to their pricing and learned their 20HP PT-355 is $5000 shipped. Their 10HP is too small, and they don't have a 15HP model.
I am also strongly considering American Rotary Phase Converter's "AD" line (their "CNC compatible" AD-30 model), but I can't get an engineer on the phone to convince me their system will balance any better than what I have now. They claim their balance is strongly centered in their idler technology (special windings and cages, etc.), but adds "digital balancing" which I think simply means switching capacitors in and out of the circuit as the load varies. However, WNY warns against using the digital technology because adding a cap at the wrong point in the swing can actually add the residual stored in the capacitor to that alreay on the line, thereby potentially double the voltage and causing spikes to ripple through one's CNC. Holy. Something I definitely need to talk to the engineer about.
General Disarray,
Thank you for measuring all that for me. Yours are all within 2% during steady state, and that explains the small imbalance in the hydraulics. It gives me a target to shoot for.
But what does "amps after Hydraulic unit overload" mean? Did you measure when the collet is open, and again when the collet is closed?
I talked to Hardinge and they too say it shouldn't increase the load - that the closer has a bypass at the end of its travel that relieve the pressure, yet I see a definite and significant measurable difference when the main closer is open. For example, in the open state, I can get readings of 228/226/226, but when the collet opens, I get 228/219/222 (or something like that). The subspindle also induces a slight load increase, but it far less prominent. When they are both open, the middle leg drops to 217.
Hardinge doesn't seem to have an answer to "why is there more load in the open position", but I only spoke with one guy. I will keep trying. I will also read over closer maintenance and see if there are any adjustments to be made on the closer mechanism.
When I first bought my machine, the sub-spindle closer was frozen. I had to use the machine's own hydraulics to force the drawtube from the spindle housing. Turned out it had frozen due to a combination of oil, rust, and what I can only describe as shellac. I thought maybe the hydraulic oil itself had turned solid, but when I opened up the collet hydraulic mechanism itself, it was surgically clean. So I don't understand what's going on... if they say each of these closers has a bypass, and it's not gummed up, what the heck could it be? Could the bypass simply be adding a small amount of resistance because of the change of the fluid path, but is being amplified by the imbalance in my system?
Looking at the oil is on my list of things to check - the oil is amber in color. I don't know if the oil is dirty, the wrong type, or what, but I do notice the button on the filter always pops out. I have a new filter, and will need to change the oil as well just to be on the safe side. At least I can eliminate one more potential issue. It would be nice and simple if someone used the wrong oil, but think the RPC balance is more likely the core of the issue.
Torin...
I think generaldisarray may be onto something here, if the oil is too thick then that could overload the motor. Given your experience in post #6 running with the shop door open kind of points in that direction. It could be that you are fighting two problems here, 1 an imbalanced RPC, and 2, high viscosity oil. I am just not seeing anything like what you are experiencing with my machine. There is no change in the pump sound, collet open or closed.
Just as a side note, my power here is crazy also. Normally runs about 245, but I looked at my high and low for today, low was 228.4 @ 03:20, and high was 300.0 @ 04:33. I was definitely not in the shop that early, but we were having some wind gusts.