What's new
What's new

Yaskawa V1000 Tripping on e-stop?

dazz

Stainless
Joined
Aug 20, 2006
Location
New Zealand
Hi
This is an opening question to find any experts on Yaskawa V1000 drives.


I have retrofitted a V1000 to a Nadini lathe fitted to a 4kW motor.

IMG_0044.jpg


The install includes an external braking resistor.
When I do a e-stop at average load, it stops very quickly as expected.
I tried an e-stop at high frequency with the lathe in high gear. The chuck spinning at high rpm has significant inertia.
The drive tripped out on over voltage/current.

I have recovered all of the error codes so I can provide those if there are any experts who can troubleshoot.

Dazz
 
Error codes are always good. Overvoltage and overcurrent are very different alarms.

The manual lists 'Adjust L3-25 (Load Inertia Ratio) in accordance with the load' as one fix for the overvoltage fault.

Playing around with L3-04 could be an option too.

Make sure the brake resistor is connected and operational.

If you're running in open-loop vector with torque control, setting L7-07 to 1 might help, as it will limit motor torque.
 
Hi

The drive tripped out on over voltage/current.

I have recovered all of the error codes so I can provide those if there are any experts who can troubleshoot.

Dazz

The DC bus voltage is going beyond design limits. Good drives have a mosfet like device that switches the bus to ground through a resistor.
Your drive is not able to handle the sudden changes.

What value is the braking resistor?

expert definiton: somebody who knows 1% more than somebody else on a subject.
 
A few other parameters to look at/try.
C6-02 Carrier Frequency Selection = 3 or 4 (8 0r 10 kHz)
C1-09 Fast-stop Time if used set to time that does not produce a fault
L3-04 StallP deceleration selection = 2 Intelligent Stall Prevention (shortest possible deceleration time is achieved while the motor is protected from stalling)
Also try 3 Stall Prevention with Braking Option (Enables the Stall Prevention function while using a braking resistor)

L3-17 "Overvoltage Suppression and Deceleration Stall and Desired DC Bus Voltage during Motor Stall" 380 V
L8-38 Carrier frequency decel selection = 2 Enabled across entire frequency range
L8-40 Reduction carrier frequency time = 2.00 sec

Autotune should be performed if not previously done. There are also parameters for if the motor is run in normal or HD mode.
 
Overvoltage trip means the resistor is not effective enough to handle that amount of brake energy.

There may be a parameter that adjusts the resistor duty cycle (resistor current, really). If so, follow manual directions for adjusting that.

Check that the resistor is basically working... it is surprising how effective even a small extra drain from a resistor can be in helping with shut-downs.

However, the higher the RPM, the more energy has to be pulled out to stop the spindle, so you may have to make some adjustment to the settings as has been mentioned. Energy goes as the square of the RPM, so it goes up fast with increasing RPM. You may even have to accept a slower stop to get it to work at higher RPMs.
 
Hi
OK thanks for the pointers. It will take me a day or so to work through them.

Here is a photo of the braking resistor and heat sink. The rated power dissipation of the resistor can only be achieved if it is fitted with a heat sink.
I made the heat sink with more capacity that required.

IMG_9013.jpg

Dazz
 
Hi
According to the Manual in the section of parameter C1-09 Fast Stop time

NOTICE: Rapid deceleration can trigger an overvoltage fault. When faulted, the drive output shuts off, and the motor coasts. To avoid this
uncontrolled motor state and to ensure that the motor stops quickly and safely, set an appropriate Fast-stop time to C1-09.

This is exactly what I saw happen. I would have thought the stall protection with braking option would have prevented faulting.

Dazz
 
It probably is helping, which would be the case if you find that the resistor heats up after some stops. At some point, there is just too much energy to get rid of, however. Then either parameters need adjusted to get maximum effect, or else you may just have run out of capability.

Make sure you have the maximum effect that resistor can produce, though. Usually they do make a very obvious difference.
 
What they mean, is that with high inertia loads it may take longer to stop and not cause a fault code. In some safety systems they use a mechanical stop which is an emergency stop to prevent/minimize physical injury, in some cases this can cause damage to the machine at the expense of faster stopping.

As far as braking resistors, I have yet to have any external lathe braking resistor even get past slightly warm with routine use. All you should need to achieve the watt rating is mount it to a metal surface to conduct some of the heat away from the resistor.

Just because you have an external braking resistor doesn't mean that you will not get an over voltage/current error. It is a function of adjusting the parameters mentioned to achieve quick stops. If you need an emergence stop you can also use the lathe foot brake. Braking frequency can also effect the rate of stopping, there is a ratio to run time vs. braking time.

I have a V1000 with an external brake resistor on my lathe and was working at 1400 RPM with a moderately heavy chuck the other day, with a 1 second stop it went into a free run with an over current bus fault. I always warn people that the braking with a VFD is not always 100% reliable. I try to set the parameters conservatively, and warn people about fast stops with high inertia loads.

The V1000 has parameters to minimize this fault and adjust the braking time to a controlled stop. I do not use this feature because I use an electronic stop system and activating this system will effect the final cutter position when stopping.

Nice install, but I would just mount the braking resistor to a metal surface and skip the bolt on wings.
 
Hi

I have played around with settings and none do what I want. I want e-stop to stop the spindle in the shortest time regardless of load (inertia).

I have a 56ohm resistor which can theoretically take ~2.5kW off the DC bus. More than that will exceed the nominal DC bus voltage. If I can set the drive to achieve maximum reverse current (= torque), then the spindle will stop in the shortest time.

I have found DriveWorks EZ. I have had a look at it, and I think it has the potential to produce a custom e-stop function that does what I want. The blocks include a PI controller. I may be able to use that to actively control torque/current during an e-stop.

Dazz
 
Do you have any actual proof that the resistor is taking power?

Many things would be explained if the system were not actually working as intended......

If it is not on a heatsink and you stop a few times, until it actually changes temp, then you would at least know it is working.

Same if you can check the voltage across it. Should be a pulse during a hard stop. It will NOT do anything UNLESS the DC bus goes up over a limit. That limit is usually settable, but it appears that it may not be on your drive, stated to be 410V approximately, which is OK, but a tad lower than it could be.

You have the option of DC injection braking, which is a little rough on the motor if done often, but which stops nicely.

Look at "stall prevention selection during deceleration", parameter L3-04, page 386 and following for some other relevant parameters.. Setting L3-04 to "1" seems to stop as fast as possible without an "OV" error, it simply stops decel until voltage comes down and then resumes decel.
 
Do you have any actual proof that the resistor is taking power?

Many things would be explained if the system were not actually working as intended......

If it is not on a heatsink and you stop a few times, until it actually changes temp, then you would at least know it is working.

Same if you can check the voltage across it. Should be a pulse during a hard stop. It will NOT do anything UNLESS the DC bus goes up over a limit. That limit is usually settable, but it appears that it may not be on your drive, stated to be 410V approximately, which is OK, but a tad lower than it could be.
I have tested the resistor. It measures 56ohm when disconnected from the drive.

You have the option of DC injection braking, which is a little rough on the motor if done often, but which stops nicely.
I have that set to engage as low rpm. It is set to be reasonably soft on the motor.

Look at "stall prevention selection during deceleration", parameter L3-04, page 386 and following for some other relevant parameters.. Setting L3-04 to "1" seems to stop as fast as possible without an "OV" error, it simply stops decel until voltage comes down and then resumes decel.
I have L7-03 and L7-4 set to 3, stall prevent enabled.
L8-01 set to 0, Brake resistor protection disabled.

Unless I have misunderstood the parameter descriptions (entirely possible), I think I have the settings doing what I what to achieve and doing what they can achieve, except for the e-stop.

Monitors
U1-08 Output Power and
U1-09 Torque Reference
both provide negative values for regen.

Preliminary testing/analysis indicates the regen power is linear. If I can link control to C1-09 e-stop time parameter, I might be able to keep the DC bus voltage within limits. I should also be able to use the bus voltage as a control input. This is all speculation. I need to do more testing and try some different things.

Dazz
 
Check that there is actually DC voltage on the resistor during a brake. It's possible the braking transistor is fried.

I haven't seen a mention of what size the drive is - is 56 ohm appropriate?
 
Check that there is actually DC voltage on the resistor during a brake. It's possible the braking transistor is fried.

I don't see any monitored variable for the braking transistor enable. I will need to measure directly off the resistor.

I haven't seen a mention of what size the drive is - is 56 ohm appropriate?

It is a V1000 220VAC/1ph/3.8kW version, driving a 4kW 3ph motor.
From what data I have found, 40ohm is the minimum spec so 56ohm should give a wide safety margin.
 
Would it be silly to ask why are you E Stopping it?

That is not a controlled stop


I would 'think' as you attached a lower value higher wattage resistor it could sink more current.

A big lathe with a big chuck is a lot of rotating energy that needs to be turned to heat very quickly
 
Would it be silly to ask why are you E Stopping it?

........

if you want an E-stop you want it to STOP NOW. The last thing you want is to have your neck or hand become the mechanical brake because the e-stop caused a fault-out and refused to stop, letting the unit coast down.

It is a sensible test.

Of course, NO the E-stop is not necessarily any different from regular stop, because it may be just a latched switch in-line with the regular stop button. It MAY be wired separately, to a different input, in which case it can have different behavior.

If it is programmable, I'd want the e-stop to slam in full DC injection braking and anything else it has. E-stop is not a case of trying to be easy on the equipment, it is a case of getting the damn thing stopped as soon as possible. Otherwise it is not an "E" stop that you want.
 
Hi
I think e-stop is always going to be a compromise. You do want the kinetic energy to be zeroed on the machine, but without damaging the machine. You don't want someone hesitating to hit e-stop because they are worried about breaking the machine.

I did some testing today.
I connected a multi-meter across the braking resistor. It is definitely taking load. It can also be seen in the data recordings attached.

The aim of the testing was to identify a signal that can be used to regulate the bus voltage. The prime candidate is output frequency. If the output frequency approaches the actual motor frequency to reduce slip, the reverse power that is raising the bus voltage will stop rising. Zero slip angle = zero reverse power.

When the motor is stopped, the output frequency ramps down to zero. At the same time the bus voltage ramps up until the braking resistor is engaged, and flattens the peak. The resistor is 56ohms but the vfd controls the resistance with switching. If the switch conducts to the resistor 50% of the time, the resistor will look like it is 56 / 50% = 112ohms. Switching makes a fixed resistor look like a variable one.

Only a couple of channels of data are recorded because the data is obtained in real time over a 9600bit/s serial port. The sample rate is well below the Nyquist sampling frequency, so the data is likely to be missing key information. For example, the braking resistor switching transients are not shown. It is also difficult to identify the precise moment the over-voltage threshold was exceeded.

This shows the results of a normal run fwd, stop, then an e-stop. The e-stop tripped the drive.
Nardini-MS 350 Trend Report 06.jpg

This shows a normal run and stop at high rpm and high inertia. The drive close to tripping after the stop command.
Nardini-MS 350 Trend Report 10.jpg


At 67.5Hz, the e-stop works as expected, as shown in this report. If the frequency is raised to 70Hz, the e-stop trips the drive. At 67.5Hz, the drive is just managing to complete an e-stop.
Nardini-MS 350 Trend Report 12.jpg

The charts show there isn't much difference between a normal stop and a tripped e-stop. I suspect it will be difficult to implement a PLC type control function to reliably complete a fast shutdown. Part of the problem is the noise not shown in the charts.

The control algorithm for the braking resistor switch will know when it can't cope with the reverse power. Making use of that would require Yaskawa to change the firmware. Unlikely to happen any time soon.'

I plan to experiment with the PLC functions but at present, I am not confident I can make it work based on the test results attached. Any advice/feedback welcomed.



Dazz
 
I suspect it is NOT the 67 vs 70 Hz that is the issue, it is the SPEED corresponding to that frequency, and the energy difference. Remember, it goes as rpm squared.

However, the setting parameter I mentioned earlier looks good, it appear to avoid the "I give up" response of simply "tripping" and instead if voltage gets high it cuts back the braking until voltage goes down, then resumes. That will stop faster than the "I give up" followed by coast down.
 
For an emergency stop, I think I would have to agree with the earlier comment that DC braking might be a better option.

Yes, it dissipates heat in the windings. But this is an infrequent occurrence, and a lathe driven by a VFD is going to be generally quite gentle on the motor anyway - little prolonged operation at full load, and no direct-on-line starts.

The heat dumped into the windings is probably far less than in a direct-on-line start of the same inertia and speed anyway.

However, the drive manual seems to have some caveats and conflicts between various features:

DC injection braking can't be used with open-loop vector (p125) - could this be your issue?

High slip braking, again V/f control only (p241). Similar to DC braking except at a frequency above zero, I think.

Overexcitation deceleration, usable with both V/f and OLV, but says don't use with a braking resistor (p241).

Whatever braking option you choose, IMHO you should really run it up to above the maximum intended inertia and speed and test it repeatedly. Having the OK vs Fault line between 60 & 67 Hz is not good if you then use a heavier chunk of steel...

It would be interesting to see the duty cycle or average current on that braking resistor. Table E1-01 on page 169 implies that the default setpoint for turning on the braking transistor appears to be higher than for reducing acceleration, which is... odd. Those parameters might be worth looking into.
 








 
Back
Top