Yaskawa drive questions

Status
Not open for further replies.
Good morning all,
Got onto a project yesterday involving a 4000watt laser that uses a Yaskawa F7 varispeed drive controlling a 20 hp motor for the turbo induction system. Machine will run for eight and a half minute as normal and then shuts down. Not sure what the time has to do with anything just trying to provide as much information as possible. After the shut down the drive has GF alarm on the display. Research on Yaskawas web states this is a ground fault issue. We have megged all related wiring to the drive and from the drive to the motor which all have passed. To continue to narrow the problem down we removed the drive in question and installed it on an identical machine. The problem of shutting down the machine again in eight in a half minute followed the drive. Currently it's off the a repair shop who is scratching there heads because they can't get it to fault out.
My first question is what can I tell them to look for or has any of you seen this problem and know what to suggest.
Secondly, did I jump to soon at blaming the drive? What else should I consider checking.
Third, I have located a American version of this drive in Chicago but the spec. show it only comes in 208 to 480 volt three phase. from the part number below the only difference is the letter "A" is replaced with a "U" which I have learn stands for Asian opposed to United States. Can I make this United State version work?
Lastly, the programming pendant is removable from the drive can I assume that it can be removed and plugged into a new drive and the programming parameters are present for the new drive? Are the programmed parameters stored in this pendant or the drive itself.
I was fairly confident when the problem moved with the drive. below I have listed the model of the drive in question.

CIMR-F7A2015IE

Thanks
LHarrington
 

StephenSDH

Senior Member
Location
Allentown, PA
Do you have an ungrounded service. Some drives have a jumper that needs to be removed, my understanding is you can get false ground faults if you leave it in. I found the user manual online, it doesn't mention anything about a ground jumper or about ungrounded systems.
 

foqnc

Member
Lastly, the programming pendant is removable from the drive can I assume that it can be removed and plugged into a new drive and the programming parameters are present for the new drive? Are the programmed parameters stored in this pendant or the drive itself.

Any Parameters entered will be stored on the onboard CPU and not the pendant. I think manufacturers are making pendants removable for safety reasons ie preventing un-authorized button pushing.
As far as a new drive, any 20Hp Drive will work, if you have a list of the parameters from the old one, you just need to cross reference them to the new one.
Good Luck
 
Its been a while since I've played with these drives. They were usually pretty reliable but sometimes could get temperamental... the fact that the problem followed it is a good indicator. Did you try the motor you swapped out in the machine with the original problem to see if it solved the GF error?
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
I agree that if the problem followed the drive, it's the drive.

If you thoroughly check the circuit it and it was working fine but it no longer is, then chances are one or more of your Hall Effect transducers (the DC CTs) has gone belly up. The GF circuit in VFDs is typically a Residual Current method where they would use the vector sum of the 3 phases and measure the zero sequence current flowing to an internal Wye point on the sensors. In VFDs, they use Hall Effect transducers because the output is not really AC, it is pulses of DC from transistors. So to measure output current in each phase, they have a DC transducer on each of the 6 transistor circuits and then in digital systems like VFDs, complex calculations are used to create a "virtual" RCD circuit in the microprocessor. If one of the DC CTs is failing, it is reading the wrong amount of current in that part of the circuit which will skew the Zero Sequence current calculation in the microprocessor and give you a GF indication even if one does not really exist.

The puzzling part is that your test shop cannot recreate it (if I read you correctly). The 8 minute issue would lead me to believe that something is heating up and causing the failure of the Hall Effect circuit, maybe when they are testing it they are not adequately recreating that condition.
 

skeshesh

Senior Member
Location
Los Angeles, Ca
I agree that if the problem followed the drive, it's the drive.

If you thoroughly check the circuit it and it was working fine but it no longer is, then chances are one or more of your Hall Effect transducers (the DC CTs) has gone belly up. The GF circuit in VFDs is typically a Residual Current method where they would use the vector sum of the 3 phases and measure the zero sequence current flowing to an internal Wye point on the sensors. In VFDs, they use Hall Effect transducers because the output is not really AC, it is pulses of DC from transistors. So to measure output current in each phase, they have a DC transducer on each of the 6 transistor circuits and then in digital systems like VFDs, complex calculations are used to create a "virtual" RCD circuit in the microprocessor. If one of the DC CTs is failing, it is reading the wrong amount of current in that part of the circuit which will skew the Zero Sequence current calculation in the microprocessor and give you a GF indication even if one does not really exist.

The puzzling part is that your test shop cannot recreate it (if I read you correctly). The 8 minute issue would lead me to believe that something is heating up and causing the failure of the Hall Effect circuit, maybe when they are testing it they are not adequately recreating that condition.

Would a sensing fault not cause a ground fault detection much ealier than 8 minutes?
 

jim dungar

Moderator
Staff member
Location
Wisconsin
Occupation
PE (Retired) - Power Systems
Would a sensing fault not cause a ground fault detection much ealier than 8 minutes?
Not if it is heat related.

The test shop probably has the drive on their test bench in open air rather than stuck in a poorly ventilated enclosure. Test shops often have a problem duplicating environmental conditions.
 

skeshesh

Senior Member
Location
Los Angeles, Ca
Not if it is heat related.

The test shop probably has the drive on their test bench in open air rather than stuck in a poorly ventilated enclosure. Test shops often have a problem duplicating environmental conditions.

I could see that, but again are CTs known to have problems that are caused by heating up to failure in a matter of minutes? I would think that a failure would happen either abruptly because of extreme heat or because of being subjected to excessive heat over a period of time until a critical failure is reached. What condition causes the sensor to function properly then get to a point of failure in these 8 minutes?
I would love to spend a some time on a good read on this topic if you have any suggestions.
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
I could see that, but again are CTs known to have problems that are caused by heating up to failure in a matter of minutes? I would think that a failure would happen either abruptly because of extreme heat or because of being subjected to excessive heat over a period of time until a critical failure is reached. What condition causes the sensor to function properly then get to a point of failure in these 8 minutes?
I would love to spend a some time on a good read on this topic if you have any suggestions.
A bad solder connection. Theses are not big CTs in the conventional sense, they are little transducers, maybe even PC board mounted, that are sending out mV signals. A weak solder joint could be fine until there is some significant thermal stress, then it starts to lose integrity. From a sensing and calculating standpoint, that is noise that may end up being interpreted as a severe current imbalance, translated to being a ground fault.
RD180735-01.jpg


Probably closer to what Yaskawa would have on a 20HP drive.
http://www.cy-sensors.com/CYHCS-B1-50A.pdf
CYHCS-B1.jpg
 
Last edited:

skeshesh

Senior Member
Location
Los Angeles, Ca
A bad solder connection. Theses are not big CTs in the conventional sense, they are little transducers, maybe even PC board mounted, that are sending out mV signals. A weak solder joint could be fine until there is some significant thermal stress, then it starts to lose integrity. From a sensing and calculating standpoint, that is noise that may end up being interpreted as a severe current imbalance, translated to being a ground fault.
RD180735-01.jpg


Probably closer to what Yaskawa would have on a 20HP drive.
http://www.cy-sensors.com/CYHCS-B1-50A.pdf
CYHCS-B1.jpg

Yes I could see that happening. I just don't know how often this kind of failure happens - do bad solder connections not fail typically unexpectedly at first due to either one critical condition or an accumulation of excess conditions follows by immediate failure when operated the next time?
I'm just asking cause I want to know but this case seems to fit the scenario, especially taking into consideration the fact that the shop was not able to recreate the fault. Maybe the OP can elaborate a bit more on installation conditions and the possibility of heat build-up.
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
Yes I could see that happening. I just don't know how often this kind of failure happens - do bad solder connections not fail typically unexpectedly at first due to either one critical condition or an accumulation of excess conditions follows by immediate failure when operated the next time?
I'm just asking cause I want to know but this case seems to fit the scenario, especially taking into consideration the fact that the shop was not able to recreate the fault. Maybe the OP can elaborate a bit more on installation conditions and the possibility of heat build-up.
Common? No.

But Murphy rules my world...
 
Good morning all,
Thanks to everyone who has commented thus far, we learned at this point my drive, because it a Asian version of the Yaskawa Drive and not the United States version couldn't be repaired here. This is noted in the part number by an "A" or a "U". We can by anyone of the components in this drive here in the States but because this is a specific Drive for a Laser the OEM has proprietary software loaded on this drive that prevent us to make any of these replacement parts work. Our bad drive has a faulty main PCB board. we received a new board and tried to program it and that?s when we learned we can't achieve the 750hz required to drive our blower. We could only get to 400hz. After pick the brains of Yaskawa in Chicago, Canada and anyone else that could speak, we mention who it was for all the lights came on and I was informed this has been the case for this particular OEM before. Really stinks being force to go back to the OEM to buy the new drive at three times the face value and hire their service tech to come in because they wouldn't sell me the drive with out him or honor a warranty.
One post has requested my feedback on the cabinet or installation conditions. It very clean environment because of the positive air pressure in the enclosures but it is at least 20 degrees warmer than ambient in the cabinet.
As of this morning I?m waiting on this tech to arrive so him and I can spend 10 minutes installing this new drive. Very, very, very, frustrating.

thanks again everyone
LHarrington
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
One thing Yaskawa has always done, and which is a key component to their success, is to make their internal designs "open" to partners for modification and customization. Years ago one of the Yaskawa "partners" in the US came up with a killer algorithm used in the Yaskawa inverters for doing what is called "torque proving" on a closed loop vector drive so that it could effectively replace DC drives and motors on hoists, a near impossible accomplishment at the time. Being that there were at the time 5 or 6 different partner / vendors of the same drive product, creative people would buy one of the versions with the algorithm in it and clone the EPROM so it could be used on the other brands. Yaskawa had to come down hard on that issue, because the people who developed the algorithm had spent a LOT of money doing so and threatened to sue Yaskawa if they didn't do something to prevent it. Ever since then, Yaskawa has been very vigilant on the issue of proprietary alterations.
 
Status
Not open for further replies.
Top