Please share some guidance : Installing VFD on PIMMs

[BUBBA271]

I've used Yaskawa brand vfds with alot of success and reliability. I have three on a dough press at work.

 

[ajay34]

if you go for fuji or yaskawa...than u no need for encoder..'as far i had use yaskawa..u not need feed back from motor to to determine the speed if you need to do that u might need to consider to place pulse sensor or promity sensor and feed back the inpout to VFD.than do setting on VFD as set point and feed back min n max,..

normal VFD the speed are determine from the HZ setting that you set ..
from potenial meter (V/R) or internal control HZ setting ..
fuji and yaskawa or any other VFD always does have voltage ,hz and amp as output to display ..

 

[jiro.iro]

Hi,
How about hitachi SJ-700 or mitsubishi, any experience?

Thanks.

 

[BUBBA271]

No on either of them. So far i like yaskawa, and Vacon x4 series. Vacon purchased tb woods drive part of the corporation. I have many machines at my job that run with these. I'm talking about a hundred or so and a few still have the original e-trak model in their cabinets. The vacon is easy to configure and yes cheaper than allen bradley.

 

[ajay34]

i agree with bubba..vacon are very easy to configure..
as for mitshubshi..yes it also easy to configure but it same as yaskawa you need to do the parameter seeting as well..i notice both VFD does save similiar parameter ..so not much diffrent ..the difrrent was on
analog input ..coz on yaskawa u can share to common but for mitshu
u cant share the common for digital or analog input

 

[tommys1210]

Do not be tempted to measure anything, yes, that's anything, on the leads between the VFD and the motor.

First, There isn't a piece of test equipment made that will accurately report Hz, effective voltage, and effective current on the complex, high frequency pulses that the VFD sends to the motor.

With all due respect I am Having trouble accepting this. If this is true how do the manufactures check there drives before shipment? I have seen many drives field service people checking drives with O-Scopes on site?

I myself have but a clamp on on the output leads of drives and the current reading is in most cases in agreement with the drive with about a half amp of accuracy.

What is in the drive internal that make it able to read correctly if there is no test equipment that will do it?

Second, it is hugely unsafe to contact the motor leads due to the voltage being far higher that expected. For example, in a 460V system, the voltage on the motor leads can be in excess of 1400V. In a 400V system that would be over 1200V. And you are trusting a CAT III 1000V meter!!!!.

I have measured many drives with a Fluke and never seen this?

How would you have 1200 or 1400 Volts and if you did i would think it would be closer to or at the motor if it is due to Ring?

 

[mhulbert]

I have used portable oscilloscopes to measure VFD leads. It works fine, and is a handy tool for learning how a drive works. I measure the voltage at the drive terminals, and have not seen voltage spikes that high.

In fact, any True RMS meter that has a frequency response at least twice your carrier frequency will measure voltage and current (with a CT) just fine.

 

[DickDV]

In the classes I teach, I routinely use a good Fluke scopemeter to illustrate the output waveforms. These are accurate and are useful for test and evaluation work as long as they have adequate voltage ratings.

However, I also use this same excellent (expensive) scopemeter to illustrate the huge error between the scope-displayed voltage and hertz compared to the drive keypad-displayed values. I have sampled dozens of clamp-on ammeters on the drive output and have yet to find one that can be trusted. I'd have to see it to believe it!

It is true that the drive only produces 660V high pulses on a 480VAC supply but, by the time those pulses proceed as little as 100 feet down the wire, they are already over 1000V. If that isn't true, then why are MG1 Part 31 motors now being tested to 1800V?

As to how the drive can get it right when expensive test equipment can't, the answer is in the fact that the values the drive displays are calculated from the DC bus and the firing signals to the IGBT's. These are unavailable to field test equipment.

For those of you that are interested and have a good scope available, take a small PWM VFD (it doesn't matter what brand), connect it to a small motor with 100 ft leads or longer, and look at the waveform first at the drive terminals and then at the motor end of the leads. Prepare to be surprised at how wild the pulses have become. And, note the peak value of the pulses. And then be prepared to put your 1000V rated test gear away!

And, if you somehow find that the voltages are still under 1000V, you may be onto something BIG! Motor manufacturers who are still struggling with insulation issues on inverter duty motors will flock to your door to figure out how that happened. It could be a fast way to get rich!

 

[DickDV]

One other comment on measuring drive output voltage and frequency: I've heard of little RC networks that can be built to filter the output pulses and produce a measureable voltage and frequency reading but it requires some good understanding of filter design to get it right. On another BBS that I frequent, there are a couple of European drive techs that often work on 1500kw drives and larger, that have described these filters.

As for me, I intend to read the keypad and go with that!

 

[Lancie1]

Tommys,

Heed and follow DickDV's advice. He is the drive expert here, and is giving you wise advice. Don't try to measure the current and voltage downstream of a VFD, except as an indication of "Yeah, there is some output there", or "No, there is nothing coming out".

 

[tommys1210]

All I can tell you is that we have hunderds of powerflex 700 drives and siemens simanics drives and I can put my Fluke 337 amp clamp on at the drive and it's reading is within 1 amp of the drive HIM display when reading right at the drive.

Never tried it doen the line. AS for the 1400 volts at the motor that is fixed by a sine filter. I have rins over 300 feet using THHN and I will bet my life you wont get anything higer than 500 volts at the motor.

But sine filters are not cheap but if you want to prevent corona damage to you motor leads and reduce Noise then it is the only way to go. We use MTE brand and Block Brand sine filters.

When using sine filters you do not have to use vfd cable although it is still a good idea if budget permits and you also do not ned shaft grounding rings and EMI filters.

We use K rated isolation transformers to keep the harmonice off the supply system and each drive cabinet has it's own transformer.

 

[tommys1210]

While I do know the amp clamp will never be 100% accurate like the Drive HIM it is a very good troubleshooting tool and it works well enough for that purpose.

From some of the posts a person with no experience would take that it would not work at all and that is simply not true.

For troubleshooting general workings of a drive and motor within a 1/2 amp is god enough for me but I can't speak for anyone else.

 

[ajsimione]

I strongly recommend that you do not attempt this project. I work in an injection molding shop. We have some machines which have variable displacement pump systems. They are complicated and costly. They require special knowledge to design, implement, and maintain. Most major machine OEM's are just now, in the last few years, combining VFD controlled pump motors and variable displacement pumps to good effect.
The systems we have on our machines combine a pump, transducers, and a dedicated control board that is seperate from the machine PLC. This is not easy stuff. I recommend talking to a reputable machine supplier in your area. New equipment is the best solution here, whether it's an electric, hybrid, or pure hydraulic. New machines are more energy efficient, capable, and easier to use and maintain.

 

[jiro.iro]

Hi guys,
Bubba271, ajay34, thanks for your feedback on inverters. I didn't know Vacon brand until just now. I'll take a look.

Tommy1210, mhulbert, dickDV, lancie1, thanks for sharing your info and experiences. Since it also relates to safety issue, we will be very careful. Is it safe to say DickDV's concern is that the reading at the motor end when the lead from the VFD and the motor exceeds 100feet long will be far greater than what's shown on the VFD?

Ajsimione, from your experience, can adding only a VFD to an old PIMM be justified on the purpose of electricity saving?

Thanks a lot.

 

[DickDV]

tommys1210, I don't want to be confrontational about measuring on the motor leads but let me respectfully point out why you think you are seeing things differently than in most applications.

First, the use of a sine filter on very long motor leads is necessary but on midrange and short motor leads, a sine filter is simply a waste of money. For most troublesome applications, a motor lead reactor or, at most, a dv/dt filter is all that is necessary. In my 22 years of drive work, I have used a sine filter only once and that was on a job where the motor was 800+ feet from the drive. Of course, if you have stock in MTE or TCI, then it would be a desireable thing to do. But, overkill otherwise.

Second, your statement about not needing shielded motor leads or EMI filters reveals a lack of understanding of their purpose. Shielded motor leads and EMI filters do nothing to reduce ringing and reflected waves on the motor leads. They simply protect surrounding sensitive equipment from radiated electrical noise from the drive output pulses. And, they make common noise problems worse by increasing the capacitive coupling between the motor leads and ground.

Third, your statement that you would bet your life on there being no voltage over 500V at the motor could cost you your life tonite yet! 460V in pure sine wave form reaches 660V twice per cycle. At the drive output, every pulse is 660V high since the DC bus (660V) is chopped into narrow pulses to make the PWM output. And that is assuming that the pulses are perfectly clean with no ringing, overshoot, or other noise. All rather foolish assumptions, I would say.

But, tommys1210, I mean no offense with any of this. I encourage you to get a good 20meg or higher oscillascope with 2000V leads and check this out for yourself. I think you will see for yourself that my claims are pretty accurate. Check before and after the sine filter too. You will see that they also are pretty good but not perfect. And, if you can get your hands on 200 feet of shielded motor leads, check the drive pulses at the drive end and then at the motor end. You will see that the claims of reducing motor insulation stress are "snake oil".