Thread: Help me Rewire A Motor From Star To Delta and To Connect the VFD in Melbourne

Originally Posted by Jayson

Just checked and it looks like I've been duped... Oh well...

20150617_172422-800.jpg20150617_172604-800.jpg

Stand by, I have contacts at Powtran. I will send them an email now and see what the reaction is.

Hi Chris,

Mine wasn't from Powtran. It is a Huanyang VFD

No worries.

Originally Posted by Jayson

Just checked and it looks like I've been duped... Oh well...

Maybe it's just an issue with your multimeter? Are you running it at over/around 100hz?

Originally Posted by Jayson

Just checked and it looks like I've been duped... Oh well...

20150617_172422-800.jpg20150617_172604-800.jpg

But you cannot measure VFD output voltage with a digital multimeter. Even if that multimeter was a true RMS tye, it would just show meaningless garbage figures, due to its limited bandwidth and the RF noise created by the VFD. The VFD output is a non-sinusoidal, pulse width modulated voltage of several kHz frequency. That meter you have there may show a voltage easily 50% higher than it is in reality. To measure VFD output voltage you need an oscilloscope, or a very specialized digital multimeter. Or in a pinch you can use an old analog multimeter, but it has to be one with very low Ohm per Volt impedance (less thn 20 KOhm/Volt) else it will not be much better than a digital multimeter. If you have an oscilloscope, that would be your best option. If you are not into electronics or do not have a mate that is and do not want to spend 1k$+ on suitable measuring equipment, it is fair to say that the output voltage of a VFD cannot be measured at all.

What you could measure with that multimeter you have is the DC bus voltage, and that would be about 1.41 times the input voltage. But again, if you are not into power electronics it may be a bad idea to try this, as the bus voltage would have to be accessed inside the VFD. One false move and it starts getting life threatening. Not long ago we had a suicide candidate guy that was testing the board of a DC variable speed controller on a steel benchtop..... most fortunately he only cooked the board..... and not himself.

Originally Posted by cba_melbourne

But you cannot measure VFD output voltage with a digital multimeter. Even if that multimeter was a true RMS tye, it would just show meaningless garbage figures, due to its limited bandwidth and the RF noise created by the VFD. The VFD output is a non-sinusoidal, pulse width modulated voltage of several kHz frequency. That meter you have there may show a voltage easily 50% higher than it is in reality. To measure VFD output voltage you need an oscilloscope, or a very specialized digital multimeter. Or in a pinch you can use an old analog multimeter, but it has to be one with very low Ohm per Volt impedance (less thn 20 KOhm/Volt) else it will not be much better than a digital multimeter. If you have an oscilloscope, that would be your best option. If you are not into electronics or do not have a mate that is and do not want to spend 1k$+ on suitable measuring equipment, it is fair to say that the output voltage of a VFD cannot be measured at all.

Many CRO's don't have sufficient V/division to reach mains power V's levels.

Originally Posted by BobL

Many CRO's don't have sufficient V/division to reach mains power V's levels.

Bob, that is correct. That is why they make 10:1 and 100:1 and 1000:1 probes to go with scopes.

If you take a scope with a maximum of 5V/division, and add a 10:1 probe, it becomes 50V per division. With 10 divisionss on the screen, you can display a full 250V sine curve, going 250V positive to 250V negative. If you add a 100:1 probe, each division becomes 500V. There are switchable probes too, so you do not have to buy 2 or 3 probes.

Be careful with scopes that do not have fully insulated input stages when measuring VFD outputs, you may end up with line voltage on the scope chassis. Fully insulated handheld scopes are the ideal tool to work on VFD's.

Originally Posted by BobL

Many CRO's don't have sufficient V/division to reach mains power V's levels.

Bob, that is correct. That is why they make 10:1 and 100:1 and 1000:1 probes to go with scopes.

If you take a scope with a maximum of 5V/division, and add a 10:1 probe, it becomes 50V per division. With 10 divisionss on the screen, you can display a full 250V sine curve, going 250V positive to 250V negative. If you add a 100:1 probe, each division becomes 500V. There are switchable probes too, so you do not have to buy 2 probes - like this one, the longish black slider selects between 1:1 and 10:1
switchable_probe.jpg

Be careful with scopes that do not have fully insulated input stages when measuring VFD outputs, you may end up with line voltage on the scope chassis. Fully insulated handheld scopes are the ideal tool to work on VFD's.

I just borrowed a mates new DSO and went to measure the voltage, turned on the VFD, hooked up the little alligator clip to one of the phases and boom, flash, #$#@....

Pointing my friend to this thread so he can see it wasn't my fault......

Lots of information has been given in this thread that could result in the above.

So, lets just leave it at no VFD will give 380/415 output from 240 in, this seems to be the consensus from those that know.

Going out to the shed to use my lathe...

Originally Posted by cba_melbourne

Be careful with scopes that do not have fully insulated input stages when measuring VFD outputs, you may end up with line voltage on the scope chassis. Fully insulated handheld scopes are the ideal tool to work on VFD's.

Honestly, you'd want to be pretty brave/competent to go scoping the output of a VFD - for starters, the output is probably floating, meaning you really want to differential probe across 2 scope channels (or use a differential probe), and even with 10:1 or 100:1 you really want a decent Cat rating on the probes - having a high divider ratio on a probe does *not* mean it's therefore designed for high voltages.

I'd expect measuring the voltage with the load attached would go some way towards dealing with the harmonic content. Also I think the DC bus is sometimes readily available for connection to a braking resistor? Again, though, I'd be really careful measuring high DC voltages with a DVM that's not properly designed for high-energy sources.

Originally Posted by Jayson

So, lets just leave it at no VFD will give 380/415 output from 240 in, this seems to be the consensus from those that know.

While I've never met one, there's no reason they can't exist, you'd just need an intermediate inverter stage (i.e. a DC-DC converter) to boost the DC bus voltage.

Originally Posted by RustyArc

Honestly, you'd want to be pretty brave/competent

I have a 20V/Div CRO with 10 times CAT 1 500V probes but I'm pretty sure it doesn't have isolated inputs and I'm neither brave or competent
I assume 500v wouldn't really be considered good enough?
If I had an isolation transformer large enough to run the VSD on would that remove the isolated input problem?
Not that I have any plans to do such ATM just interested.

Originally Posted by Jayson

So, lets just leave it at no VFD will give 380/415 output from 240 in, this seems to be the consensus from those that know.

Want to bet?

Stuart

Originally Posted by RustyArc

While I've never met one, there's no reason they can't exist, you'd just need an intermediate inverter stage (i.e. a DC-DC converter) to boost the DC bus voltage.

I believe(though cant recall where I got the info from) that my VSD uses a voltage doubler* on the mains side then uses a standard 415V VSD.

As I said earlier but have never put a meter on it...... but I have run the same motor on a 240V in/out VSD an you can certainly tell the difference.

Stuart

*If I recall correctly the person that supplied said info was not all that impressed with the design of said doubler, but so far it seems to be working just fine.

Originally Posted by Stustoys

I believe(though cant recall where I got the info from) that my VSD uses a voltage doubler* on the mains side then uses a standard 415V VSD.

As I said earlier but have never put a meter on it...... but I have run the same motor on a 240V in/out VSD an you can certainly tell the difference.

Stuart

*If I recall correctly the person that supplied said info was not all that impressed with the design of said doubler, but so far it seems to be working just fine.

Drives Direct in the UK sells a range of 240V to 415V inverters. They seem to use a voltage doubler at the input stage. Drives direct must have had them manufactured on order, as no one else seems to sell them. Here is a video that shows the insides. I think it is an earlier model, not exactly the ones they sell now:

https://www.youtube.com/watch?v=VnLRhEHVLxA

Towards the last third of the video, the drive is shown inside. At the top left is the voltage doubler stage, with the capacitors. The guy uses a Fluke 867B to measure the VFD voltages - that instrument is ideally suited for the task.

Given the steep price of thes Drive Direct units, I wonder if it would not be cheaper to use a normal 400V input VFD and feed it from a 240V/400V step-up transformer up front.


Here some good information on measuring (safely!!!) on VFD's:
http://www.testequipmentdepot.com/ap...289-dmm_an.pdf
http://en-us.fluke.com/community/flu...o-a-motor.html
http://assets.fluke.com/AppNotes/Ele...r/GO416b_u.pdf