How harmonics affect a drive system with caps versus diode rectifiers.

Status
Not open for further replies.

aelectricalman

Senior Member
Location
KY
So, in past conversations I have mentioned a particular drive system which was at 4160V having exploded due to harmonic issues inflicted on the system from a utility source. The capacitors in the drive unit for this 2000 HP drive ruptured (through routine maintenance) and the debris was sucked through the system by a line side vent fan. The result was the transformers, capacitors and for that matter, the building, burned.

The old Harmony Robicon VAC drive was replaced with a Toshiba T300MVI system. This new system operates using diode rectifiers and eliminates the capacitors.

The harmonic source that is affecting the system in only elevated outside of IEEE 519-2014, on the 5th order (6-8% average) and 7th order (3-5% average).

My question is, understanding that this could destroy a capacitor arrangement if a parallel resonance exists...

what can happen to the new type of system. How tolerant are diode rectification techniques in drive systems? Should I worry? Also, are traditional remedies (reactors) the best way to mitigate harmonics for this type of drive system?

Thanks for your help, as always.
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
So, in past conversations I have mentioned a particular drive system which was at 4160V having exploded due to harmonic issues inflicted on the system from a utility source. The capacitors in the drive unit for this 2000 HP drive ruptured (through routine maintenance) and the debris was sucked through the system by a line side vent fan. The result was the transformers, capacitors and for that matter, the building, burned.

The old Harmony Robicon VAC drive was replaced with a Toshiba T300MVI system. This new system operates using diode rectifiers and eliminates the capacitors.

The harmonic source that is affecting the system in only elevated outside of IEEE 519-2014, on the 5th order (6-8% average) and 7th order (3-5% average).

My question is, understanding that this could destroy a capacitor arrangement if a parallel resonance exists...

what can happen to the new type of system. How tolerant are diode rectification techniques in drive systems? Should I worry? Also, are traditional remedies (reactors) the best way to mitigate harmonics for this type of drive system?

Thanks for your help, as always.
I'm not sure you have all of your info right, or I am not understanding what you are asking. The Robicon Perfect Harmony VFDs are not substantially different from the TMIC (Toshiba) T300MVi as far as the rectifier / inverter topology goes. BOTH of them use diodes for rectification into a DC bus that has capacitors in it. Neither of them should be particularly more susceptible to line side harmonics as the other, in that they both just rectify the AC to DC anyway.

So if someone told you that your Robicon drive was damaged "by harmonics", then there are one of two possibilities: A) they didn't know what they were talking about and blamed it on something that is hard to define, or B) you did not actually have a Robicon Perfect Harmony drive, but rather a much older version Gen 1 Current Source Inverter (CSI) drive, which did use tuning capacitors on the front-end ahead of the GTO thyristor rectifier, to help make the GTOs fire correctly. Because those caps were ahead of the rectifier, they did have a tendency to interact with supply side harmonics, one of the reasons why Robicon stopped building those years ago and went to Voltage Source Inverter technology, which is what the Perfect Harmony and T300MVI drives both are now.

Much more likely, in the case of the Robicon PH drive failing so catastrophically, is that something caused one or more of your IGBTs to rupture. The down side of VSI drives that use IGBTs is that the failure mode of them is extremely violent, often enough to cause collateral damage, which then has a tendency to cascade. If your capacitors ended up damaged, it might have been secondary to the initial failure. But I'm not there looking at it, so it's really all just conjecture on my part.

Reactors ahead of the T300MVi would be pointless, that drive has a 24 pulse front-end, meaning there is a custom phase shifting multiple output isolation transformer inside of it that provides much much more inductance than you would add with a reactor.
 

Besoeker

Senior Member
Location
UK
I'm not sure you have all of your info right, or I am not understanding what you are asking. The Robicon Perfect Harmony VFDs are not substantially different from the TMIC (Toshiba) T300MVi as far as the rectifier / inverter topology goes. BOTH of them use diodes for rectification into a DC bus that has capacitors in it. Neither of them should be particularly more susceptible to line side harmonics as the other, in that they both just rectify the AC to DC anyway.
I have a vague recollection that the Robicon was current source.
The Toshiba is PWM.
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
I have a vague recollection that the Robicon was current source.
The Toshiba is PWM.
Your memory serves you well. OLD Robicon drives were CSI, using GTOs on the front end, hence the filter caps. They switched to Cascading H Bridge VSI when they released the Perfect Harmony drives in the 90s. That's what the new Toshiba drives are now as well.
 

junkhound

Senior Member
Location
Renton, WA
Occupation
EE, power electronics specialty
damaged "by harmonics", .

I have seen vfd drives with blown capacitors due to line harmonics. At no output load, the inductance of the EMI filters resonated with capacitors in the circuit to a degree that the PFC output voltage was even exceeded and the caps overstressed by voltage and failed, so the harmonic failure phenomenon is not unknown. Fix was to add damping resistors across the EMI inductors to prevent peak ringing voltages above the PFC circuit output voltage.
 

big john

Senior Member
Location
Portland, ME
...At no output load, the inductance of the EMI filters resonated with capacitors in the circuit to a degree that the PFC output voltage was even exceeded and the caps overstressed by voltage and failed, so the harmonic failure phenomenon is not unknown....
This was a line-side filter? I recently read a problem someone reported about having inexplicably high voltage on the front end of a drive, and my guess was resonant ringing, but I'll admit I'd never seen it happen on the line side before.
 

aelectricalman

Senior Member
Location
KY
I'm not sure you have all of your info right, or I am not understanding what you are asking. The Robicon Perfect Harmony VFDs are not substantially different from the TMIC (Toshiba) T300MVi as far as the rectifier / inverter topology goes. BOTH of them use diodes for rectification into a DC bus that has capacitors in it. Neither of them should be particularly more susceptible to line side harmonics as the other, in that they both just rectify the AC to DC anyway.

So if someone told you that your Robicon drive was damaged "by harmonics", then there are one of two possibilities: A) they didn't know what they were talking about and blamed it on something that is hard to define, or B) you did not actually have a Robicon Perfect Harmony drive, but rather a much older version Gen 1 Current Source Inverter (CSI) drive, which did use tuning capacitors on the front-end ahead of the GTO thyristor rectifier, to help make the GTOs fire correctly. Because those caps were ahead of the rectifier, they did have a tendency to interact with supply side harmonics, one of the reasons why Robicon stopped building those years ago and went to Voltage Source Inverter technology, which is what the Perfect Harmony and T300MVI drives both are now.


I didn't see any capacitors on the DC bus with this MVI300,t but maybe I missed them. My observation of the GE Drive was that there where about fifteen compartments containing caps. That is where the explosion resonated. I will look at the drawings again. Here is a link to view them if you would please. https://drive.google.com/folderview?id=0B0oyWVwRE8m3c1RudDNsR2ZuSjA&usp=sharing . Any additional clarification would be worthy of a fruitcake this Christmas! Thanks again.
 
Last edited:
Status
Not open for further replies.
Top