Using DC voltage to operate a VFD

[electrified_cc]

Hello all,
This is my first time here. I am now registered. I have a question. Can you operate a AC 75HP Variable Frequency Drive by using a DC input voltage? What I am trying to say is can you input DC voltage on the DC bus on an AC variable frequency drive to get an AC output voltage to operate a AC motor. What I am trying to do is operate a 60HP 208-230 volt 3 phase motor using DC voltage from lead acid batteries. I am involved in an experimental project of building a completely electric powered trolley used to transport people in the City of Norwich, CT., and to be donated to the city. I am part of the Rotary International here in Norwich CT. My gut feeling is yes you can do this cause most VFD's have a DC output bus. So I figured with no AC Input voltage to create the DC output voltage on the DC bus, you could input DC voltage on the DC bus.
Also if this is possible, my estimates is that an approximate DC input voltage should be around 300 - 330VDC.

 

[G._S._Ohm]

Post a schematic of your specific VFD. The generic VFD block diagrams from the Web aren't detailed enough.

 

[gar]

110306-1749 EST

electrified_cc:

Almost certainly the answer is yes. But there are many considerations.

You used the term VFD. This may not be the kind of system you want. You need to look at the speed, torque, and acceleration factors of both the load (vehicle -- bus) and the motor. Quite possibly what might be described as a brushless DC motor would be preferable to get better low RPM torque. A brushless DC motor system is a variable frequency system, and is essentially an AC synchronous motor with other items added. It has many similar characteristics to a DC permanent magnetic motor system. You also want a regenerative system.

The batteries plus the already existing filter capacitors should work fine. Note that you may want to study the safety considerations of a DC system. Namely there is no AC cycle to extinguish an arc. Thus, external to the motor drive system you need to study what happens on a short circuit load on the batteries.

There are others on the forum that are directly involved in high power motor control and servo systems. These people can probably give you the best advice.

GM claims the Chevy Volt will out accelerate the Corvette and near 0 speed. This means lots of zero speed torque.

.

 

[broadgage]

Yes it can be done, though a pedant might consider such use to be a violation as it is almost certainly contary to the instructions.
OTOH ,does the NEC apply to vehicles ? I suspect not.

The battery voltage must be similar to the DC voltage within the VSD this may be determined by measurement or enquiry.

 

[Open Neutral]

Is there a full current terminal for a DC choke? That would be one possible input. You might parallel all the 'AC in' terminals and go in there, but only one set of diodes are doing all the work.

 

[gar]

110308-0747 EST

Assuming the VFD is has a three phase delta input, then there are only three input wires and to parallel all is to make it a one wire input. Won't work.

But if you connect to any two input wires, then it will work and use only two diodes. Full power load would probably overheat the two diodes being used.

The way to connect the battery to the system is by direct connection to the DC bus. Then you need to determine if there is a small AC power supply that is used to supply low voltage power to the electronic circuitry. It is unlikely that this low voltage power is supplied from the DC bus. Thus, the low voltage supply needs to be replaced by one that can use the DC bus as its input. It is likely the DC bus is designed for a nominal 325 V value. This can be supplied directly from 230 V AC.

.

 

[gar]

110308-0808 EST

Continuing this subject, but a digression.

For a long time I have felt that a logical solution to energy use reduction in a manufacturing plant doing volume production work with typical CNC machines would be to use a moderate battery bank and a common DC bus to all the CNC machines in an area. A battery pack per 10 machines might work.

The logic behind this is:

1. Most CNCs do not have regeneration back to the AC mains. There is limited regeneration to the DC bus capacitor bank. But when too much energy is returned to the capacitors, then the capacitor voltage rises too high and at an excessive voltage threshold the excess energy is dumped into a resistor. This is wasted energy.

2. If a battery is paralleled with the capacitors in a single machine, then that battery has to have the capacity to absorb all the excess energy of the one machine.

3. Combine one battery with multiple machines and the battery size probably can be less on a per machine basis, than if done on a single machine basis. The reason is that it is unlikely that all machines are running in synchronization, and unlikely that they are running the same program. Thus, peak loading and dumping would be random between machines.

4. Besides energy reduction another advantage is that short power outages would not abort a machine's operation. Aborting a cycle may just require restarting the machine, although it might be time consuming. But aborting also may cause damage to the part being made or the machine.

Then there are jobs run lights out or on weekends when no one is around. If the machine aborts, then there can be major lost production time.

5. Continuing further if solar power is a source on the building, then it can be designed to supply DC to the DC battery-bus combination.

.

 

[kingpb]

Interesting perspective. How have you calculated into the equation that the greatest detriment to a battery is the number of charge and discharge cycles?

I think an economic analysis would be valuable.

 

[gar]

110308-2115 EST

kingpb:

No calculations.

The Chevy Volt battery is classified as a 15 KWH unit. But by system design is operated over a middle portion of its charge range, 10 KWH, to provide long battery life. In a fixed installation where weight is not a factor the battery chemistry can be based on cycle life, and time period life.

.

 

[Open Neutral]

110308-0747 EST

Assuming the VFD is has a three phase delta input, then there are only three input wires and to parallel all is to make it a one wire input. Won't work.

.

gar is correct; I really should not post answers when I'm half asleep. Not sure what I was thinking except 3-phase auto alternator diodes were on my mind. But I do think the DC choke terminal would work, if there's a matching return terminal on the bottom end.

 

[Besoeker]

Hello all,
This is my first time here. I am now registered. I have a question. Can you operate a AC 75HP Variable Frequency Drive by using a DC input voltage?

Yes it is possible. Been there, done that.
There are a few things to consider.
You will no doubt want to be able to isolate the VFD and have an isolator on the input. Bear in mind that this needs to be able to switch DC. It's a different beast to the more usual AC rated isolator or breaker.
VFDs have DC link capacitors that need to be pre-charged so you will need to include circuitry for that.
AC output from the VFD can't be more than 0.74 of the DC supply voltage.

 

[beanland]

Commercial DC Link VFD

Commercial DC Link VFD

Ther are commercially available VFD inverters with DC link capability so that a common DC bus can be used. This works well with 4-quadrant drives so that energy recovered from one drive can be used to power a different drive. This is really nothing more than an updated version of the DC motor system with a common DC bus used in the past.

 

[Besoeker]

Ther are commercially available VFD inverters with DC link capability so that a common DC bus can be used. This works well with 4-quadrant drives so that energy recovered from one drive can be used to power a different drive. This is really nothing more than an updated version of the DC motor system with a common DC bus used in the past.

The old Ward-Leonard system you mean, maybe?

The DC fed systems we have put in have had a 24-pulse rectifier to limit harmonics on the supply. You can regen into the DC bus provided the total load is positive. I see this on paper mill drives where there might be a dozen or so different drives and it isn't uncommon for some of them to get pulled round by others in normal operation. The DC to AC inverter section is usually inherently capable of regeneration anyway.
We still include dynamic braking resistors for stopping - all the drives on the machine might be stopped at the same time in an emergency.