Page 1 of 2

Fuji variable speed drive any good?

Posted: Thu, 19 Mar 2009, 00:32
by pieceofstring
Hi guys, I'm a virgin at this so be gentle!
Would a Fuji 45KW variable speed drive be suitable for an EV? Are they any good?

Fuji FRN45P11S-4JE
Source 3ph 380-440V 50Hz, 380-480V 60Hz 124A
Output 3ph 380-460V 0.1-120Hz 69KVA 91A 110% 1min
34Kg Dimensions 375W * 550H * 270D

Any comments appreciated.

Fuji variable speed drive any good?

Posted: Thu, 19 Mar 2009, 01:11
by Johny
Hi pieceofstring - welcome.
The 110% overload capability tells me that it might be an HVAC unit. The disadvantages of HVAC is that they often do not have torque control and sometimes do not give you easy access to the DC Bus.

91 Amps is just enough for 11 to 15kW but not quite enough if you are planning to rewind to a lower voltage. That being said, my VFD can only do 89 Amps and I AM planning a rewind, it will just be controller limited.

If the price is good and it mentions the DC Bus connections in the manual then it's a good buy. The torque control, while nice, is not currently being used by the 2 AEVA members (I know of) running VFDs.

Fuji variable speed drive any good?

Posted: Thu, 19 Mar 2009, 08:58
by acmotor
Fuji FRN45P11S-4JE price is good US$2587 ex Singapore. (from a quick google)

http://fa.jonweb.net/fuji.htm

data at http://stech.vn/products/fuji/Catalogue ... 20P11S.pdf Top of page 9 is this model.

Don't worry about the mains input number of 124A, it is without DC reator so power factor is bad (none of that matters with EV system anyway)

Regen braking is only 70% with brake resistor or battery pack on DC bus but that is 70% of 45kW .... heaps.

DC bus can be connected to P(+) and N(-) terminals. I have done this on a Delta brand controller (stangely, very similar to the fuji) Works fine. There is inrush protection built in.

This controller 'could' be quite suitable for EV.

Just read through all the data and user parameters. They seem like a fairly standard set.

This model P11S only does speed control, G11S does torque control as well.

Oversizing compared to motor (use this one on 11 to 15 kW as Johny suggests would be fine) 110% overload is only a limit at full rated motor size. Controller can apply 200% (or more) as long as it is within the output current limit.

Image

Fuji variable speed drive any good?

Posted: Sat, 21 Mar 2009, 00:39
by pieceofstring
That's pretty good guys.   Image
You will have to give me a week or so to digest everything ( and work out what it all means). I appreciate your help.

Fuji variable speed drive any good?

Posted: Fri, 28 Jan 2011, 12:44
by brandon_costa
I'm using a Hitachi SJ300-550LFU. I got this barely used drive for a really good price. I believe Hitachi doesn't manufacture this drive anymore, but some websites still show it as available. When I called Hitachi for tech support they mentioned that it's been obsolete for a while. So far it's been very good and I'm happy with the tech support from the factory.

Fuji variable speed drive any good?

Posted: Fri, 28 Jan 2011, 13:42
by Johny
When you say "I'm using" do you mean that you are using/going to use it in an EV Brandon?

Fuji variable speed drive any good?

Posted: Fri, 28 Jan 2011, 20:14
by T2
If you are going to lay out $2000 + I would look at Curtis they do a range that will allow you to use 144Vdc battery supply. IMO 300Vdc supplies and above are somewhat daunting for a noobie and need an experienced user. Those previous comments obviously came from experienced users of course.

Success is also comcomitant on getting your motor rewound. The volts per HZ for an idustrial 460v 60 Hz motor is 8.0. That's no good for 144Vdc or even 320Vdc You need closer to 0.5 V/Hz and a rotor balance for 8000 rpm as ACMOTOR pointed out in one of his earlier posts somewhere on this site.
That will give a base speed of around 6000rpm for 144Vdc. This is the approach I am taking, together with a 10:1 ratio.

Fuji variable speed drive any good?

Posted: Fri, 28 Jan 2011, 23:17
by antiscab
actually it is ideal to hit be able to run into the bus voltage limit somewhat before max rpm.

This allows a constant power area, meaning you aren't paying for a peak controller poweryou aren't likely to be able to use.

The only time you wouldn't do this is if your motor physically isn't large enough.

for instance, in a directdrive situation, with a vehicle that needs a top speed of 120kmh, weighs 1500kg.

the mass means the minimum tractive effort is ~5000N.

the top speed means minimum power @ 120kmh is ~50kw (peak).

so with a 120kva VFD, that means you need to hit constant power at ~72kmh.

because you can reduce the slip, your shaft power at 120kmh is still ~60kw (the additional voltage needed to deal with the inductance at higher frequency causes progressively poorer PF).

for the above example, smallest possible motor that will do the job is a ~11kw nominal 4-pole. so 60kw @ 3000rpm max, calls for 1V/Hz (100vac @ 100Hz from a 144vdc bus).

Matt

Fuji variable speed drive any good?

Posted: Fri, 28 Jan 2011, 23:22
by antiscab
to clarify, top speed determines minimum power of the vfd.

minimum tractive effort determines the speed of the start of the constant power zone.

Matt

Fuji variable speed drive any good?

Posted: Sat, 29 Jan 2011, 04:52
by coulomb
antiscab wrote: actually it is ideal to hit be able to run into the bus voltage limit somewhat before max rpm.
Yes, for direct drive, this is essential, IMHO.
the mass means the minimum tractive effort is ~5000N.
I assume you're talking linear force here, in Newtons, not torque in Newton-metres. That makes it independent of the wheel size, which is good.
the top speed means minimum power @ 120kmh is ~50kw (peak).

so with a 120kva VFD, that means you need to hit constant power at ~72kmh.
Um, so 72 km/h is 20 m/s, so mechanical power = F.v = 5000 . 20 W = 100 kW. So a 120 kVA drive at a power factor of around 0.83 puts out about 100 kW of mechanical power.

So you'd arrange your wheel size and overall reduction ratio to get about 5000 N at the wheels from your peak motor torque, which you should get from zero to some 72 km/h. From there on, you get around 100 kW of mechanical power or a bit less as the frequency increases and the power factor worsens. Actually, I'm not sure that the effects of inductance show up as a worsening (lower) power factor. But the mechanical power available from the motor will decrease gently as the speed increases past the point where you run out of voltage (nominal speed, I guess that's still called; it gets a bit weird after a rewind and with overvoltaging / overspeeding).

Have I got this about right?

Fuji variable speed drive any good?

Posted: Sat, 29 Jan 2011, 11:59
by antiscab
yep, all correct, though I think I over estimated how much the power drops after going past the point where torque drops off (not necessarily the nominal speed).

that means you could get away with a smaller drive, and reaching nominal speed somewhere below 72kmh.

I think we need to come up with new names for the post rewind nominal :)

I picked numbers that were already bouncing around my head.

Matt

Fuji variable speed drive any good?

Posted: Mon, 31 Jan 2011, 11:06
by T2
I think we are on the same page as regards the usefulness of a rectangular power profile as opposed to the triangular. My industrial experience has been with (DC) drive salesmen who needed some prodding to get them to accept field weakening. They would attempt to get you to purchase the next torque size up. This forces a higher amperage controller and motor. Rather than expense the upgrade to 50Hp motors, I was able to persuade motor manufacturers to stamp my 40Hp motors with 1800-2400 rpm nameplates. Of course the timing belt gear ratio was changed to allow 2400 rpms but that 25% extra torque proved useful on Monday morning starts when encountering a cold machine that had become stiff from standing.

For electric vehicles it's even more relevant - drivers don't want to wait for their peak power to arrive at vehicle max speed either.

With the power source I am using the final objective is towards a base speed of around 6600rpm for meeting 144Vdc at a road speed around 42mph. This is the approach I am presently taking and an integral gear reducer with a 10:1 gear ratio is in the process of construction.
I assumed that you would understand the vehicle would be exceeding base speed in constant power. In fact the goal is to reach 75mph at 12000 rpm passing thru' 60mph at 10000 rpm.

I have come to learn that the difference between an industrial and an automotive controller is the expectation that with the automotive you can also extend constant power in the other direction. That is, BELOW base speed by utilising overcurrent margins of 300%.

This is typically done specifying a 300 amp controller, for example, with a battery source capable of 100 amps max. In order not to exceed 100 amps these controllers have battery side limit protection.

The effect is that from a rolling start the motor current can be 300 % nominal until the vehicle reaches one third base speed (14mph). At which point the controller begins reducing motor current inversely to speed such that it has reduced to 200% by half base speed (20mph) and down further to 150% at two thirds base speed etc etc.

The advantage is to give an enhanced launch for reduced 0 to 60mph times compared to the more traditional constant current to base speed enjoyed by industrial controllers, with the benefit that the battery need not be subjected to abuse.


Fuji variable speed drive any good?

Posted: Mon, 31 Jan 2011, 14:34
by T2
Jan 30 2011 #2

Your mention of an 11.7Kw drive reminded me that last year I walked away
from an FX-8500 servo drive complete with all encoder and communication cables intact plus the 15Hp motor of course. From internal markings we saw it was of 1999 vintage which explains why the procured interface software needed a WIN95 OS. A visit to our local greenTech recycler - actually a clandestine night time trip to the dumpsters situated at the rear of the building - located the perfect machine. More digging and soon a mouse monitor and keyboard arrived to complete the ensemble - who doesn't like free ?? On previous visits we've picked up gently used Hi Fi items and other gems with great results, although there was one meanie who had jammed a sharp object into the video connector rendering the whole monitor useless.

The FX-8500 was only suitable as a direct drive into the differential since the controller was good to just 100Hz which limits the 4-pole motor to 3600rpm. The intention was to spoof the AC and DC inputs which were clearly identified on the PCB and thereby enable operation at 144Vdc. Originally we had intended to feed 460Vac from our microprocessor controlled 4-stroke generator but this was already giving vibration issues at only 110Vac and wasn't going to reach 3000rpm without a mounting redesign. It was also much louder than expected and would have been impossible inside the van. One day my hearing will return I hope !
On top of all that I knew from everything I know that even if it all worked perfectly the performance would still be mediocre.

In the meantime I read about the Curtis from this site and the success with induction motor rewinds and suddenly our FX-8500 didn't look the most pragmatic choice anymore. We would have had a lot invested to make the FX-8500 work and if it failed in service the rebuild could end up being prohibitive on a drive no longer supported.

Regarding actual motor frame sizes, it is not a business in which I am acutely aware. Consequently a 60 page report from Rohr Industries back in September 1978 must have flown under my radar. This is from a NASA contract entitled "Preliminary Powertrain Design for a State of the Art EV" if anyone can find it. They had 3 AC motors rewound for 68v @ 240Hz. The term V/Hz probably not in vogue then. The following 0.28V/Hz 4 pole motors were ordered.
pp 39-40   
GE 184T             31.3Kg     cont power 19.2Kw @ 9000rpm   44.7Kw 30 sec
GE 215T             38.6Kg     cont power 38.3Kw @ 9000rpm   88.1Kw 30 sec
RELIANCE 215T 56.7Kg cont power 38.3Kw @ 9000rpm   88.1Kw 30 sec
Their 30 sec rating probably represents 230% rated torque.

Presumably modern machines with copper replacing the aluminum on the rotor bars could allow up to 390% rated torque.

Rohr chose the aluminum framed GE215T for their reference design and at 55mph constant speed noted a temperature rise of 15 deg C. with 250 cfm cooling. They don't mention the ambient temp.

I noticed that the brand I examined rates their industrial motors at full load with a 60deg C rise in a 40deg C ambient.

Had you asked me last week I couldn't have told you this but now...
FYI a 184T frame is a 5Hp @1800rpm 3.7Kw
        a 215T frame is a 10Hp@1800rpm 7.4Kw

Given what these two frame sizes can yield at 9000rpm what is all this talk regarding an 11.7Kw frame size ? ?

Fuji variable speed drive any good?

Posted: Mon, 31 Jan 2011, 20:38
by antiscab
T2 wrote:
Given what these two frame sizes can yield at 9000rpm what is all this talk regarding an 11.7Kw frame size ? ?


9000 rpm would require a gearbox, which itself is not exactly light.

The larger induction motors are also more efficient (actually its more like a 30kw continuous motor at 3000rpm has the same efficiency as a 30kw continuous motor at 9000rpm).
its weird how you can push an induction motor harder and it becomes more efficient.

The gearbox also has its own losses.

bigger motors have more thermal mass, so the 1 hour rating is higher (handy if you find yourself going up everest)

The bigger motors aren't that more expensive (at least second hand).

to use the 30kw continuous motor at 9000rpm, it effectively means you can't use field weakening, meaning a big VFD you only get use at full power at top speed.

Those are the reasons that come to mind...

Matt


Fuji variable speed drive any good?

Posted: Mon, 31 Jan 2011, 20:58
by Richo
T2 wrote: GE 184T             31.3Kg     cont power 19.2Kw @ 9000rpm   44.7Kw 30 sec
GE 215T             38.6Kg     cont power 38.3Kw @ 9000rpm   88.1Kw 30 sec
RELIANCE 215T 56.7Kg cont power 38.3Kw @ 9000rpm   88.1Kw 30 sec
Their 30 sec rating probably represents 230% rated torque.

Presumably modern machines with copper replacing the aluminum on the rotor bars could allow up to 390% rated torque.

Given what these two frame sizes can yield at 9000rpm what is all this talk regarding an 11.7Kw frame size ? ?

I agree there are motors around with even higher specs than 390% rated torque.
So over 100kW is possible from a 40kg motor.
But the issue for most converters is down to two problems.
1. How to gear 9000RPM to something useful (most ICE rarely see 6000RPM)
2. Getting the rewind low enough or getting the battery voltage high enough.

So the compomise is to use a larger/heavier ACIM with lower RPM lower battery voltage that fits to a regular gearbox or direct drive.
So an 11kW ACIM is the generic compromise.

Don't forget that 9000RPM would be the peak power.
Most motors rev past peak power to match the next gear.
So now you are talking 12000 max RPM (Like a Tesla)
Good in theory but not practical for the average converter.

How much extra are you spending on a high speed motor and custom reduction drive vs a bigger motor and standard box?

Fuji variable speed drive any good?

Posted: Tue, 01 Feb 2011, 12:46
by T2
Let me do a friendly rebuttal to your comments here, and put this industrial equipment to be or not to be thread to rest.

I think I have mentioned twice that an integral gear reducer with a 10:1 gear ratio is in the process of construction. This gear reducer will mount directly to, and be concentric to the end frame of the motor.
There are several reasons I'll be "rolling my own" reducer.

Most auto transmissions won't handle 7000+ rpms without complaining. I have first hand experience of that being attempted. Furthermore there are significant losses incurred simultaneously spinning all those gearsets enclosed inside the gearbox since they are all in mesh full time. You do understand that only one gear is dogged on to the shaft by the gear shifter fork and is actually transferring power at any one time leaving the other unused gears to continue to churn away. How efficient is that ?

I expect a single reducer planetary to be simpler and lighter in comparison. The fastest rotating component will be the motor pinion (sun gear) itself on a 1.5" protruding shaft from the motor.

All I have for you now on the project is that the intention is to use straight cut spur gears with a diametrical pitch of 16. The sun gear is to be a 12 tooth and the planet gears will be off-the-shelf 48 tooth.
The assembly will fit on to an 11" motor flange which is standard for 184T and smaller frame sizes which gives the freedom to choose from many machines. Putting the sun on the standard 1.125" shaft is going to be interesting ! Initially the intention is to start small with a more manageable 182T size.

Neither the trainee machinist with CNC access nor myself have ever built such a device. It's going to be a learning experience. For that matter few machinists ever get called upon to produce internal tooth gears. I have a sample reducer which is undergoing a tear down to give an idea of what considerations might be in order. From that exercise there will be a better understanding of what is doable with the facilities available.

Expectations of longevity is not on the table at this time but the completed unit will be rotated at no load and high speed for several hours before the decision to do a motor rewind is taken.

Yes rewinding does improve efficiency - its been discussed here several times. Anyone interested search this site for "104v motor" The main reason is that although the Copper loss remains exactly the same after the rewind the motor now puts out about 4 or 5 times the power. Unfortunately Iron loss does increase with frequency. The way I see it is that there was never any magic reason why 60Hz/50Hz should be the sweet spot for efficiency. It's not. The use of inverters has shown that motors are easily capable of 5X more power without bursting into flames or requiring gale force winds of fan cooling.

But all this aside, it is of no interest to industrial designers - I happen to have been one - since most want their power at 600 rpm. Placing ten to one reducers everywhere would be inconvenient to say the least. But then most industrial equipment is not expected to race between stop lights in busy traffic either ! Efficiency and lightness are rarely a consideration.

What needs to be changed here is the culture of thinking that has grown up around the 50/60Hz tradition. Aircraft have used 400hz for years. It is happening slowly but inverters no longer contain 60Hz control transformers. In fact the FX-8500 uses two off-the-shelf switchers with 220Vac input. No tap changing there. The problem in controls is exacerbated by the fact that most second hand equipment available is 110Vac whereas industry has now migrated to 24Vdc sensors and controls.

"But the issue for most converters is down to two problems.
1. How to gear 9000RPM to something useful (most ICE rarely see 6000RPM)
2. Getting the rewind low enough or getting the battery voltage high enough."

1. The gearbox I am proposing won't be anywhere near the weight and size of anyone's 5-speed manual. I'll be posting details of my costs on a separate thread in the future.

2. Figuring out ways to get high voltage from battery arrays I see as a problem caused by trying to escape from the expense of motor rewind. Incidently while I was looking to go for 0.5V/Hz but on reading that report from 1978 it seems that even that is too kind.
At 68v @ 240Hz they are stipulating 0.28V/Hz !

Just in case someone may be getting left behind on these figures. V/Hz is a measure of motor torque, besides motor current that is. It is not clever to just quote lower values of V/Hz since the unspoken understanding is that you must now supply more amps just to get the torque you had before. No free lunch. Fortunately there are now companies that understand this and provide high output current controllers at a reasonable price. Industrial sellers on the other hand think they are doing you a favour when they supply a 200 amp unit that weighs 200lbs and requires mortgage backed lending !


Fuji variable speed drive any good?

Posted: Tue, 01 Feb 2011, 21:00
by Richo
Ah so you are an idustrial designer with acess to CNC facilities.
That would be why you would build a custom box.
I would do the same in your shoes.

There are a lot of people who rewind motors.
But the issue is most will NOT rewind a motor less than 200V here.
Too much trouble - not enough money in it.

Sorry pieceofstring for the hijack Image

Fuji variable speed drive any good?

Posted: Wed, 09 Mar 2011, 10:52
by T2
Hey Richo,
Don't be apologising for me - I didn't hijack the thread.

The OP specifically stated Any comments appreciated.

And I answered directly that there are better ways to go.

If you are going to lay out $2000 + I would look at Curtis they do a range that will allow you to use 144Vdc battery supply. IMO 300Vdc supplies and above are somewhat daunting for a noobie and need an experienced user. Those previous comments obviously came from experienced users of course.

Success is also concomitant on getting your motor rewound. The volts per HZ for an idustrial 460v 60 Hz motor is 8.0. That's no good for 144Vdc or even 320Vdc You need closer to 0.5 V/Hz and a rotor balance for 8000 rpm as ACMOTOR pointed out in one of his earlier posts somewhere on this site.
That will give a base speed of around 6000rpm for 144Vdc. This is the approach I am taking, together with a 10:1 ratio.


Then antiscab clarified the meaning of Base speed. And I subsequently made clear that indeed base speed (of the voltage saturation kind) should occur at about half max speed. I also offered the more modern concept which suggested extending the constant max power profile below base speed by utilising the short term overcurrent of 300% that the latest controllers like those from Curtis are designed to support.

Finally I demonstrated my disdain for industrial controllers whose only redeeming feature is "free". They present constructors with the need to provide dangerous and expensive battery voltages of 330Vdc and higher. They also constrain constructors to utilise 60Hz 230Vac motors. Such machines have very poor power to weight ratios.

All these reasons are why I kicked my particular gift horse of an MX-8500 in the teeth. I would suggest to all noobies to at least take a look at the alternative culture as presented on the website 400HZ.net

Fuji variable speed drive any good?

Posted: Wed, 09 Mar 2011, 20:59
by Richo
True - he did say any comments! Image

Regarding 400hertz.net.
Many many moons ago I asked for a quote on this:
150kW motor
Strange that they couldn't since they still avertise it.
Any chance you could give it a go???

Fuji variable speed drive any good?

Posted: Wed, 09 Mar 2011, 21:31
by Tritium_James
Richo, that motor looks exactly the same as what AC Propulsion sell with their system - it looks identical to what we had in the Porsche. Might give you another avenue of enquiry...

Fuji variable speed drive any good?

Posted: Thu, 10 Mar 2011, 06:56
by Richo
That's right.
Which is the probably the reason why 400hertz.net wouldn't give a $$$
So is 400hetz.net selling unobtanium?
I don't want the system just the motor. Image
AC Propulsion
I liked the Tzero - been a while since that has come up.
AND they had a range extender trailer.

We're all livin in their shadow
All bow down to the AC gods Image Image

Fuji variable speed drive any good?

Posted: Thu, 10 Mar 2011, 16:22
by T2

   LOL, Richo you are ON !!

I e-mailed 400Hz.net earlier this week regarding the pricing of any machine they may have available in an approx L184 frame size, although I had considered doing a rewind locally. I briefly stated that the application required a machine with a 0.5V/Hz spec with constant power beyond 6000rpm and that one of their 400Hz 200Vac motors to be operated on a 144Vdc bus would seem to be a good fit.

The proprietor e-mailed me back advising that his products incorporate thinner lams for reduced iron loss - comparable rewinds would not. I would also be disappointed with the performance of the rewind's aluminum rotor.

All very nice if I can afford it no doubt - and I am not averse to going back and doing some gold plating later, if the rest of the project succeeds - but for now I still have to contact the guy again - this time I'll use the phone - for the pricing. I'll put in a word, richo, for your 150Kw as well.

Expect a report here soon.




Fuji variable speed drive any good?

Posted: Fri, 11 Mar 2011, 06:14
by Richo
I'm sure i'll keel over and start involantary twitching when I hear a price Image

I'd be lucky to get half that power from the same weight.
Go the copper rotor and thin lams!

Fuji variable speed drive any good?

Posted: Fri, 11 Mar 2011, 14:37
by Mesuge
Interesting similarities there, 12000rpm multi-lol; however ACP introduced some dielectric inside their ver. of emotor - so it worked as charger with their inverter as well. Also they supposedly outsourced the manuf. based to asia few years ago, not sure if that meant both emotor + inverters.

Fuji variable speed drive any good?

Posted: Fri, 11 Mar 2011, 14:42
by Johny
Richo wrote:Go the copper rotor and thin lams!
Can't resist the thin lams can you Richo. Image