powerful stepper /brushless motors up to 12Krpm 750+W

[Simon]

I'm looking for stepper motors and controller if available capable of 750W of power and up to 12'000rpm @ 28V. Is this something that can be obtained easily or do they tend to be bespoke made ?. We don't need many so having stuff made is a bit of a non starter. Potentially we could look at a controller that will bump the voltage up say 2x if it's customed made or available in order to get what we want.

Can anyone recommend any particular manufacturer / supplier ?

[Simon]

yep. We could use a belt to increase speed.

It's not fast, your hard drive motors do up to 15Krpm and the dyson air blase hand driers have a 90Krpm motor in them.

Is stepper the correct name ? we need speed and power with no brushed, positioning accuracy is not a problem.

[Simon]

So what is the difference between a brushless and a stepper ? my recent investigastion seems to show that "brushless  motor" just describes a stepper of some sort + a controller.

So where do I get a brushless motor with these abilities?

[Andy Watson]

15k RPM stepper would be some motor, especially at 750W. I think the term you are looking for is "spindle motor".

[Simon]

two caveats, reliability is of the upmost importance. I see funny looking specs when i look at RC motors that remind me of things like the laughable PMPO ratinmgs of compter speakers.

I need reliability and peak power for prolonged periods so not something thats made to give a shot of thrust and then rest for a while.

[Andy Watson]

Put "spindle motor drive cnc" into your favourite search engine.

[Simon]

And if you read the brushless motor page it describes a stepper type motor with built in controller and says about as much:

"Brushless motors may be described as stepper motors; however, the term stepper motor tends to be used for motors that are designed specifically to be operated in a mode where they are frequently stopped with the rotor in a defined angular position. This page describes more general brushless motor principles, though there is overlap."

Yes a stepper can be specifically positioned and is used for that sort of precision, a brushless will usually be controlled to just turn. "brushless" motors I have seen driving fans actually index and go through a setup procedure before they will run because they have a built in procedure. They usually index to work out the position they are in and check that the fan is not obstructed before starting up full power. I would assume that brushless motors have much few steps than precision positioning stepper motors.

[oldway]

Why "brushless" ?
As you can use a belt to increase the speed, what about a very reliable vacuum cleaner motor?
I mean a Nilfisk GM80C motor....power is hight, 1300W but need force cooling. No problem to reach 750W.
Workt on 240Vac (also on 240vdc).
I do'nt now how much rpm, but it is high rpm. You should ask this information to the manufacturer.
You should use a dc/dc converter 28/240V if grid is not available.
And a rpm sensor for rpm regulation.
A Mc Gijver solution.... 

[Simon]

Yea, I'm a bit like Mc Gijver around here, always end up being asked at the last minute to see if I can fix it. I did suggest a vacum motor but power is the problem. I think we may have emi problems too.

[G7PSK]

Try this company.

http://www.sem.co.uk/products/

[SeanB]

What you want is basically a 3 phase squirrel cage motor (regular one without brushes) that will have a drive from an inverter that typically has a mains input and 3 wires to the motor.  You having to run from 28V means either you are looking at trucks or aircraft, and are looking at a custom wound 400Hz 3 phase motor and a drive system to power it. You can get IGBT modules that will handle the 40A or so current draw that the motor will need, and there are a lot of VFD inverter drive chipsets around that can do the PWM drive. You just need to join the dots, but the drive is going to be a fairly big chunk of heatsink with a board attached as an afterthought. Probably about the size of a loaf of bread.

Is it not possible to use a series wound, shunt wound or permanent magnet motor with a tacho pickup and a speed regulator? A 24V one will likely be an off the shelf item from a motor manufacturer like Bonfiglioli, WEG or one of the other motor manufacturers.

[Simon]

I think i might have to ask more questions about what is actually required. My aim is to have the 10-12000rpm as we are using a belt and pulleys at the moment which I'd like to avoid (recovering some cost for a better motor solution). It's land vehicles btw.

[Rufus]

So what is the difference between a brushless and a stepper ?

Steppers are usually 200 steps per rev, brushless DC motors usually 3 so the difference is usually 197 poles.

[Simon]

Right, I was unclear as to actual construction of the raw motor before it has a controller, however any combination is possible i guess.

[NiHaoMike]

What about a large model aircraft motor? Just derate it for continuous operation.

The fastest motor I'm aware of in a consumer product is a water cycle air conditioner compressor that runs at up to 100kRPM, but that's a switched reluctance motor.

[Simon]

I'm just concerned that model aircraft motors will not have the durability and reliability, this motor really really must not fail.

[Kremmen]

I'm looking for stepper motors and controller if available capable of 750W of power and up to 12'000rpm @ 28V. Is this something that can be obtained easily or do they tend to be bespoke made ?. We don't need many so having stuff made is a bit of a non starter. Potentially we could look at a controller that will bump the voltage up say 2x if it's customed made or available in order to get what we want.

Can anyone recommend any particular manufacturer / supplier ?

It is not something you can get at all, never mind easy. A stepper like that is not obtainable for love or money. Nor would it make any sense or be in any way the right choice for such requirements. Steppers are generally used in low performance positioning applications and similar. And in any case, 28V would not be nearly enough to force feed a stepper to the astronomical step rates required for such a shaft speed.
The real killer is the low voltage. Cetain kinds of brushed DC motors can do RPM like that on such a low voltage, but then you might be against the reliability requirement. If it "really, really" must not fail, then your best option is a 3 phase induction motor or AC servo. Only AC servos generally do max 3000 or in extreme cases 6000 RPM and induction motors no more than 3000 or 3600 RPM. But insert a 200 Hz capable inverter drive into the equation and you can do your 12 kRPM with the right motor. Unfortunately it won't do it with 28V.

What you need to do is put your requirements in priority order. Fulfilling all of them could become awfully expensive, but fulfilling the most important 50 to 80% could be manageable.
Others have included useful links to mfgs that make motors in the categories i mentioned above. The solution, if any, will be somewhere in that bracket

[penfold]

It does sound like a bit of a custom job.  There is a company in the UK called Controlled Power Technologies (CPT) who make some high speed and high power motors for automotive super-chargers, so these will also have with them a pretty descent reliability, could be worth contacting them.

[Simon]

What you need to do is put your requirements in priority order. Fulfilling all of them could become awfully expensive, but fulfilling the most important 50 to 80% could be manageable.

Well my priorities are:

1) power - without it nothing will be happening

2) voltage, I don't think we want to be using inverters as we are mechanical not electronic so the very idea scare them

3) speed - being a mechanical company we will use gearing

Personally I'd swap 2 and 3 but my boss will have it as above. I can't see much wrong with setting a target for 46V using a voltage doubler off a H bridge.

I'm not neccessarily after fine step steppers but something that can be controlled to do what we want.

As for it being near impossible we already have a motor that does the power and voltage but it needs gearing up and it was not really designed for what we want so stalls on startup as it does not expect the mechanical load we are applying to it and has it's custom controller hence me seeing this as a two part thing, motor (of whatever type you care to call it) and driver.

[Paul Moir]

You're not gonna generate motor power with a doubler.  Forget that.

I don't think you say how much power you need, but if it's outside of a dremel motor than it's not going to happen with 28v.  Motors are inductors and you need voltage to get power into it at high speeds.

I think someone mentioned aircraft motors.  By which I think he means 400Hz proper aircraft motors.  That is a good McGuiver idea as they're small high speed items and you might be able to find an old one cheap.   

10-12K rpm means 10k right?  If you mean 10rpm it's gonna be a brushless servo motor.

[Simon]

like i just said we have a motor that is 750W at 28V but it's not setup to run with the type of load we are giving it so stalls on start thinking that it's jammed because we cut it out of something else and adapted it

i have seen a number of motors rated to 48V already


it's 10K - 12k I'm after, although the power is more important than the speed

[Paul Moir]

Oh sorry, but I think that shows the point about the doubler:  750/28 is a lot of current to make with a doubler.

How fast does your current motor spin?  To spin it twice as fast with the same torque should be twice the power then right?

PS - Hard to beat aliexpress sometimes!  Check out this: http://www.aliexpress.com/item/1-5KW-Water-cooled-spindle-motor-220V-ER11-24000rpm-3-bearing-1-5kw-Inverter-VFD-2HP/1252450743.html 
(Just googling around..)

[Simon]

I think the current one is around 2500-3000rpm so we have geared it up, the torque is there but it was used before to drive a light plastic fan, now that we have connected it to a pulley and belt it sort of gets confused and is fussy about starting so i suggested we get something setup properly.

[Rufus]

2) voltage, I don't think we want to be using inverters as we are mechanical not electronic so the very idea scare them

The electronics to drive anything but a dumb brushed DC motor from a 28v DC supply is going to be as complicated as an inverter anyway.

I think your choice is limited to brushless DC which which you might be able to get at 15k rpm and 28vdc drive (but how well does it have to perform when your 28v is only 22v?) or 3ph induction.

You can get 2 pole 3ph induction motors good for 15k rpm which requires 250Hz drive which is available from some off the shelf 3ph motor drives. Some 3ph drives can be fed around 350vdc instead of 120/240v single phase ac. You would need a 28v to 350v DC-DC converter (at more than 750W).

http://www.groschopp.com make a lot of motors - much more than gets listed on their web site. I know they do 2 pole 3ph motors good for 15k, don't know if they go up to 750W with them.

[peter.mitchell]

I think the current one is around 2500-3000rpm so we have geared it up, the torque is there but it was used before to drive a light plastic fan, now that we have connected it to a pulley and belt it sort of gets confused and is fussy about starting so i suggested we get something setup properly.

Does the unit use a controller? (I assume it does) The controller probably has an expected startup RPM curve, so when you change the load, the RPM curve does not match and it changes poles too quick/slow - either that or it goes failsafe.

A lot of controllers in the RC world instead of following a fixed curve, use inductive feedback from the motor windings to measure RPM and advance/retard the timings of pole changes appropriately, because in the RC world, any man and his dog can put this stuff together, with random gearing (and load), motor and power supply selection.

Either change the curve or change the startup method.

[Simon]

like i say we hacked this motor off a fan and then gave it some heavy gearing to drive. It come with a built in controller so no chance of changing it as that is not meant to be and ultimately we don't want to keep cutting motors off of fans we want a motor that is bought as a motor.

[G7PSK]

If you are mechanical engineers its most likely that you have compressed air service. So what about an air motor, they have high speed and are available in quite high powers Ingersoll Rand do up to 30KW) and they wont have any emi problems.

http://www.ingersollrandproducts.com/eu-en/products/air-motors

http://www.pneumatictoolservices.co.uk/industrial_air_motors.htm

[Simon]

except most automotive vehicles don't have compressed air.

[Paul Moir]

That would have been helpful to know several pages ago.

I give up.

[Ton]

Best I have found is this, maybe share the burden between two ?

400W 24V 14011RPM @ no load

http://www.peromatic.ch/d/produktelinien/moog/bldc_motoren/bn34hs.pdf
reseller of Moog.com motors
http://www.moog.com/literature/MCG/bnhsseries.pdf

Most BLDC motor manufactors stop their standard line around 400W, especialy on the high RPM line

but dont think they are cheep 

you could check http://www.maxonmotor.com
their EC line look fine, but stop at 250W 24V 10400RPM (EC = electronically commutated)

so a search after "electronically commutated motor" might find something, but now I am off to bed

good luck

[codeboy2k]

...must never fail

You've lost already.

If you're a mechanical engineer (did I read correctly? ) then you should know that moving parts will fail.

Your best defense against this :
(a) a time-limited lifespan where it is guaranteed not to fail because you've done the math on the wear and tear already, or;
(b) a maintenance program with specified replacements that is strictly adhered to in order to get more life out of the moving parts. 

When you say it must not fail, then if it does fail is someone likely to die? or be hurt? I read about automotive something earlier on...

By the way.. that Wikipedia page that says brushless DC motors are like steppers... I really want to edit that out. Brushless DC motors rotate continuously, are never stepped.  They may be used in a closed loop feedback as part of a servomotor system, but this does not make them "steppers" just because the feedback loop allows them to be "stepped".  They actually make really, really good servo motors because they are mostly RPM linear with voltage and torque linear with applied current, so if you know the torque you need to move a mass X meters, it's easy to calculate the drive requirements. If your bed or pickup has a load cell on it you can feedback the mass to the motor controller and it can apply needed torque to move a changing load with constant velocity.  This makes BLDC motors in servo loops really good for robotic arms on factory floors.

[strangelovemd12]

Can't you just post a picture of the current setup, or describe what mechanical action needs preforming?  Three pages of teasing out what you are trying to accomplish is driving me crazy.  So far it sounds like you just need to add a .15 duty starter to your current rig to overcome static load.

At the very least , fill in the blanks:
Available Voltage: 24v
Available Start Current*:
Available Run Current*:
Final Shaft Speed: 10k-12kr/m
Ramp-up Time (max/min):
Starting Torque:
Static Load (y/n):
Average Run Time**:
Max Run Time**:

*   Is 750w your max power supply output?  Because measuring the actual mechanical requirements of a motor as a function of it's input current doesn't really work.
** "Continuous" can mean a lot of things.

Also wouldn't hurt to know about the thermal properties of the space the motor would occupy.  And the motor itself will fail at some point, never mind the chain of electronics/mechanics on either side of it.  The closest you'll get to non-fail is a highly redundant system.  You probably don't want to do this whole thing in triplicate.

In other words, take a picture of the current setup.

[Simon]

That would have been helpful to know several pages ago.

I give up.

To know what ? about no compressed air ? that was an offshoot and never part of the discussion

[Simon]

Can't you just post a picture of the current setup, or describe what mechanical action needs preforming?  Three pages of teasing out what you are trying to accomplish is driving me crazy.  So far it sounds like you just need to add a .15 duty starter to your current rig to overcome static load.

At the very least , fill in the blanks:
Available Voltage: 24v
Available Start Current*:
Available Run Current*:
Final Shaft Speed: 10k-12kr/m
Ramp-up Time (max/min):
Starting Torque:
Static Load (y/n):
Average Run Time**:
Max Run Time**:

*   Is 750w your max power supply output?  Because measuring the actual mechanical requirements of a motor as a function of it's input current doesn't really work.
** "Continuous" can mean a lot of things.

Also wouldn't hurt to know about the thermal properties of the space the motor would occupy.  And the motor itself will fail at some point, never mind the chain of electronics/mechanics on either side of it.  The closest you'll get to non-fail is a highly redundant system.  You probably don't want to do this whole thing in triplicate.

In other words, take a picture of the current setup.

Unfortunately being a work project I can't tell all or post any pictures, sorry about that. The load is another fan but heavier than the original plastic fan on the motor which i think is upsetting the controller. Apparently "bolting 2 together" has not helped as they load share and we get the same speed as before and the same power and infact it's like having only one motor I'm told. I'm the electronics geek around although we usually subcontract electronics design and manufacture but I help with general specifications and making sure we are all on the same page. As a company we are mechanical only really with basic electrical knowledge but for me being the "local anorak"

"At the very least , fill in the blanks:
Available Voltage: 24v
Available Start Current*: battery - big ones so for our purpose unlimited
Available Run Current*: as above
Final Shaft Speed: 10k-12kr/m we are currently getting around 9000rpm but I'm told we are looking towards 10000-11000
Ramp-up Time (max/min): that's very secondary at this point, current setup is not quick anyway
Starting Torque: no idea but the load turns freely
Static Load (y/n): not sure what that means but don't know
Average Run Time**: no idea - as long as needed
Max Run Time** as above

[Kremmen]

Basically you have 2 alternatives:
1) A brush DC motor. Somewhere in the world a motor exists that can do what you need, if only someone can find it.
2) A "brushless DC" motor, i.e. in fact an electrically commutated 3 phase AC motor, either a BLDC or PMSM. The power output will be no issue but max RPM will be, and the input voltage is also quite a challenge.

Pros and cons as i see them:
A brush DC motor is a no-brainer to control so the electronics will be easy. If the numbers match you can let it run directly from the battery after the initial start&acceleration.
The downside is that it has a mechanical commutator with carbon brushes that require regular maintenance. The intervals can be long(ish) but they will be there. Other than that and properly applied such a motor can be quite reliable however.

A brushless motor has 1 moving part, making it essentially indestructible. So the only maintenance is related to bearing life which probably exceeds the lifetinme of the rest of the application with a comfortable margin.
The downside is that industrial quality BLDC motors with the required output power combined with this low supply voltage are few and far between. And you will not get 12 kRPM out of one so prepare to set up a gearbox.

Basically that's it. An induction motor would be the #1 candidate for this app, but lacking an AC/line voltage supply it is apparently not an option.
A stepper was never an option for many reasons that have already been pointed out.

I at least don't know of anything alse that could work.

[Simon]

A stepper is not what I was after but I was unsure basically the 3 phase motor with electronic comutation is what I'm after. i will have words and explain that for the power/speed a voltage increase is a must.

but to give an idea of how possible it is to get the power at this voltage I can quote the current motor we are using that is 750W off 24V and we have used a bosch fan in the past that if the dogs privates in commercial fans and will pull 48A at 12V (it was designed for high end audi's but we have used them often where power is required and locked them into max speed rather than used a controller).

I think the limiting factor is the speed here, we need more voltage for that or gearing. I personally would recomend a brushed motor unless lifetime and emi is a problem. At this point I need more details as I'm not heavily incolved in this project I've just pocked my finger in having seen them try to do it the hard way because they don't seem to know any better (as if I know much more which I don't but when dealing with electrical things i tuirn to electronic solutions before i look at mechanical ones where as they know mechanics so start making bits of metal to solve the problem first).

[poorchava]

There are two fundamental differences: steppers have thing inside what I can only describe as a circular bumpy spring washer which makes the motor "snap" to certain angles, BLDC motors don't have that. The second thing is that steppers have many more poles than BLDC's do.

I think for a stepper you can try A4989. It's specced up to 500W, but it's a N-MOS controller, and all the power elements are external, so in practice the limit may be much higher.

[Simon]

Right, I've done some convincing and explained that even if we get a custom solution they will be just bumping up the voltage and then driving as motor so now my fixes are 750W and up to 12Krpm and we shall see what voltage we need.

So spindle motor may do it.

Any people good at doing off the shelf DC/DC inverters ?

[ovnr]

Kelly Controls does some acceptable, fairly reliable motor controllers. Make sure you stay inside the maximum frequency range; they give up when you get too close to it.

[G7PSK]

except most automotive vehicles don't have compressed air.

Sorry I was under the delusion that this was some form of test rig. 

These people might have what you are looking for.
http://www.transmotec.com/brushless-dc-motors/no-gear/B86-Series.aspx?gclid=COvar5P1ibsCFZLKtAodyjAAew

http://yzkg.en.alibaba.com/product/413397031-201032127/KG315_high_power_brushless_DC_hub_electric_motor.html

http://www.adlee.com/brushless_dc_motors_with_dc_driver.htm

[Simon]

No it's to go in a vehicle, we are using a testing rig now so that we have things to bolt it all to but it will be vehicle kit. I'll take a look at those suppliers thank you, I'm also open to coupling 2 motors together.

[Kremmen]

[...]
Any people good at doing off the shelf DC/DC inverters ?

Technically that would be a boost converter then. Boosters take an input voltage and produce an equal or higher output voltage, up to something like max ~8 times the input voltage. Doubling or tripling the input voltage should be quite feasible.
One thing that makes this application easier is that you don't need galvanic isolation between input and output. That said it is still the case that implementing a proper production quality SMPS is no task for a novice. If you want i can do a quick check on what kind of efficiency you can expect from a converter like this. Offhand it could be between 80-90%.

[Simon]

anything like that would need to be bought off the shelf or designed for us. I don't do electronic design for the company other than mess with things and poke sticks at the subcontractors work.

[Kremmen]

anything like that would need to be bought off the shelf or designed for us. I don't do electronic design for the company other than mess with things and poke sticks at the subcontractors work.

My sentiment exactly I would not recommend you to try this, but wanted to mention it just to let you think about the feasibility of such a solution. I have no idea if you could locate a true off the shelf, standard device that can do this. Someone will make it for you certainly, but the price is a good question and certainly depends on the length of the production run and other variables.

[Simon]

Well most brushless motors and higher power stepper motors seem to work of fairly standardized higher voltages and 24v is fairly standard so i'm hoping that there is an off the shelf DC booster out there. I mean even 110V is quite common on motors and that coincides with mains voltage as does 220V. We could use a spindle cnc motor if we get the speed.

[oldway]

I remember that such high power and high speed brushless motors are used for flywheel energy storage.

[Kremmen]

Well most brushless motors and higher power stepper motors seem to work of fairly standardized higher voltages and 24v is fairly standard so i'm hoping that there is an off the shelf DC booster out there. I mean even 110V is quite common on motors and that coincides with mains voltage as does 220V. We could use a spindle cnc motor if we get the speed.

24V to 110-120V or 220V inverters should be findable without too much trouble.
You will certainly get the speed with a spindle motor, but there might be other issues. At least be careful of the motors with very high power density because they tend to be liquid cooled. Also at those power densities probably not very long life in continuous use unless you pay through the nose.

[Simon]

Well i have spoken to one supplier who is suggesting water cooling which is not a problem for us as we are primarily a heat exchanger manufacturer.

[Kremmen]

OK that should make things considerably easier then. Just include the cooling system in your reliability loop since failure there will kill the motor quite fast.

[strangelovemd12]

Unfortunately being a work project I can't tell all or post any pictures, sorry about

The downside of  a proprietary, private system is that you can't find much help for it in a forum, you have to actually know what you are doing.

The load is another fan but heavier than the original plastic fan on the motor which i think is upsetting the controller.

Taking off some plastic blades and replacing them with heavier metal ones sounds safe.

Apparently "bolting 2 together" has not helped as they load share and we get the same speed as before and the same power and infact it's like having only one motor I'm told. I'm the electronics geek around although we usually subcontract electronics design and manufacture but I help with general specifications and making sure we are all on the same page. As a company we are mechanical only really with basic electrical knowledge but for me being the "local anorak"

I guess we all have to trust that the mechanical know-how floating around your place is enough to keep this thing from turning into a bloodbath.  Spinning a fan up to 12krpm is not a trivial thing.  Even at a foot in diameter a 12krpm fan has tips moving at twice the speed of sound.

Available Start Current*: battery - big ones so for our purpose unlimited
Available Run Current*: as above

What?  Are they just some big car batteries with impressive CCAs?  Presumably they are lead-acid.  Are they deep cycle?  What do you need their service life to be?  How are they recharged?  Do you intended to run the fan while charging them?  All of these factors (and more) are important. 

Average Run Time**: no idea - as long as needed
Max Run Time** as above

Jesus Christ on a cracker.  Do you mean you guys might run it all the way through a single work day, or similarly short period of time?  Are you trying to run it for months on end?  And when it fails, what are the implications of it not running?

Final Shaft Speed: 10k-12kr/m we are currently getting around 9000rpm but I'm told we are looking towards 10000-11000

Why?  Because that is what will provide the airflow you need with the current blades?  Can those blades be run at those speeds?  Saying you need a "10krpm fan" is like saying you need a "48 pin MCU", it's verging on meaningless unless you provide a whole bunch of context.  (Same goes for the wattage rating frankly)

Starting Torque: no idea but the load turns freely

Well yeah, you can't put a brake on it if you want it to start.  You mentioned you had troubled getting it started with the current motor.  We need to know how much torque is required to start it.  Easily tested with a torque wrench.

Static Load (y/n): not sure what that means but don't know

It just means whether or not the load demands on the motor change.  Since we are apparently looking at a can, it would be a constant/static load.


I think what you really want is a fan.  Places sell them, go buy one.  You seem to be stuck on this avenue of tweaking this and that to boost your current system, and I think it's the wrong approach.  Define the airflow characteristics you need the device to accomplish, then find a fan that meets them.  My money is on it being faster and cheaper, and all the smart money is on it being a hell of a lot safer.

[Simon]

The downside of  a proprietary, private system is that you can't find much help for it in a forum, you have to actually know what you are doing.

So far people have been quite helpful and i've gleaned some useful information from this discussion.

Why?  Because that is what will provide the airflow you need with the current blades?  Can those blades be run at those speeds?  Saying you need a "10krpm fan" is like saying you need a "48 pin MCU", it's verging on meaningless unless you provide a whole bunch of context.  (Same goes for the wattage rating frankly)

the fan being driven has been specifically speced, it just needs driving. The power rating has been given in relation to the speed required, what is so out of context ?in effect I've practically specified a torque if you do the math. Measuring the torque required to turn a bearing is not something you do with a mechanics torque wrench

Jesus Christ on a cracker.  Do you mean you guys might run it all the way through a single work day, or similarly short period of time?  Are you trying to run it for months on end?  And when it fails, what are the implications of it not running?

no it won't run for more than a few hours but then that is enough to eliminate any possible use of something that can take a higher load for a short period. whatever is used needs to withstand it's specs for prolonged use not run up and run down again.

Taking off some plastic blades and replacing them with heavier metal ones sounds safe.

yes but when you have a motor that has a controller specifically designed to start with the torque required to turn a light plastic load it sort of doesn't like it when you put a metal pullet on it and use it to belt drive another pulley at a higher speed with a metal impeller on it particularly when the controller of the motor is designed to detect clogging f the original fan and not try to run the motor full blast and burn it out.

[strangelovemd12]

So far people have been quite helpful and i've gleaned some useful information from this discussion.

Yeah, there has been a lot of gleaning in this thread.

the fan being driven has been specifically speced, it just needs driving.

So someone built/designed/sold/picked a fan for you to use, saying "This will meet your requirements when spun at 11krpm,"?  Alright, that's what I'll go with.

The power rating has been given in relation to the speed required, what is so out of context ? in effect I've practically specified a torque if you do the math.

Information provided:
Around 10krpm
Please show you how one would "calculate the torque".  Make some guesstimates based off of your "750w" requirement?  It will depend on fan diameter, mass, system resistances, etc.

Measuring the torque required to turn a bearing is not something you do with a mechanics torque wrench

You missed my boat.  I was saying that since you had trouble starting the system it would be handy to find out how much energy was required to crank the shaft when cold.  A torque wrench is a quick and dirty method that has served me well many times.

no it won't run for more than a few hours but then that is enough to eliminate any possible use of something that can take a higher load for a short period. whatever is used needs to withstand it's specs for prolonged use not run up and run down again.

Awesome, this is the sort of info that's needed.  Obviously we were never looking at just tossing a starting motor on there, even if it would get you the 750w consumption you want.  Nor do you need a proper extended up-time motor.  Just a continuous duty 1.2 service factor 10krpm unit.  Dime a dozen.

yes but when you have a motor that has a controller specifically designed to start with the torque required to turn a light plastic load it sort of doesn't like it when you put a metal pullet on it and use it to belt drive another pulley at a higher speed with a metal impeller on it particularly when the controller of the motor is designed to detect clogging f the original fan and not try to run the motor full blast and burn it out.

I understand, you want your improvised system to spin harder faster longer.  Reading this thread I pray that you guys are big fans of cotter pins and Loctite.


Something like this might work:

http://www.grainger.com/product/2M191?gclid=CLDCuf3BirsCFZFxQgodvQ8AiA&cm_mmc=PPC:GooglePLA-_-Motors-_-Definite%2520Purpose%2520AC%2520Motors-_-2M191&ci_src=17588969&ci_sku=2M191&ef_id=UfKNxwAABdCufJqv:20131129172604:s

Requires 120v, easily achieved from your car batteries and a boost converter.  Looks like 10krpm@SF1, but easily derated for 12k.  40*C ambient rating mean you probably wont need cooling at SF1.2  (Also, notice the brush life?  Things like this are why running for a workday and running for an extended period make a huge difference.)  Is it powerful enough?  Who can tell.  You could certainly talk it into drawing 750w though.

[Simon]

I understand, you want your improvised system to spin harder faster longer.  Reading this thread I pray that you guys are big fans of cotter pins and Loctite.

Oh yes we are. What we need is more power. our 500w motor will not run quite fast enough and is uncertain on startup due to the way it's controller is configured and stops if it decides we have overloaded it so we want more power margin (as derating will give better durability) and control over the controller so that we can configure it for what we want.

i note your suggested motor states an rpm range of 10'000 - 20'000 and being a dc brushed motor I guess that are around 10-12000 rpm we are looking at 50% power which is 1/2 Hp where are we are looking for 0.8-1 hp (around 750W of power usage)

[Simon]

another option of course can be a motor with integrated gear box to increase speed, of course we don't want the gearbox failing

[strangelovemd12]

i note your suggested motor states an rpm range of 10'000 - 20'000 and being a dc brushed motor I guess that are around 10-12000 rpm we are looking at 50% power which is 1/2 Hp where are we are looking for 0.8-1 hp (around 750W of power usage)

Guessing sucks, read the datasheet.  Actually, read a lot of things.  Reading is fun.  I'll make that motor draw 1kw@0rpm if you really want me to.
And yeah, motors sometimes come with integrated gearboxes.  Was this not being considered before?   

Please just tell me you're working on an amphibious vehicle so that I can just laugh off the stress.  (Silence means "yes")

[Simon]

well there was no obvious datasheet but there's a catalogue page and yes it says full load speed 20-100% so if i may guess again they are saying that it will retain it's full power down to 20% speed.

another requirement is that they obtainable, we don't want to develop a system around a motor only to find that the motors are no longer made.

i honestly don't know what sort of vehicle this is. We don't work on the actual vehicles only supply subsystems for them and i'm not that deep into this project just looking at motors for them.

[strangelovemd12]

another requirement is that they obtainable, we don't want to develop a system around a motor only to find that the motors are no longer made.

Call Grainger.  Nice folks.

[G7PSK]

I assume from the speeds being talked about that we are talking about some form of radial rather than axial fan unless you have something akin to the front end of a gas turbine. So how much air flow is required  ? cubic meters a minute at what pressure, given that you can go to a fan supplier and they should be able to list your options.

[Simon]

if by radial you mean something akin to a centrifugal water pump yes. It's an industrial high pressure air blower, cast impeller and housing with it's own bearings.

[free_electron]

And if you read the brushless motor page it describes a stepper type motor with built in controller and says about as much:

"Brushless motors may be described as stepper motors; however, the term stepper motor tends to be used for motors that are designed specifically to be operated in a mode where they are frequently stopped with the rotor in a defined angular position. This page describes more general brushless motor principles, though there is overlap."

lemme guess .. crappypedia ?

a stepper motor has poles that line up with the rotor.
a bldc does not ! if a bldc rotor has 8 poles the bldc stator has 9 ! a bldc is always driven by a 3 phase current. a stepper motor not.

[strangelovemd12]

I assume from the speeds being talked about that we are talking about some form of radial rather than axial fan unless you have something akin to the front end of a gas turbine. So how much air flow is required  ? cubic meters a minute at what pressure, given that you can go to a fan supplier and they should be able to list your options.

I've tried this form of logic, it seems to be a nonstarter.

[Simon]

I don't know what air flow specs are required, I told them to use a vacum motor but between it being brushed and high voltage they weren't having it and it may not even of done what they wanted. I will inquire further on monday about that as an option as if it would work it would really simplify things but I think preference is on a brushless motor as there are less mechanical parts to worry about.

[SeanB]

You do get 24V vacuum motors, commonly used in commercial wet extractors as a safe vacuum which will not shock even if it gets wet. About the same power as a mains motor, and about the same life.

[Simon]

same life ? because I'd be awfully worried about that amount of current being commuted on those brushes at 24V. There are 110V vacum motors and at the end of the day it won't be too hard to get 24V to near 100V using a H bridge and voltage doubling twice or a proper inverter. I think the concern with brushed motors is also the emi emmisions but I'll explore the possibility on monday

[SeanB]

I have had them run for over 5 years used daily ( about 2-3 hours a day) and they are cheap, about 40 pounds each as generic spare motors. Took one out last weekend that died, it did about a decade of work, and had worn the brushes down to springs, along with the commutator being about 3mm thinner where the brush ran. Aside from that and noisy bearings the motor was still fine, just really dirty. Easier to replace than put new brushes and bearings in.

[Simon]

any idea where to get them ?

[SeanB]

Any place that sells white goods spares to the trade, look in your yellow pages. Bosch, kArcher and WAP are units that are using these, though they tend to use a 36VAC motor inside, so as to still have a  SELV  rating even when wet.

[Simon]

Yes I've just googled and seen them all over the net and looking just like a unit i used at 240V so we are already familiar with the basic type but we will need look at the power, I've seen a 500W one. Monday I will contact the people I got my 240V one from and see what they have.

[SeanB]

If you need more than the single one can flow and have the space then use 2 in parallel with equal length and bend ducting on inlet and outlet to allow them to share the load. Used on commercial machines to get the flow rate up. you also get 2 stage units that have 2 series impeller units that give extra suction within a single housing.

[Simon]

yes I had considered that and i suppose it adds safety margin in redundancy. the one i used in a hand drier is 2 stage and there are 3 stage ones too.