Multi-speed DC brushless motors?

[RichLeimbach]

I am currently using a brush-type DC motor to drive a relatively high inertia flywheel from 0 to 15K RPM in approx. .5 sec with a DC source (battery).
The problem I am running into is that with a single, always-on gear ratio, we spend too much time in the high current, low speed, low efficiency range (< 1/2 the max motor speed) and I'm having durability and heat related problems with the current motor. Electrically modulating the applied voltage does not help, as it just increases the time the motor spends in this inefficient range.
Besides finding a way to mechanically shift gear ratios, is there any way to increase the low-end efficiency of a DC motor? I am thinking of something along the lines of 2 speed AC induction motors, but in DC form. Any ideas?

 

[ve7brz]

Could you give us an idea as to the magnitude of the load? Perhaps info on the motor currently in use or an estimate of the inrush current?

 

[djs]

Hello;
Look into field weaking a shunt type DC motor... In any case if you limit the motor current to the rated max, you should be able to control the overheating.

 

[RichLeimbach]

The load is all inertial (~15x the rotor inertia), there is no applied torque outside of the flywheel during the ramp-up. Motor is a 35mm can and we are throwing a max of ~ 100A through it at stall. Basics on the motor: R= .15 ohm, Tmax= 8 in-LB, RPM max = 25K.

DJS... I did some calcs. and ran some tests re: current limiting in this application and found that it actually INCREASES motor heating because as the applied voltage is modulated, the motor spends more and more time in the hugely ineffiecient low RPM range. Hitting it hard at the start gets it out of the low end and decreases the total heat generated in one cycle.
I'll look into field weaking a shunt type DC motor. Thanks for the suggestion.

 

[ve7brz]

Thanks for the motor info; it gives a good indication of the motor size. One approach I would suggest considering is to use a three phase AC motor. I know this sounds ludicrous, but there are very small (about the same size as your 'can' motor) permanent magnet three-phase AC motors which would be far more efficient than your 'can' motor.

The controllers are designed to operate on anything from 6 to 36 VDC. The controllers use a 'soft start', limiting current flow to the motor as it comes up to speed. I don't know offhand what the ramp-up timing is but it's quite short.

Like any small motor they don't have a great deal of torque, although more than a brushed motor. On the other hand they're good for 50 to 60 grand so a 3:1 or 4:1 gearbox is feasible. There are also a couple of 'rotating can' versions which offer much greater torque at lower RPM.

The main drawback of this approach is a high initial cost, however the lifespan would be orders of magnitude greater than a 'can' motor. If you want more details I can dig out a few websites where this equipment can be found.

Good luck, Bruce

 

[jbartos]

Suggestion: How about getting the dc motor with appropriate parameters for your load?
Please, could you post the motor nameplate data and shaft load data?
The motor-load marriage should not be a major problem; however, it may lead to a custom built motor.

 

[jbartos]

Small encore: Visit

http://www.electrosales.com/bodine/DcInformation.html

etc. for typical dc motor data

 

[RichLeimbach]

Thanks for the info.
FYI: Motor performance (with a PMDC motor) is already very close to optimum sizing for this app. (already custom built). Making it larger (size-wise) is very difficult/ expensive because of packaging issues, but would admittedly help with some of the durability issues associated with the brush. However, total efficiency (another big piece of the puzzle because of the battery power) is not really increased because the motor is still driven through the standard efficiency curve during the ramp-up.
Decreasing current through the motor (either by soft start or increase dynamic resistance) decreases performance (time from 0 to 15K flywheel RPM) and does not reduce efficiency problems (motor spends even more time in the sub-50% speed range).
Increasing current through the motor (by decreasing dynamic resistance) past its current point also leads to problems because of the voltage drop associated with the internal resistance of the battery and electronics attached. 100A is about the max. we can expect.
Motor RPM is chosen for max efficiency (gear ratio tops the flywheel out with the motor at approx 90% of max speed). Any lower in max speed and we won't get there. Any higher and we spend more of our time in the low efficiency range and increase total heating during one cycle.
The main issue is the poor effiency of the motor (ANY PMDC motor) in the low RPM range. Because our load is all inertia and we are directly connected, there is not much way around having to load the motor heavily when it is still in the sub-50% RPM range. I'm trying to figure out a way to increase output power at the low end above the standard parabolic output power curve. Any suggestions are appreciated.
ve7brz... I'll check into 3 phase AC. Thanks for the suggestion.

 

[ve7brz]

Hi again,

Here are a couple of sites you might check out for brushless AC motors. You can find motors here which fall within the constraints for direct drive. My main concern would be the relatively high idle current given the low impedance of the windings.

http://www.modelmotors.cz/index.php?id=en&nc=produkty_vypis&url_ps=axi_2808_16&id_odkazy=m_ac

http://www.aveox.com/index.html

Ignoring viscous coupling which I'm sure you've considered, I can think of two other avenues of investigation:

Use a flywheel and an electrically operated clutch. The flywheel is driven by the motor to a velocity greater than 15 grand. The inertia of the flywheel must be such that when the clutch is engaged the velocity of the combined mass drops to 15K. Simultaneously the motor is throttled back to maintain the 15K until the clutch is disengaged, at which time motor accelerates the flywheel back to working speed.

Stupid idea number two: Use an automatic two speed gearbox. This is an idea I came up with about 40 years ago. A computer simulation at that time validated the concept.

The basic two speed unit integrated a planetary gearset, spraque clutch and centrifigal clutch. The motor drives the pinion and the planet carrier the load. The outer (inside tooth) gear is prohibited from counter-rotating by a sprague clutch integrated into it, working against a stationary outer ring. The centrifigal clutch located on the output or integrated into the planet carrier engages the outer gear when the carrier reaches a given velocity. This unit could be cascaded to give any number of stepped ratios.

I don't know if anyone has ever built such a mechanism but it looked good in the simulation. I hope I'm not using too much space or time on this thread with hare-brained ideas. If I am please tell me.

Bruce

 

[RichLeimbach]

Bruce,
Good input. We ARE looking at using a 2 speed gearbox with a centrifugally engaged high gear. Calculated efficiency increases in this set-up with a 2:1 difference between low and high gear is ~30%. FYI: similar products are sold commercially from several maunfacturers for use in RC cars. Method of operation consists of 2 pinion gears directly connected to the flywheel. The high gear is centrifugally engaged to the motor shaft above a trip speed and the low gear is allowed to over rotate on the shaft when the high gear is engaged by use of a one-way bearing. Pic of the parts of one model can be seen here:

http://www2.towerhobbies.com/cgi-bin/wti0001p.pgm?Q=1&I=LXBTB6&P=7

I'll have to think a little about the efficiency of the flywheel and electrical clutch idea. By the way, I did model the viscous coupling and found that it just transferred heating from the brushes to the other end of the shaft, without much positive result.
Thanks for the ideas.

 

[jbartos]

Suggestion: This topic can also effectively be discussed under mechanical engineering since this section is reserved for electrical motors and associated electrical enginering. The interface solution in terms of gear boxes is on the mechanical side. This could lead to faster solution for you since the average electrical enginering department passes this type of integration to the mechanical department or systems department. It also indicated above that the motor is optimally sized and custom built. Incidentally, is that custom motor built low speed? What seems to be so secret about the motor nameplate data? The cooperation just is not there with volunteering expertise.

 

[cbarn24050]

Hi, if you use a dc motor with a field winding you can change the &quot;electric gear ratio&quot; by altering the field current. You use a high field current at low speed to give you high torque and low field current at high speed.

 

[RichLeimbach]

Jbartos... Thanks for the suggestion. I'll post in the Mechanical section too. The reason I posted here was that I already had a number of mechanical ideas in the works, but wanted to know if there were some some possible motor / controller ideas out there that may be of help also. FYI - motor is built by Johnson (all the other particulars needed to develop the motor curve are in my Oct 30 post, except for free run current which is 1.5A @18V).

 

[jbartos]

Suggestion: If RPM needed are up to 15K, then the motor should be designed for RPM=15K. What about increasing the motor shaft horsepower, i.e. oversize the motor?

 

[kamenges]

I'm kind of new to motion systems and most of my experience is in AC induction and brushless motors so unfortunately this is a question as opposed to any helpful comments.
It is stated up front in this thread that the current DC motor used in this application is having problems because it &quot;spend too much time in the high current, low speed, low efficiency range (< 1/2 the max motor speed)&quot;. While I have heard that brush commutated DC motors should not be run at extended low speeds due to localized commutator heating, I have never heard that they inherently generate more heat at low RPM. Granted, the efficiency number will be low because the motor RPM is low and the mechanical energy generated is low relative to the losses. But the thread seems to point to increased heating Is this heating due to full voltage starting and the accompanying high starting currents? If this were the case all motor types would suffer from this loss. is there something inherent in the brush DC design that casues this inefficiency?

Sorry for the sidetrack, but any insight would be helpful.

Keith

 

[UKpete]

Perhaps unlike mechanical transmissions, electric motors generally don't work at high efficiency across a wide range of operating conditions including speed - I think this is true of all motor types.

There is a similar problem in electric traction systems. The overall efficiency is reduced and the motor rating, determined by the winding temperature rise, is dependent on the RMS current taken over a period less than the thermal time constant (thus the repetition rate of starts can have a large influence on rating).

I don't think there is any reason to think that a different motor type to brushed dc would significantly improve overall efficiency, except that there are moderate electrical and mechanical losses with carbon brush/commutator systems. The usual reasons for going to brushless (ac or dc, the differences are a bit subtle) are things like higher speed operation because there is not commutator, low EMC, improved power density and some improvement in efficiency.

Incidentally, you're original post says that the acceleration time is 0.5secs - I presume you mean 5secs.

 

[RichLeimbach]

UKPete... No, we really do it in 1/2 sec. FYI - flywheel has the approximate inertia of a 2&quot; dia x .5&quot; wide steel disk (if this helps you visualize the sizing any).
The problem I have is that motor starts from stall and has very bad efficiency for a ~ 1/2 of the cycle. For example, during the time from 1 to 1K RPM, I put in an average of 1000 Watts from the battery, but receive only an average of 60 Watts of useful output. The rest is wasted as heat. Although the timeframe is short, we're talking about a lot of heat generation.

Kamenges... You're right, any brushed vs. brushless standard designs should not make any huge efficiency or heating differences here. What I was looking for here were any non-standard controls or motor construction ideas that might improve low end efficiency or bump up low end power.

Jbartos... The current motor is ~25K on a 1.59 gear ratio = 15.7K max possible RPM at flywheel.
How would oversizing the motor help me? I understand that I could get more output power, but it would come at the expense of more input power, since the same basic efficiency curve would still be in effect. Please expand on this.

Thanks for all comments / suggestions.

 

[cbarn24050]

Hi, I find those numbers hard to believe, if they are correct then you are misusing the motor badly.Try using a chopper to limit the current to something sensible.

 

[jbartos]

Suggestion to RichLeimbach (Mechanical)Nov 2, 2002:
To abandon the gear, the motor should be build for about 15K RPM. One could reason the oversized motor following way, for simplicity: Supposing that one looks at motor product catalog, e.g. Baldor Motors and Drives Data Catalog 502, April 1992 &quot;AC Motor Date Catalog&quot;
HP RPM EFF Type
1.5 3450 75.5 0516M
2 3450 78.5 0524M
1 1140 75.5 0524M
It is clearly seen on the AC motor example that the motor design makes a big difference in RPM, and efficiency varies marginally. Therefore, your DC motor custom builder or designer should go back to basics and design the DC motor of truly optimal parameters for your application or find a better DC motor designer.

 

[RichLeimbach]

cbarn24050 ... I know we are abusing the motor badly. However, the trade-offs associated with current limiting to a low current (<30A) are not acceptable with regards to performance. The current design actually functions quite well, but runs hotter and is more inefficient than I would like. As discussed previously, current limiting doesn't really make either of these two issues any better (although it will decrease or eliminate the slight electrical errosion of brushes from sparking at some point).

Jbartos ... This is exactly my point. Increasing motor power does not give me any increase in efficiency. Since I need a certain total energy to come out of the motor and transfer to the flywheel, I will generate almost exactly the same amount of heat using any of these motors. The only difference is that with the more powerful motor, I am generating all of the heat energy in a shorter period of time.
Comparing this to the current situation as described above: Doubling the motor power would roughly cut my ramp-up time in half, but would mean I am putting 2000 watts of energy in and only getting 120 watts out to go from 0 to 1K RPM. I just doubled both sides of the equation. No gains, it just compressed the whole event.
I do agree that if the gear ratio was 1:1 instead of 1.59:1, our motor should be set at ~ 15K max.