DIY Electric Car Forums banner

Why a gearbox in an EV?

23K views 46 replies 9 participants last post by  lazzer408 
#1 · (Edited)
Should a transmission be considered an "optimization", or a necessity in an EV that uses a series DC motor?

I thought one of the main plus points to a series DC motor was that it has lots of torque at low RPM. If that's true, then it would seem to reason that it could be connected directly to the drive shaft, and provide plenty of startup torque.

The top end RPM for these motors is well in excess of 2k RPM, which by my rudimentary calculations should allow over 70mph directly connected to the shaft, while still putting the 2500 RPM sweet spot in the "highway speed" category. Then again, maybe my calculations were screwed up.

I looked at the titles in the DIY EV wiki section of these forums and didn't see any articles that seemed like they would cover the reasons for having a gearbox, or especially a variable gearbox in a DC motor EV.

thanks in advance for edumacating me.

cheers,
jp
 
#2 ·
that assumption might be true... but you have to consider a few things....


What is the power needed to start a vehicle from stop and accelerate to 25mph? It requires a certain torque at the rear wheel. Is that torque, from the motor, to the differential to the axle going to be enough to accelerate it? Now, put that through a gear reduction and see how much torque you get. Lots more. Transmissions are torque amplifiers, and even though the motor may get all its torque at 0rpm, it may not be enough to push it very quickly without a transmission.

Another thing to consider, is that it takes TONS of amps to put max torque on a motor at 0rpm. Thats not very great for the battery pack.... as your peukert exponent rears its ugly head. The more amps you draw, the exponentially less your battery pack lasts.
 
#3 ·
that assumption might be true... but you have to consider a few things....

Another thing to consider, is that it takes TONS of amps to put max torque on a motor at 0rpm. Thats not very great for the battery pack.... as your peukert exponent rears its ugly head. The more amps you draw, the exponentially less your battery pack lasts.
Thanks. Very interesting. Does the controller have to put out high voltage to get tons of current at 0 RPM? Or is the impedance of the motor lower at low RPM? I guess I need to find some 4 dimensional graphs that show volts, amps, and torque at different RPMs.

jp
 
#4 ·
impedance of the motor doesn't change (well, actually it does vary with heat some, but not tons)....

volts is RPM, torque is current... thats the easiest way to look at it. At low RPMs, its lower voltage (well, lower duty cycle on the FETs/IGBT's, which shows as a lower average volage)....when accelerating, or holding speed, you'll have a torque requirement to achieve that, and that draws amps.

Hope that helps... do some research on this site about controllers, and search for "torque curve".
 
#5 ·
Having now looked at the torque curves for the Warp9, it looks like efficiency might be the biggest reason for at least a gearbox (even if not a variable ratio gearbox).

The motor seems to have a very definite efficiency sweet spot between 3k and 4k rpm. Outside that range, the efficiency drops like a stone.

So I'm thinking you need probably a two speed gearbox really. One speed for getting going, and another one for highway speed. The efficient range of rpm's is wide enough that you could probably cover from say 45mph to 75 mph with pretty good efficiency.

Forgive me if it seems like I'm answering my own post. I'm just thinking out loud.

cheers,
jp
 
#6 ·
Having now looked at the torque curves for the Warp9, it looks like efficiency might be the biggest reason for at least a gearbox (even if not a variable ratio gearbox).

The motor seems to have a very definite efficiency sweet spot between 3k and 4k rpm. Outside that range, the efficiency drops like a stone.

So I'm thinking you need probably a two speed gearbox really. One speed for getting going, and another one for highway speed. The efficient range of rpm's is wide enough that you could probably cover from say 45mph to 75 mph with pretty good efficiency.

Forgive me if it seems like I'm answering my own post. I'm just thinking out loud.

cheers,
jp
no problem :D
 
#7 ·
The average gearbox has approximately the following ratios:
1st- 3/1
2nd- 2/1
3rd- 1.5/1
4th- 1/1
5th- .8/1

So to put that in perspective, my miata comes stock with about 115hp and 90 ft lbs of torque. That gives me approximately:
1st- 345hp, 270torque 0-30mph
2nd- 230hp, 180 torque 30-45mph
3rd- 172.5hp, 135 ft lbs torque 45-60mph
4th- 115hp, 90ft lbs torque 60+ mph
5th- etc...

This is the power to the differential not the wheels, we are assuming that most people want direct drive to the differential or some other form of gear reduction.

To match the gear multiplied power without a transmission would take lots and lots of amps. Off the top of my head, a 9" motor will make around 100ftlbs of torque at 2000 rpm at 650A. (I could be wrong on the numbers, but the concept is the same). 2000 rpm direct driving the diff on my car takes you to 30mph. To make over 300 ft lbs of torque to match the first gear multiplied torque of a stock miata motor would take close to 2000 amps to the motor. To match the second gear multiplied torque of a miata would take 1300A at the motor. Third gear 650A, etc...

Those numbers are all possible with a Zilla and around 200V. The problem is thats alot of money to match the stock performance of a miata. Simply adding a transmission will drastically up your performance.
 
#8 ·
So to put that in perspective, my miata comes stock with about 115hp and 90 ft lbs of torque. That gives me approximately:
1st- 345hp, 270torque 0-30mph
2nd- 230hp, 180 torque 30-45mph
3rd- 172.5hp, 135 ft lbs torque 45-60mph
4th- 115hp, 90ft lbs torque 60+ mph
5th- etc...
Agreed on the torque and the cost of performance points. Gearbox does not multiply/divide HP though.;) It trades off rotational speed vs. rotational force, which, both being proportional to HP, cancel each other out and leave the HP unchanged.

cheers,
jp
 
#9 · (Edited)
Miata? Did I hear someone say Miata? :)

The point of a gearbox/transmission is to keep an ICE in it's power band. An ICE doesn't produce much torque at low rpm. A Miata motor might only make 25-50lbs of torque at 1000rpm where an electric motor could make 200+. In that case it already has 4x the torque. You don't need a trans to multiply that again. Electric motors, for the most part, have a flat horsepower curve. As rpms drop torque climbs, as rpm climbs torque drops. The exact opposite of an ICE.
 
#10 ·
While I was going from memory on motor torque curves-I happen to have a miata dyno graph right in front of me. The stock torque curve may be low, but it is amazingly flat. A stock miata produces 90 +-5 ft/lbs of torque from 1500-6000 rpm. An electric motor can triple that at high amps, but you cant carry those amps to very high rpms so you end up with a torque curve that crushes a miata at low rpms and barely matches it at high rpms.
 
#11 ·
While I was going from memory on motor torque curves-I happen to have a miata dyno graph right in front of me. The stock torque curve may be low, but it is amazingly flat. A stock miata produces 90 +-5 ft/lbs of torque from 1500-6000 rpm. An electric motor can triple that at high amps, but you cant carry those amps to very high rpms so you end up with a torque curve that crushes a miata at low rpms and barely matches it at high rpms.
Right. So if you have an electric motor that could do nutty torque but only at low rpm then you'd want to keep it there. That would mean shifting thru the gears rather quickly. Why not just leave it in 4th (1:1) by just eliminating the trans? Most of the motorcycle conversions I see are direct drive with only the final reduction in place. I had an EV Metro I and to be honest it went like hell in 1st-3rd but the R's came up quick. The car shot out of the hole like a terd on taco night I'll tell you what. When it all boils down to it, it takes X watts to push Y weight to Z mph. Do you know someone near you with an EV who maintained the gearbox so you can drive it to see how you like it?

I own a boosted '94 Laguna blue Miata btw. 275hp in 2300lbs is funnnnnnnnn :D I'm still working on my 500hp build. Must be a deathwish.
 
#12 ·
While I was going from memory on motor torque curves-I happen to have a miata dyno graph right in front of me. The stock torque curve may be low, but it is amazingly flat. A stock miata produces 90 +-5 ft/lbs of torque from 1500-6000 rpm. An electric motor can triple that at high amps, but you cant carry those amps to very high rpms so you end up with a torque curve that crushes a miata at low rpms and barely matches it at high rpms.
If one increased the voltage then not only will the series wound motor crush the Miata's stock engine torque at low rpms, but it will also crush it at high rpms. Remember increasing the voltage will also increase the torque-speed range of the motor (and therefore horespower). Don't forget too that the controller limits the current to one value under full throttle conditions that is not exceeded and so the torque curve will stay at one value over the RPM range up to the cut off point which can be moved further down into the higher RPM's by simply increasing the voltage to the motor.

http://img519.imageshack.us/img519/7722/torquelimitln5.png
 
#13 ·
Don't forget too that the controller limits the current to one value under full throttle conditions that is not exceeded and so the torque curve will stay at one value over the RPM range up to the cut off point which can be moved further down into the higher RPM's by simply increasing the voltage to the motor.

http://img519.imageshack.us/img519/7722/torquelimitln5.png
Interesting. That graph, as I read it, says a controller will current limit (and therefore torque limit) at low RPMs. I find that counter-intuitive. I would have expected current limit to come into play only at high rpms, rather than only at low rpms.

So without current limiting, with a fixed voltage, a series wound motor does have very low impedance at low rpms, thus high current at low rpms, thus the high torque at low rpms is explained. (but is limited by the controller)

I think I'm starting to catch on.

thanks,
jp
 
#15 ·
orthogonal whut? Does it happen at 5250 rpm? ;)

Sorry. Larry was on TV the other night. Forum politics need an ice-breaker once in awhile.

Stall (locked shaft) or peak current on a series would motor can exceed 1000s of amps. If the esc didn't limit current to the motor, one could easily damage the esc, motor, wiring, or batteries. I had a GE motor rated at 8hp 102amps 72v. I ran the motor at 120v and pegged my little 200a gauge on a regular basis. I didn't have current limiting on my controller yet. The motor's stall torque was rated 302ft-lbs at 72v 1200amps. At 120v you can bet it's closer to 400lbs. The car would smoke the tires in 1st-3rd. It just didn't care. :) Anything higher and I was afraid of grenading the gearbox.

A Curtis may be rated for 500amps and cost ~$500. On the other hand a Zilla 2k may handle 2000amps but you pay the insanely overpriced cost of ~$4500. This is where we see the world's problem show it's ugly head again. Company greed. -Someone- has -something- that no one else has so they take advantage of the customer and fill their pockets while they can. I'm a bit off topic again. My point was that the current limit is there for safety reasons. You have the choice of using a higher voltage pack and an esc, like the Zilla, with higher current capability, but it comes at a price to both your wallet and your run time. The only advantages I can see retaining a gearbox is the fun factor and possibly some range gain being able to accelerate quickly without as much current draw. This may help reduce plunket(?) effect on the batteries as well. This is only do to the minimal differences in the electric motor's efficiency at varying RPM. The difference is minimal compaired to an ICE. An electric motor rated at 5hp puts out ~5hp at 10rpm or 2500rpm. An ICE rated at 100hp only puts out 100hp at it's PEAK no matter what the torque is at any other RPM.
 
#16 ·
Check out Joule Injecteds dyno graph:


Hes running 2 9" motors, a Zilla, and 25 batteries for 300V (im assuming the motor only sees about 200V max. With all of that he has a MASSIVE 650 ftlbs torque at 1000 rpm, but can only muster 200ftlbs at 3000 rpm. In my Miata a direct drive setup with his driveline would be at 45 mph at 3000 rpm. So at 45 mph he would have 200 ft lbs of torque.................and because of the benefit of a gearbox, so would I with the wimpy stock ICE in second gear.
 
#17 · (Edited)
Check out Joule Injecteds dyno graph:
http://www.dragtimes.com/ima So at 45 mph he would have 200 ft lbs of torque.................and because of the benefit of a gearbox, so would I with the wimpy stock ICE in second gear.[/QUOTE]

I'm a sparky, not a gearhead, so correct me if I'm wrong on this...

Torque is needed to accelerate. HP is needed to overcome wind drag at high speeds. Your miata ICE might equal his WarpZilla at 45, but your HP peaks there (and again and again as you shift up) where as his maintains relatively constant (ok, it droops a little bit) HP up past 75 mph. So even though is torque is headed for the pavement at high rpms, his power is hanging in there pretty good.

I see your point though. Gearbox allows you to get similar performance for less money. It really seems like just a two position gearbox would be pretty close to ideal for a series wound DC motor configuration. It would allow you to start well with less current, and would allow you to get better efficiency at cruising speed, and a higher top speed.

cheers,
jp
 
#19 ·
Time for a hugfest! :D
Despite my arguments for keeping the stock transmission, I agree with you on a 2 speed being perfect, Im even building a dual electric motor no transmission miata for drag racing and autocrossing.

My reasons for going against my own advice:
-Lose 75lbs for the transmission
-Lose 75lbs for the diff (im doing a 1 motor per wheel gear reduction mess)
-Gain tons of room for batteries (frees up the trans tunnel and all of the engine bay
-perfect limited slip
-the motors are small so I need 2, and they dont lend them self to 2 inline.
-I seem to have access to moderately inexpensive yellow tops. They are low ah each so I need a bunch for decent range.
-wont have to worry about a hipo clutch
 
#21 ·
Im up in the air about the controller situation. I have about 8 regular contactors and 4 reversing contactors so a contactor controller is an option. I may hit up the guy on ebay with the 72V 400A curtis controllers and use 2 of those with either a contactor bypass for 144V at full throttle or a lever I pull for 144V bypass. Another option is for the open source motor controller guys to come up with a design I can solder together out of surplus parts.

The single reduction drive is probably going to be sprocket/chain drive to each inner CV with a sprocket machined to fit the CV flange. I may try to incorporate the gear drive setup out I saved from the forklifts, but that looks more complicated.
 
#24 ·
I have about 8 regular contactors and 4 reversing contactors so a contactor controller is an option.
Translation please? Us new guys don't know a contactor from a compactor. It sounds kinda like a switch, maybe even a heavy current relay??????

If so, that would parallel an idea I've been toying with of replacing controller with just a switching mechanism to vary how much of the series battery stack one connects to the motor. Is that what a "contacting controller" is???

thanks,
jp
 
#23 · (Edited)
Gearless motors used in elevators are about that much torque although the horespower is more along the lines of 18 or greater horespower. The poles make a difference in torque for 3-phase AC motors, but also effect speed assuming frequency is held constant. It's not impossible to have a 720 pole 3-phase motor rated at full load horespower of 5, but it is impractical........


Here are the specs to gearless elevator motors:

http://www.imperialelectric.com/pdfs/ac_gearless.pdf



Interesting. That graph, as I read it, says a controller will current limit (and therefore torque limit) at low RPMs. I find that counter-intuitive. I would have expected current limit to come into play only at high rpms, rather than only at low rpms.

So without current limiting, with a fixed voltage, a series wound motor does have very low impedance at low rpms, thus high current at low rpms, thus the high torque at low rpms is explained. (but is limited by the controller)

I think I'm starting to catch on.

thanks,
jp
It has to do with whats called BACK EMF voltage that the motor generates when it turns that opposes the supply voltage. Really what runs a DC motor is the differential voltage (Vsupply - Vgen). It's when the back emf starts to get so high that the differential voltage is not at the value needed to maintain the same current draw so it starts to fall off as it normally would if just hooked up to batteries with no current limit. Controllers limit current by reducing the average voltage of Vsupply such that the current flowing that is measured by some current sense element whether it be a shunt or current mirror is at the value that was set by the user via programming of the controller's parameters.
 
#26 ·
Translation please? Us new guys don't know a contactor from a compactor. It sounds kinda like a switch, maybe even a heavy current relay??????
A contactor is a device that is designed to make or break large current loads like motors without destroying itself in the process unlike a relay which is designed for small loads. Contactor's as well have two break points where as your typical SPST relay has one break point where the other side of the moving part is secured to a pivot point.

If so, that would parallel an idea I've been toying with of replacing controller with just a switching mechanism to vary how much of the series battery stack one connects to the motor. Is that what a "contacting controller" is???

thanks,
jp
Why even do this? It would be a good idea in the late 1800's to early 1900's, but it's 2008...Well it would be able to survive a nuclear blast's EMP wave since it is mostly electromechanical, but your not trying to build a navy submarine are you.:D
 
#30 ·
If you use a controller, you ramp up the voltage, and the motor may still take the same amount of current, but at a lower voltage..... but in both cases you'll be accelerating the same.
Not true because the current will be less with the electronic controller. With no electronic controller the full 144 volt pack is being applied to the motor and thus any current limitation is provided solely by the sum of the resistances of the wires, batteries, and motor at t = 0 (motor at stall). The torque will be insanely high with no controller assuming the motor was not destroyed in the process of such a high current surge. With a controller the current that flows through the motor circuit will not exceed what the controller limits it to.
 
#31 ·
When you chop 144vdc to a motor, at say 50%, you are indeed sending 144v pulses 50% of the time. The motor itself only 'sees' 72v across it's inductive windings. You will actually get more torque this way then if you gave it 72v at 100%. During switching, the counter emf (flyback) from the motor's windings continues to flow through diodes in the controller even though there is no draw from the pack. This maintains the current flow in the motor keeping the field from collapsing until the next pulse.

$.02
 
#32 ·
When you chop 144vdc to a motor, at say 50%, you are indeed sending 144v pulses 50% of the time. The motor itself only 'sees' 72v across it's inductive windings. You will actually get more torque this way then if you gave it 72v at 100%. During switching, the counter emf (flyback) from the motor's windings continues to flow through diodes in the controller even though there is no draw from the pack. This maintains the current flow in the motor keeping the field from collapsing until the next pulse.

$.02
The duration of the ON time of the pulses is so small at 15 KHz that it's effects of "sending 144 volts at 50% duty cycle is not the same as 144 volts at 100%. This is simple integral calculus of finding the area under a rectangular curve wave or in this case a sqaurewave in which case the average or "area" under the waveform is 72 volts. I know about flyback topologies, but I disagree that this gets you more torque than simply applying pure 72 volts DC with no current limit.

You do know how current limit works right? The controller basically has a current sense element like a shunt or current mirror that is monitored by comparator circuit that compares the reading from the current sense element to that of what the set point is that the user wants the current limit to be at. If the value is exceeded then the comparator's output will be in its active state (it could be active high or active low) that will send this state to the PWM section of the controller to turn off the MOSFETS completely. This is called "dead time" and the result is the average voltage will be less than what you would expect at 50% duty cycle. So at t=0 or motor initially not rotating there will be lots of dead time in the pulse waveform in a current limited controller versus 72 volts directly which has the various voltages drops across the various electrical resistances of the motor circuit to limit its current.
 
#33 ·
I had this same question. I had thought of doing a direct drive with a big reduction diff. Then I started thinking how my ICE engine could start the car out from a 3rd or 4th gear and drive around all the time but I don't do it. Sure Electric motors have more low end torque but if I could get more.....But if there is more torque then the tires could handle it's useless. I didn't know what to do.

Then I was reading about the White Zombie (for those that don't know a really fast street legal dragster.) which is direct drive and I realized that the owner John wayland's daily driver EV had a transmission in it. So he converted the dragster but not the daily driver despite the daily driver being reworked a few times for better performance. So I e-mailed him and asked him why. In a nut shell he said with the same motor and batteries the EV with the trans will be faster. I'm all about speed and I'll trust someone that's been building EV's for years and years and setting records.

My EV conversion will have a manual trans and I'll even keep the clutch. The extra mass hurts but did I mention I'm all about speed :D
 
#35 ·
My EV conversion will have a manual trans and I'll even keep the clutch. The extra mass hurts but did I mention I'm all about speed :D
I think that's a wise choice. If your lazy some day and don't feel like shifting just pop it into 3rd or whatever gear suits the driving conditions. If you feel like showing those tires who's boss then by all means set them alight with 1st. :D
 
#34 ·
The current limit does NOT reduce the voltage. It simply backs off the PWM percentage. The voltage "output" from a controller is the same as your pack voltage -period-. The controller is only a switch connecting the motor- to battery-. That's it's main purpose. It has 2 states. On and Off. The duty cycle is the "on time" measured in percent. It chops a 120v pack voltage into pulses but the pulses are 120v during the on state and it's only the inductive characteristics of the motor itself that reduces the shaft speed.
It takes time to build a magnetic field in the windings and it takes time for that field to collapse. The pwm switching frequency is chosen to have minimal fluctuation in the magnetic field. Imagine a bowling ball to represent the magnetic field of a motor. Now imagine that bowling ball hanging from a bungie cord in your hand. Your the pwm. If you move your hand up and down slowly the ball will move. If you move your hand up and down quickly the ball will not move. That's the idea when choosing the switching frequency. The inertia of the ball will hold it in place. If the time you pull up on the bungie is longer then the time you relax, the ball will rise and visa-versa. That it the pwm percentage of on vs. off. That's the best analogy I have for you.
There IS more torque from 120v@50% then from 72v@100%. 2x Get on a dyno and try it. There is also less current draw but I think we all knew that.
 
#36 ·
There IS more torque from 120v@50% then from 72v@100%. 2x Get on a dyno and try it. There is also less current draw but I think we all knew that.
I'm not gonna say you're wrong about that (just yet), but I will say it don't make no damn sense. :)

If that's the case, then what percent at 120V will equal 100percent at 70V? (feel free to approximate). And please explain your answer.

thanks,
jp
 
#38 ·
The current limit does NOT reduce the voltage. It simply backs off the PWM percentage. The voltage "output" from a controller is the same as your pack voltage -period-. The controller is only a switch connecting the motor- to battery-. That's it's main purpose. It has 2 states. On and Off. The duty cycle is the "on time" measured in percent. It chops a 120v pack voltage into pulses but the pulses are 120v during the on state and it's only the inductive characteristics of the motor itself that reduces the shaft speed.

It takes time to build a magnetic field in the windings and it takes time for that field to collapse. The pwm switching frequency is chosen to have minimal fluctuation in the magnetic field. Imagine a bowling ball to represent the magnetic field of a motor. Now imagine that bowling ball hanging from a bungie cord in your hand. Your the pwm. If you move your hand up and down slowly the ball will move. If you move your hand up and down quickly the ball will not move. That's the idea when choosing the switching frequency. The inertia of the ball will hold it in place. If the time you pull up on the bungie is longer then the time you relax, the ball will rise and visa-versa. That it the pwm percentage of on vs. off. That's the best analogy I have for you.
The "backing off the PWM percentage" sounds to me that you are varying the ratio of ON time to the OFF time which in effect is reducing the voltage AVERAGE. I promise you if you dynoed a series wound motor with a no current limited power supply it will have torque to the moon compared to current limiting of a motor controller.

This statement: "There IS more torque from 120v@50% then from 72v@100%. 2x Get on a dyno and try it. There is also less current draw but I think we all knew that..", is contradictory to the very definition of how a DC motor develops torque. If you have studied DC motor behavior then you should realize the torque developed at the motor shaft is proportional to the current. So more current will equate to more torque in a motor. So at stall or t=0 the motor (no back EMF) that has 72 volts applied to it will generate greater static torque force assuming no damage from the high current surge than a motor at 60 volts average (120 volts at 50% duty cycle squarewave) which a motor controller will not apply to a stalled motor anyways:rolleyes:, else it would destroy the MOSFETS and/or the motor.
 
#39 · (Edited)
The "backing off the PWM percentage" sounds to me that you are varying the ratio of ON time to the OFF time which in effect is reducing the voltage AVERAGE. I promise you if you dynoed a series wound motor with a no current limited power supply it will have torque to the moon compared to current limiting of a motor controller.

This statement: "There IS more torque from 120v@50% then from 72v@100%. 2x Get on a dyno and try it. There is also less current draw but I think we all knew that..", is contradictory to the very definition of how a DC motor develops torque. If you have studied DC motor behavior then you should realize the torque developed at the motor shaft is proportional to the current. So more current will equate to more torque in a motor. So at stall or t=0 the motor (no back EMF) that has 72 volts applied to it will generate greater static torque force assuming no damage from the high current surge than a motor at 60 volts average (120 volts at 50% duty cycle squarewave) which a motor controller will not apply to a stalled motor anyways:rolleyes:, else it would destroy the MOSFETS and/or the motor.
Less current draw from the battery but as I explained there is more current in the motor then what your drawing from the battery when switching.

I'm not speaking about current limiting. All things equal. If it's 200a@72v then I'll assume 200@144v@50%

As I've also explained before, the controller is only connecting M- to B-. It's a low side switch. What is the B+ supply to the motor? What is the voltage at the M- terminal during the off state?
 
#40 ·
I'm not speaking about current limiting. All things equal. If it's 200a@72v then I'll assume 200@144v@50%
It makes no difference above. The stall torque will be the same. We can argue about this all day.....Why not go ahead and throw two series wound motors on a dyno designed for electric motors (gas engine type dynos cannot catch low RPM torque).



As I've also explained before, the controller is only connecting M- to B-. It's a low side switch. What is the B+ supply to the motor? What is the voltage at the M- terminal during the off state?
LOL... I know where you are going to go with this so I am will play along...The B+ connects internally to the cathode side of the freewheeling diodes and to one of the motor's terminals and voltage there is at B+ with respect to ground. This is where you "think" you are going to say AHAH I got you with this next question... So the voltage M- WILL NOT be what you are thinking. I bet you're thinking that the B+ and the back EMF of the inductance of the motor will add together and the voltage at M- with respect to ground will be higher when the switch is off. Guess what? WRONG! The freewheeling diodes clamp any back emf caused by the inductance of the motor....(the other back EMF caused by motor acting as a generater is in the other polarity and is not clamped)...
 
#41 · (Edited)
I wasn't going there at all. I know how controllers work, I build them. I was just trying to let people think about it for themselfs. B+ is B+ say 120v. The motor always has 120v to it. B- is 0. M- gets connected to B- in pulses. It's on or off. Nothing in between (other then rds-on drop). I'm trying to clear up the confusion about controllers controlling the voltage to the motor. It doesn't exactly work that way. I could get picky and say emf is higher then pack voltage by the diodes forward drop but that's getting anal. :D

If you understand how the emf works it's easy to see why a higher pulsed voltage creates more power then steady lower voltage. Motor rpm plays into this. It's at low rpms you'll get more torque. That's when forward, or reverse flyback, current is highest. 4qd has a nice explination of it somewhere on their site. Let me see if I can find you guys a link to it. bbl.

PWM:
120v@200a = 24,000w
Pulse sent to motor 120v@200a (during on state 50% of the time) so lets say 60v*200a=12,000w
Pulse returned from motor = 12,000w (during off state 50% of the time)
12,000w+12,000w=24,000w maintained in the motor's windings. (60v*(200a[fwd]+200a[rev])=24,000

DC:
60v@200a = 12,000w

As I said there is 2x the current in the motor windings, then being drawn from the battery, when at 50% PWM. Current=torque :)

To quote 4QD:
"...if the drive MOSFET is on for a 50% duty cycle, motor voltage is 50% of battery voltage and, because battery current only flows when the MOSFET is on, battery current is only flowing for 50% of the time so the average battery current is only 50% of the motor current!"

http://www.4qdtec.com/pwm-01.html

Note: 4QD's "A" and "C" waveforms are backwards from how they explain it. Maybe they had an inverting mosfet driver and sampled the input pulse for the image. :shrugs: They are also showing a mosfet in place of the flyback diode. It's treated the same.

It makes no difference above. The stall torque will be the same. We can argue about this all day.....Why not go ahead and throw two series wound motors on a dyno designed for electric motors (gas engine type dynos cannot catch low RPM torque).





LOL... I know where you are going to go with this so I am will play along...The B+ connects internally to the cathode side of the freewheeling diodes and to one of the motor's terminals and voltage there is at B+ with respect to ground. This is where you "think" you are going to say AHAH I got you with this next question... So the voltage M- WILL NOT be what you are thinking. I bet you're thinking that the B+ and the back EMF of the inductance of the motor will add together and the voltage at M- with respect to ground will be higher when the switch is off. Guess what? WRONG! The freewheeling diodes clamp any back emf caused by the inductance of the motor....(the other back EMF caused by motor acting as a generater is in the other polarity and is not clamped)...
 
#42 ·
I wasn't going there at all. I know how controllers work, I build them. I was just trying to let people think about it for themselfs. B+ is B+ say 120v. The motor always has 120v to it. B- is 0. M- gets connected to B- in pulses. It's on or off. Nothing in between (other then rds-on drop). I'm trying to clear up the confusion about controllers controlling the voltage to the motor. It doesn't exactly work that way. I could get picky and say emf is higher then pack voltage by the diodes forward drop but that's getting anal. :D
Actually forward voltage diode drop will not be higher than pack voltage. It will be at probably no more than 1 volt under high current flow into the diode. When a diode clamps, it will be at the voltage potential of the diode not the back EMF supply....




If you understand how the emf works it's easy to see why a higher pulsed voltage creates more power then steady lower voltage.
While the instantaneous current in the motor will be high every time the pules is ON, (and same when off, since inductors try to circulate the same current) it averages out the same as having 72 volts of direct battery connection... You have studied Calculus right?


Motor rpm plays into this. It's at low rpms you'll get more torque. That's when forward, or reverse flyback, current is highest. 4qd has a nice explination of it somewhere on their site. Let me see if I can find you guys a link to it. bbl.
The back emf caused by the motor turning is RPM dependent. When the RPM's go low caused by a load upon the motor shaft, the the back EMF will be less and therefore the differential voltage (Vsupply-VEMF) will be higher. Since resistance stays fairly constant the current as well will rise resulting in more torque.....You still have not proved anything. Current will be just as high with 72 volts applied to the motor as it would with 144 volts at 50%. Those high current spikes caused by the 144 volt peaks average out the same as having current flow from a 72 volt battery pack..

Again Calculus is your friend, use it..
 
#43 ·
You still have not proved anything.
I don't need to. I already know. If you don't understand I'm sorry. I tried my best to explain it. I just sold the only matched set of motors I had or I could have run them side by side (one pwm, one dc) and posted a video for you. It doesn't take a calculus major to visualize what's happening. Try reading 4QDs site about it. It might make more sense then I do.

emf is higher under load btw. Just as forward current is higher at low rpm, so is reverse current.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top