I confirm... IRFB4110 Mosfet CAN BLOW !!!

[The7]

my setup was:
I limit to 75A
Vbatt =65V
Wire lengh between motor and controller 2 ft of 3x 12 AWG
battery volt at 75A: 63V
"motor wound resistance: 0.142ohms"
4110 mosfet RDS on: 0.0045ohm
stoped to one phase only time estimated 3 sec

The C-motor is Y-connected.
Do you mean that 0.142 ohm is the resistance between one phase and the star-point?
If so, the maximun value of the motor current at stall will be calculated as :

62V / ((2s2p 4110 mosfet) + (2x 2ft of AWG12) + (2x 4ft of AWG 10) + (r of switch contactor) + (2 x r phase)) = current to one phase

62V / (0.0045ohm + 0.006ohm + 0.008ohm + 0.01ohm + 0.284ohm) = I

62V / 0.313 ohm = 198 A (Not 363 A)

Note: At any conducting instance, the battery voltage is only applied across any two phase windings in series when using 6-step controller.

 

[The7]

Taking the maximun motor phase current to be 198A:

Assume that the motor is stalled.
1) If the throttle is at 100%, the average battery current would also be 198A if there is no battery current limit.
But if the battery current limited is set at 75 A, the current limiting will activate to maintain the average conduction period to be 75/198 (say about 38% [NOT 25%] ). So the average battery current is still 75A . But each motor phase winding still experinces a maximun current of 198A.

2) If the throttle is less than 38% , the average battery will be less than 75 A. So the current limiting will not activate at all.

3) The current limit will activate if the throttle is greater than 38%. Each motor phase winding would still experinces a maximun current of 198A.

The heating power of the motor for condition (1) and (3) is 62x75 = 4650 W.

Edited: 75/198 should be about 38%( NOT 25%). Sorry for my calculation.

 

[dirty_d]

if the duty cycle was 25% the motor current couldn't be 198A because then the voltage the motor sees would not be 65V it would be 16.25V, i think the 0.142 ohms would be the resistance between any two phase wires because it doesn't make sense to provide the resistance from a phase to the center of the Y because you never apply power to the center, so you would always end up having to multiply that by two, so considering that 0.142 is the phase-phase resistance you get a max current of about 440A. i can use maxima to solve for the actual current through the motor for the condition the motor was in when it blew. ill edit when i figure it out.

edit: ok i think i figured it out, i don't know what this kind of math is called, but i think it works. heres what i came up with.

A = motor current
D = duty cycle
75.0 = battery side current limit
0.146 = phase-phase resistance
65.0 = battery voltage
1.1 = estimated backemf at 8.6 rpm(from crystalyte website 5305 48V 365rpm no load speed)

A = (65.0 * D - 1.1) / 0.146), (D = 75.0 / A)

to solve it with maxima i used: float(solve([(A = (65.0 * D - 1.1) / 0.146), (D = 75.0 / A)], [A, D]))

and i get A = 179A, and D = 0.42(42%)

if this is right, why would 2 4110s in parallel blow with 180A going through them?

 

[eP]

Taking the maximun motor phase current to be 198A:

Assume that the motor is stalled.
[cut]
2) If the throttle is less than 25%, the average battery will be less than 75 A. So the current limiting will not activate at all.

3) The current limit will activate if the throttle is greater than 25%. Each motor phase winding would still experinces a maximun current of 198A.

The heating power of the motor for condition (1) and (3) is 62x75 = 4650 W.

Hi The7
Are You sure 198A thru 284 mohm is still within 4650W power budget ??
Would You like to calculate above ?

Would you like to explain how current limit works at stall condition ?

Best regards

 

[The7]

Hi The7
Are You sure 198A thru 284 mohm is still within 4650W power budget ??
Would You like to calculate above ?

At stall (lock-rotor condition)
4650 W should be the average heating power to the motor plus the resistance loss in FETs and plus resistance loss in the connecting cables)

The average power heating power to the motor = 4650 W x 0.284 /0.313 = 4234 W.

During the 38% conduction, the power into the motor = 198 x198 x 0.284 = 11134 W
Averaging the power over 1 cycle = 11134 x 0.38 = 4230 W.

The two results are identical.

Would you like to explain how current limit works at stall condition ?

The following simple analog way could be used.
A shunt is used to sense the average battery current. If the current is over the limit, it will over-ride the applied throttle voltage and reduce the throttle voltage until the battery current is within the limit. When the throttle voltage is reduced, the duty-cycle of the PWM is also reduced.

There could be some other more complex schemes/ways, either analog or digitai, to control the duty cycle directly. .

 

[eP]

The average power heating power to the motor = 4650 W x 0.284 /0.313 = 4234 W.

During the 38% conduction, the power into the motor = 198 x198 x 0.284 = 11134 W
Averaging the power over 1 cycle = 11134 x 0.38 = 4230 W.

The two results are identical.

Would you like to elaborate above ?
Particulary :

During the 38% conduction, the power into the motor = 198 x198 x 0.284 = 11134 W

I'm not sure what this mean.

Thus this mean 198A flows thru coils only 38% time of PWM cycle ? (duty cycle)
What about the rest 62% of time ?
What is the average current for that part of PWM cycle ?
Zero amps ?
So how fast current drop from 198A (for duty period) to zero (for the rest of the cycle) ?
Are the motor's windings (coils) acting the same way as resistors ?

One strange statment = so many questions.
Elaborate it with all details please.

 

[The7]

Are the motor's windings (coils) acting the same way as resistors ?

WE both understand that there is inductance as well as resistance in the motor.

For simplicity, the example is done on the assumption that the inductance efffect is negligible when compared with its resistance.
The inductance effect may be negligible because the motor frequency tends to zero at stall.

You could still use the "exact" method by including its inductance; PWM frequency; Duty cycle at the battery current limit; Exponential rise of motor current during ON through FETs; Exponential fall during OFF through body diodes of FETs .

In the worst case, the maximun motor current is limited by the resistance (NOT by the inductance).
The current limit mechanism could still act in a similar way whether the motor has inductance or not!

 

[eP]

For simplicity, the example is done on the assumption that the inductance efffect is negligible when compared with its resistance.
The inductance effect may be negligible because the motor frequency tends to zero at stall.

So, my next question: is PWM frequency=motor frequency ???
Is the winding inductance depends on rpm or back EMF ????
Would you like elaborate answer those questions and elaborate yours assumptions ???

You could still use the "exact" method by including its inductance; PWM frequency; Duty cycle at the battery current limit; Exponential rise of motor current during ON through FETs; Exponential fall during OFF through body diodes of FETs .

I don't want to use "exact" method , because i want to use only right method or methods.

In the worst case, the maximun motor current is limited by the resistance (NOT by the inductance).
The current limit mechanism could still act in a similar way whether the motor has inductance or not!

If your motor has No inductance, so tell us good man what should be PWM frequency and what is the reason for particular duty ?

If your motor has No inductance, so tell us good man what would be ripple torq if motor has NO inductance ?
Is your motor act this way ?

 

[Deepkimchi]

As I said before,as many parts as Mouser.com and other places have, one would think that if you could identify the type of current spike that is frying the controller, you could buy a transient surge protector...

Maybe look in the lightning department..... :wink:

 

[dirty_d]

ok as i showed in my work i figured that with a 75A battery current limit when doc was going 1km/h with the wheel rotating at 8.6 rpm the motor would be experiencing an average current of approximately 179A at 42% PWM duty cycle, with two 4110s in parallel conducting that current for 0.14 seconds at a time. this means each fet in the controller was passing 90A 33% of the time since only two fet pairs are conducting at a time for 6 step commutation, that brings the average current per single fet to 30A. since were using average current not RMS current the heating value will be higher, so just say 50A. thats still well under the 4110s max current, and they are screwed to a big aluminum case. so why did they blow? did something else go wrong?

 

[motor_builder]

New IRF 4110's are package limited to 120A, older ones 75A.

 

[The7]

So, my next question: is PWM frequency=motor frequency ???

My answer is same as your answer.


Wonder if I could ask you a question instead?

An PWM circuit is used to control the power into a simple RL series circuit from an DC source .
We all know how to determine the waveform of the current for any % duty cycle of PWM.

What is the waveform of the current at 100% duty cycle?
Does this current waveform depend on L at steady state?

 

[dirty_d]

doc, where did you go??? did you measure that resistance of the motor yourself? if so was it phase-phase? from the crystalyte websites data i figured about .5 ohms.

 

[The7]

The current limit mechanism could still act in a similar way whether the motor has inductance or not!

Sorry for my poor English by giving you a wrong impression that an motor could have NO inductance.

I should rewrite as:
"The current limit mechanism could still act in a similar way whether the effect of the inductance of motor is negligible or not."

 

[eP]

So, my next question: is PWM frequency=motor frequency ???

My answer is same as your answer.


Wonder if I could ask you a question instead?

An PWM circuit is used to control the power into a simple RL series circuit from an DC source .
We all know how to determine the waveform of the current for any % duty cycle of PWM.

What is the waveform of the current at 100% duty cycle?

Do we talking about stall condition (motor frequency = 0 Hz ) or not ?
Is this 100% duty cycle at stall condition or not ?
Is yours 198A motor current at 100% duty cycle or at 38% duty ??

If that current is at duty less than 100% at stal point (rpm=0) than you should keep in mind that current is valid for WHOLE PWM period, not only duty period. (FOR the sake of HIGH PWM frequency and motor's inductance ).

If you are not able to understand above, so ask someone to explain you some basic principles (maybe Fechter would like to do that for you). And stop creating a mess !!

Does this current waveform depend on L at steady state?

Yes it does !!!
For the sake of PWM, current waveform depend on L at each state for whole rpm range (from 0 to max rpm ).

The PWM frequency is always so high that the delta I (difference between current's hills/peaks and valleys) for one PWM period is negligible small to average current for the same period of time.

Sorry for my poor English by giving you a wrong impression that an motor could have NO inductance.

Frankly speaking my English is poorer than yours (is really poor in fact as i was always to lazy to polish my English ), but real source of your's problems is your lack of basic principles how controller work, what is the PWM frequency etc.

So once again: for stall point case the motor current never goes up to 198A if the average curent limit is set up for 75A and the resistance is 280 mohm at 38% duty for the sake of high PWM frequency and high motor inductance.

If you are not able to understand so basic principles you should stay far away from such discussions because your childish assumptions creating mess and desinfromation.
You acting the same way as troll's by makings so stupid assumptions that PWM frequency = 0Hz at stall point , or motor inductance is negligible for stall point.
Sorry but it is true. It seems to me you trying mimic some safe's lessons at his absurd's academy.

Best regards

 

[The7]

but real source of your's problems is your lack of basic principles how controller work, what is the PWM frequency etc.

Yes, eProf. Back to school!

Is it valid to make an assumption that the PWM freq = O Hz when the duty cycle of the PWM is 100%?

 

[eP]

but real source of your's problems is your lack of basic principles how controller work, what is the PWM frequency etc.

Yes, eProf. Back to school!

Is it valid to make an assumption that the PWM freq = O Hz when the duty cycle of the PWM is 100%?

Such assumption is correct only if current limit mechanism is not based on PWM and rpm=0. (the same one phase active all the time)
In other case if PWM based limiter is active, which forcing duty interval to 38% of PWM period, the PWM cycle time must be a finite time interval as a result PWM freq must be > 0 especially at stall case (rpm=0).

So for Doc's case (limit=75A based on PWM) your assumption is wrong !!!


Winding (motor's coil ) see always higher one (frequency = f_PWM or f_motor) and such freq must be positive (greater than zero) if limit mechanism is active. So if f_motor= 0 Hz for rpm=0 than f_PWM must be > 0 Hz if duty is less than 100%. (38% for example).

I have to add f_PWM must be high enough to keep motor current ripple as low as possible to avoid motor's core saturation.


Is that clear Mr. Pupil ?

 

[Doctorbass]

doc, where did you go??? did you measure that resistance of the motor yourself? if so was it phase-phase? from the crystalyte websites data i figured about .5 ohms.

no.. I measured that myself with precise method using 4 wire current sensing and i got: 0,142 ohms (phase to phase)

see my thread about the electrical caracteristic of the 5305:
http://endless-sphere.com/forums/viewtopic.php?f=2&t=4404&p=65193#p65193

Doc

 

[dirty_d]

doc check out this data, http://crystalyte.com/x5specs.htm. it shows the X5305 at 48V with a no-load of 365rpm, thats 0.132 V/rpm, so at 287 rpm the motor would generate 287 * 0.132 = 37.8V of backemf, in the data it shows the motor draws 20A at 287rpm at 48V, ( 48.0 - 37.8 ) / 20.0 = 0.510 ohms. whats the deal here, their data shows .5 ohms but you measured 0.142 ohms.

 

[eP]

doc check out this data, http://crystalyte.com/x5specs.htm. it shows the X5305 at 48V with a no-load of 365rpm, thats 0.132 V/rpm, so at 287 rpm the motor would generate 287 * 0.132 = 37.8V of backemf, in the data it shows the motor draws 20A at 287rpm at 48V, ( 48.0 - 37.8 ) / 20.0 = 0.510 ohms. whats the deal here, their data shows .5 ohms but you measured 0.142 ohms.

That remember me my discuss with Lowell over one year ago.

http://endless-sphere.com/forums/viewtopic.php?f=3&t=944&p=14795&hilit=guarantee#p14795

 

[Doctorbass]

doc check out this data, http://crystalyte.com/x5specs.htm. it shows the X5305 at 48V with a no-load of 365rpm, thats 0.132 V/rpm, so at 287 rpm the motor would generate 287 * 0.132 = 37.8V of backemf, in the data it shows the motor draws 20A at 287rpm at 48V, ( 48.0 - 37.8 ) / 20.0 = 0.510 ohms. whats the deal here, their data shows .5 ohms but you measured 0.142 ohms.

Yess, but i measured 0.142 ohm directly on the winding end where winding to wire solder are.. not on the anderson connectors at the end the yellow, blue green wires....

I suspect that they measured that on the anderson connector.. not me...i prefer to measure data that can be a reference so if people decide to ameliorate the Y-B-G wire to a larger size, the motor winding data will stay the same.. and not take account of the cable...

I measured that these 3 wire a 12AWG with thiny teflon insulation. they are 0.001588 ohm per ft and each phase wire have around 1.5ft from the motor windint solder to the anderson connector.. so that's 3ft total phase to phase that give 0.0048ohm for the wire loss

Adding this to the motor windig resistance give: 0.0048 + 0.142 = 0.1468
ohm phase to phase at ambient temp.

I'm sure about my results. no doubt.

Remember that i used 4 wire sensing method wich is alot more precise!.. maybe Crystalyte used a normal multimeter with precision of 0.1ohm and that they had not done the zeroing beforte measurement... who know...

 

[eP]

I'm sure about my results. no doubt.

Remember that i used 4 wire sensing method wich is alot more precise!.. maybe Crystalyte used a normal multimeter with precision of 0.1ohm and that they had not done the zeroing beforte measurement... who know...

It is not a matter of precision. It is matter of responsibility.

The next question: is the hot winding at work frequency have similiar resistance as cold one for DC ?
For RC motors they are differ at 2:1 to 3:1 ratios mostly.

Best regards

 

[dirty_d]

the heat would increase the resistance but not by much, the commutation would be at about 460Hz at 287 rpm. according to a skin depth calculator i found for copper at 460Hz the skin depth is 3.54mm, thats thicker than the wire so it wouldn't really have any effect right?

back to the mosfets, why did they fail?, ive calculated the average current through each of the 12 mosfets to have been 30A, with the motor current at 179A at 42% PWM duty cycle, according to docs measured phase-phase resistance of 0.146 ohms. is not a steady 30A just average, its 0A 66% of the time and its about 90A(with pwm ripple) 33% of the time. thats about 52A rms, still under the 4110s specs. and thats worst case rms it was calculated for a square wave (0A, 0A, 90A).

 

[The7]

So for Doc's case (limit=75A based on PWM) your assumption is wrong !!!

I have to add f_PWM must be high enough to keep motor current ripple as low as possible to avoid motor's core saturation.

Hi eProf.

Thanks for pointing out these two vital points.

I would like to seek your opinion on the following two cases.

Case 1
If the rider tries to FULL throttle at stall, the response time of the current limit may not be fast enough to reduce the 100% duty cycle to a lower value.
During that short period of 100% duty cycle, the motor current could have reached a very high value (very close to the max current as if the motor has NO inductance) because the frequency of the PWM will have no effect in 100% duty cycle.

Case 2
The frequecy of the PWM will always in effect if the rider is partial throttle at stall. The "steady state" PWM duty cycle of this case is same as that of case (1).

The steady state PWM duty cycle is not 38% any more. However, it could be computed with all the known values of R, L and f_PWM by iteration. I would guess it to be 40% to 60%.

Your stupid ePupil.

 

[dirty_d]

The7, its likely that the current wont be able to reach the max, depending on how many pwm cycles it takes the motor current to go from 0A to maxA, if it reaches maxA in 1 cycle the MCU will lower the PWM cycle the next cycle. it likely works like this:

1. read throttle voltage
2. read battery current
3. if battery_current < 75A, PWM_duty = throttle_voltage / 5.0V; else if battery_current >= 75A, PWM_duty = PWM_duty * (75A / battery_current)
4. repeat

its probably much different than that but you get the idea, if the motor current ever reaches the max it will only take 1 pwm cycle for the MCU to lower the duty cycle to bring it down.