IRFP4468PbF - Adaptto Max E

[steveo]

Hey Everyone,

I was wondering if anyone knows how many -- phase & battery amps can be produced in this case out of a IRFP4468 TO-247 Package Mosfet with the Adaptto Max E ?

In essence, i'm trying to find out what is the mosfet itself safe limit to determine exactly how far i can push the power of the Max E..

The only thing i've ever run to compare and have a idea with the power limits .. was a infinion 12 fet modified with irfp4568 mosfets that ran with a 24s lipo pack.

from what i tested on the same motor x6 hub motor .. i was able to peak around 200amps @ 88.8v approx 260 per phase or 780 total phase amps... with just a 12 fet square wave pcb and irfp4568 fets..!!!

so.. assuming i have the ultimate thick copper traces & wires & connections & cooling .. should i not be able to hit 300amp dc current on an adaptto ? and push 1000phase amps say with 12s or maybe 18s lipo?

please correct me if i'm wrong .. if anyone can explain what exactly to look at to determine this on the spec sheets!

http://www.irf.com/product-info/datasheets/data/irfp4468pbf.pdf

http://www.irf.com/product-info/datasheets/data/irfp4568pbf.pdf



-steveo

 

[fechter]

The 'package limited' rating would apply to a steady continuous current. More than this and the legs melt. In a 3 phase controller, since each FET bank is only on 1/3 of the time at most, you can go higher than this, probably more like the 'silicon limited' rating.

 

[macribs]

Pardon my ignorance but are you planning to upgrade Max E? Will this be sort of the steveo of Adaptto like Lyen did for those chinese controllers?
What I mean you are after pushing more current and more voltage for Adaptto Max E? What if any mods/upgrades are you planning? Would you need to solder copper bars or other mods to make a more powerful version of Adaptto, and how will you tackle the firmware? Adaptto uses closed source firmware, right? Guessing there could be max limits sat in the firmware?

 

[steveo]

The 'package limited' rating would apply to a steady continuous current. More than this and the legs melt. In a 3 phase controller, since each FET bank is only on 1/3 of the time at most, you can go higher than this, probably more like the 'silicon limited' rating.

so 290a silicon limited is for the 4468 fets.. how do i translate that into knowing how many phase and battery amps i can pull

the max e uses 18 of the 4468 fets

thanks
-steveo

 

[fechter]

The 'package limited' rating would apply to a steady continuous current. More than this and the legs melt. In a 3 phase controller, since each FET bank is only on 1/3 of the time at most, you can go higher than this, probably more like the 'silicon limited' rating.

so 290a silicon limited is for the 4468 fets.. how do i translate that into knowing how many phase and battery amps i can pull

the max e uses 18 of the 4468 fets

thanks
-steveo

If there are 18 FETs then each leg will be 3 parallel FETs. These have to handle the phase current, so somewhere around 870A phase current. Depending on the PWM duty cycle, this would translate to about 300A on the battery. This is probably an oversimplification, but will be in the ballpark.

What you should design for and what it takes to make them blow up are two different things. I've pushed lots of transistors to the smoke-release point.

 

[Doctorbass]

-Measured on the adaptto by their engineer:
10V overshoot ABOVE the instantanous batt voltage per 100A phase current


-As fet temp goes high the max 100V rating goes up a little bit , and same opposite occur on cold temp

For the max rating 350A phase, this mean 35V overshoot above the instantanous battery voltage! so if the fet are hot and max Vds raise to 110V the maxinstantanous battery voltage should not exceed 75V (75 + 35 = 110V)

I guess it might vary depending on the motor inductance too.

So i guess that the max phase current depend on the batt voltage you use too.. the lower Vbatt the higher the phase amp is .. but witch ration give best perfomance..... i guess that 20-21s is probably the best to use...

the copper busbar on the shunt side are still thin for 140A rated controller... so my guess is that if it is not adressed these might melt the solder and move on the pcb and modify the shunt value witch could fall over the programmed range, making it to drive fets beyond his controlled limit.


Here is what i wrote in my 32kW ebike thread:

Adaptto said they measured that these fets at 115V in breakdown. They recommand optimum voltage of operation at 85V( about 20-21s for 4.15V per cell) ( so that's about 30V margin before breakdown. Now that's not counting the sag voltage that contribute i guess to increase that margin to maybe 35 or 40V higher than the Vbatt state. Jeka, their engineer said they measured that these fets get 10V spike per 100A phase. that seem to match with the 30-40V witch is 350-400A phase max spec.

When and the only time i blew the adaptto was with the 18kv motor at too high voltage and too high phase current. The 5302 is a true controller killer !!. I killed about 5 Lyen 18 fets controller with that motor in the past. But last automn, when i blew the adaptto it was not due to high temp, in fact the adaptto remain very cold compare to the Lyen 4110. The adaptto 18 fets produce 4 time less heat than the Lyen 18 fets. The 4110 are 3.7 miliohm rds ON while the 4468 of the adaptto are half of that, plus the 4110 to-220 case have less surface in contact with the heatsink than the 4468 have. The Adaptto is monitoring the mosfet temp and it can limit the current if temp goes too high.

Doc

 

[steveo]

-Measured on the adaptto by their engineer:
10V overshoot ABOVE the instantanous batt voltage per 100A phase current


-As fet temp goes high the max 100V rating goes up a little bit , and same opposite occur on cold temp

For the max rating 350A phase, this mean 35V overshoot above the instantanous battery voltage! so if the fet are hot and max Vds raise to 110V the maxinstantanous battery voltage should not exceed 75V (75 + 35 = 110V)

I guess it might vary depending on the motor inductance too.

So i guess that the max phase current depend on the batt voltage you use too.. the lower Vbatt the higher the phase amp is .. but witch ration give best perfomance..... i guess that 20-21s is probably the best to use...

the copper busbar on the shunt side are still thin for 140A rated controller... so my guess is that if it is not adressed these might melt the solder and move on the pcb and modify the shunt value witch could fall over the programmed range, making it to drive fets beyond his controlled limit.


Here is what i wrote in my 32kW ebike thread:

Adaptto said they measured that these fets at 115V in breakdown. They recommand optimum voltage of operation at 85V( about 20-21s for 4.15V per cell) ( so that's about 30V margin before breakdown. Now that's not counting the sag voltage that contribute i guess to increase that margin to maybe 35 or 40V higher than the Vbatt state. Jeka, their engineer said they measured that these fets get 10V spike per 100A phase. that seem to match with the 30-40V witch is 350-400A phase max spec.

When and the only time i blew the adaptto was with the 18kv motor at too high voltage and too high phase current. The 5302 is a true controller killer !!. I killed about 5 Lyen 18 fets controller with that motor in the past. But last automn, when i blew the adaptto it was not due to high temp, in fact the adaptto remain very cold compare to the Lyen 4110. The adaptto 18 fets produce 4 time less heat than the Lyen 18 fets. The 4110 are 3.7 miliohm rds ON while the 4468 of the adaptto are half of that, plus the 4110 to-220 case have less surface in contact with the heatsink than the 4468 have. The Adaptto is monitoring the mosfet temp and it can limit the current if temp goes too high.

Doc

Hi Doc,

Most likely i will go with a 46.2v or 69.3v nominal voltage lipo pack with minimal voltage sag..

10v overshoot for every 100 amps phase .. so with my 46v.2 battery pack say.. i should be able to get up to at least 500 amps per phase wire on the motor or a total of 1500 phase amps am i calculating that correct?

The highest phase amps i've seen on a stock max e controller with a 57.75 nominal lipo pack was 275 amp per phase or 825 amps total phase amps... ( i think this was around 170 dc battery amps )

thanks
-steveo

 

[steveo]

I did some quick calculations comparing the 4468 mosfets to my experience with the famous irfb4110 fets...

a to-247 package 4468 mosfet is capable of 195A max @ 25 degrees - wire bound ---

a to-220 package 4110 mosfet is capable of 120A max @ 25 degrees - wire bound ---


i know on a irfb4110 24 fet infinion controller -- the max dc amps i've pulled was - 200a - safely
this is 42% of its wire bound limit .. or 50 amps a fet

comparing this to the 4468 .. 42% of its wire bound limit would be 81 amps minimum should not be a problem.. right now stock from adaptto is about 66amps..

so i believe 250amps dc should not be a sweat.. 300amps is defiantly possible!

-steveo

 

[steveo]

update

so with 18s4p 20c trunigy pack i was able to turn out 197 battery amps and 330phase amps per motor phase wire on a 90% charged battery.. so we are already at 990 total phase amps

i will unlock its firmware tonight and tweak the shunt and see what i can do to get towards 400 phase amps..


-steveo

 

[steveo]

update #2

ok... so ive unlocked the firmware and ive determined that you can not reduce the resistance of the shunt past .306mohms.. weither it controller is locked or unlocked..

however the battery amp limits do increase in the profile settings to a max in my case 246 dc battery amps.. and phase amps seem pertty much unlimited im able to go 500a +

it seem that even though the software is unlocked .. its actually still locked based on the battery amps

just as i tested today before unlocking firmware i was able to set battery amps around 197... and even with phase amps set at 400 or so.. i only was able to get 330phase amps or 990 total phase amps to the motor..

i will retest the unlocked controller tommorw.. i have no doubt i should be able to hit 400amps+ per phase or 1200+ total phase amps ... to bad im capped at 246 dc battery amps.. gahh

is there another version of the unlocked firmware that does not lock the dc battery amps???

with 18s4p pack based on the limiter of dc battery amps.. i should be able to pull 17.7kw battery... but i estimate 412 phase amps

-steveo

 

[fechter]

Actually, the phase amps don't add since only one phase is active at a time. They take turns. 300 phase amps is 300 amps.
Phase amps will be higher than the battery amps at partial duty cycle and equal when at full throttle and not in current limiting.

 

[Brake]

Steveo, thanks for this update. Finally! It's great to hear about pushing the limits of this thing. Right now I'm running around 12+ kw. 21s12p tesla pack. These are my settings right now in unlocked mode.

I did upgrade the caps and controller wires, not sure how much that gets me. Maybe 14kw reliably. But I just don't know how far too plush the pack? I think my voltage sag under max setting right now is down to 74v. So roughly 160 battery amps. I think I'll try upping the battery amps and keep my eye on what you are doing.

Thanks man

 

[Punx0r]

If you want to believe datasheet ratings for real-word applications then for your FETs it states a maximum repetitive pulsed current of 1120A per device, with the pulse width limited by junction temperature.

 

[steveo]

I did some quick calculations comparing the 4468 mosfets to my experience with the famous irfb4110 fets...

a to-247 package 4468 mosfet is capable of 195A max @ 25 degrees - wire bound ---

a to-220 package 4110 mosfet is capable of 120A max @ 25 degrees - wire bound ---


i know on a irfb4110 24 fet infinion controller -- the max dc amps i've pulled was - 200a - safely
this is 42% of its wire bound limit .. or 50 amps a fet

comparing this to the 4468 .. 42% of its wire bound limit would be 81 amps minimum should not be a problem.. right now stock from adaptto is about 66amps..

so i believe 250amps dc should not be a sweat.. 300amps is defiantly possible!

-steveo

Its weird that what i predicted here based on the --wire bond-- limits.. is almost exactly what Adaptto has limited in the firmware -- i estimated 243 battery amps... where as software from Adaptto is limiting 246 battery amps LOL....

 

[steveo]

update

so the weather here finally cleared up..

and the results are in..

with a pack at about 80% charged... 18s4p turnigys .. 20c lipos ... and my x6 mhm602 hub

I was able to peak.. 300amp Dc battery amps for a split burst.. under a really nice burnout then the tire gripping..

however under normal acceleration from a dead stop.. i was avg about 260 dc amps...

the highest phase current i recorded on the motor side was around 352 Phase amps

the adaptto display said i peaked at one point at 19.5kw..

lets say lipos are sagging.. approx 66.6v or 3.7 a cell..... 66.6v x 300amps is.. 19,980w..

I did bump of the pwr timing slightly.. 2 clicks to advance it but not much noticeable difference in power

a couple things i will play with and adjust to see how much i can improve its performance even more...

as per the manual:

LS current - Limits the starting current.

LS start - The speed at which the current is allowed to start increasing.

HS start - Speed above which maximum current is available (that current
specified in the profile). That is, the current gradually increases, from the
speed LS start and maximizing at speed HS start.

PWR timing - Sets the lead angle shift in relation to the phase current. The
greater the inductance of the windings of the motor, the stronger the control
signal phase currents. This setting compensates BEMF

i will use this when i reconfigure my pack to 8p12s lipo 40ah pack

OVS timing - Setting responsible for extra motor speed when the battery
voltage is not high enough to achieve the maximum vehicle speed. Works
like this: When the PWM cycle is nearly 100%, the controller increases the
lead angle, which enables additional speed (and reduced efficiency of the
motor). Only if the voltage is limiting the maximum motor speed, not the
wind resistance, will this work.

this should be already maxed

PWM rise lim - regulates the maximum speed of the motor (or the maximum
rate of increase in cycle of the PWM) and to some extent the rate of
acceleration when cracking the throttle. The larger the number, the faster.
The highest setting is 80, for extreme driving and maximum performance

probably gonna tweak regen as i notice its to strong when the trike is close to stopping

Torque rise – Optional setting for reverse motor braking. Allows you to
configure a smooth response at low speeds. The higher the value, the stronger
the braking at low speed. 80-150 is recommended.


there is prob more stuff.. but that's enough for now...

Does anyone know which firmware version will allow past 246 battery amps in software?..

its interesting how mine is limited like that in software.. however i managed to shoot about 50amps past that on a nasty burnout

 

[Brake]

That's awesome. 19.5 kw! Wow. I like the setting expansion too. You've tempted me to start tweaking again.

 

[Brake]

Just to make sure, is your traction control off? Not doing the traction control trick, correct?

 

[steveo]

Just to make sure, is your traction control off? Not doing the traction control trick, correct?

i don't think it is.. solid high load burnouts so far..

-steveo

 

[steveo]

pics ..

I have to say acceleration is reallyyy good.. when your in a pertty decent turn .. i can get one of the front wheels to come off the ground ... i'm not gonna say its insane compared to my dual Max E Adaptto E-bike with the hub monster.. That ebike.. i have to hold on for dear life at WOT.. with only 57.75 Nominal pack voltage -- 6p 15s or 20c lipos.... as my ass will slid off the dam seat LOL.. i need a back rest hahah..

Pics below showing some of the low esr caps i added.. i would of add more.. but just don't have anymore atm .. either way the resistance is about 2.5 mohms.. from originally like 5 mohms.. i've upgraded to 8 gauge silicon turnigy wires.. very very well soldered.. and i just added a tad more solder to that little shunt..

-steveo

 

[Doctorbass]

Wow that's serious numbers!

By how much did you reduced the shunt resistance? i'm asking because there is only a narow margin that the controller sofware can be calibrated too so you might fall in the reagion where controller shunt value is over that area and it just think you have let say 160A batt while i reality you have like 250A..


Doc

 

[steveo]

Wow that's serious numbers!

By how much did you reduced the shunt resistance? i'm asking because there is only a narow margin that the controller sofware can be calibrated too so you might fall in the reagion where controller shunt value is over that area and it just think you have let say 160A batt while i reality you have like 250A..


Doc

I tried to calibrate the shunt current best i can.. i used 2 20w bulbs... ran a constant load with a power supply at 12v rougly 4 amps.. and adjusted the reading on the adaptto with the reading on my amp clamp..

What i found though was the adaptto current reading jumps alot .. within half a amp.. it does not give a steady reading limit like my fluke amp clamp of my xantrex psu.. thus i can adjust it approximately.. but not exactly as what it should be.. i hope this makes sense..

i think on that 300dc current pull.. i had warmed up the shunt a bit.. after a couple nasty burnouts ..

pic of my trike in its current form below.. i'm loving it so far.. gotta sort out the independent front brakes.. gets a little scary under high speed braking

-steveo

 

[Merlin]

anything new? added latest firmware?
looks like you can setup higher battery current.

 

[macribs]

Got to say Steveo you explained the tweaks of Adaptto better in one post then what I could harvest at that endless adaptto thread.
19 kw is insane but I wonder if you are able to break the TWO ZERO limit with further tweaks?

How is the hubmonster working with the dual adaptto's?

 

[steveo]

i sold the adaptto on the trike to a friend of mine.

im sure you can go beyond 20kw power with a single unit.. id probably stop at 25kw output.. but im happy i got to 20kw.

the trike has gone back to a eb324 infinion 4110 matched fet.. tunned at 214 battery amps.. 267phase amps with its 24s4p 20c lipo pack.. i actually reassembled it with the controller today.. ive redone all the conntections to xt150 vs pp75.. and added a tiny bit more power then when i last ran a eb224 fet at 200batt 260phase..and between a slight power bump.. and less sag in the connections.. it hauls some serious ass with my whole 240lbs on it..

the dual adaptto hubmonster bike is doing very well.. ive upped its voltage from 57v nominal to 79v nominal and either its just me but it doesnt feel as responsive as it did at the lower voltage. im seeing about 10v sag at about 300amps..i only when a bit higher on the voltage to get a bit more capacity on the pack as i could not fit 8p.. so instead i did 7p24s a123 cells 2.3ah ..

pack is definatly bigger.. infact i had to make my bike side covers come out 1/4" on each side so i could fit the pack... and its also a bit heavier of couse around 35lbs.. vs the lipo pack at about 25lbs

ive peaked about 300 battery amps total.. about 14kw as per adaptto display.. i dont know... just doesnt seem to get there as quick as it did with the 57v 6p15s 20c lipos i had in it...

im gona try removing the acceleration limiter next and my power settings are maxed in boost mode.... and see if that will do it.... and if it doesnt.. ill consider 22s or 20s pack.. .. but i really dont want to to maintain the wh of the pack..

-steveo

 

[Merlin]

iam on your site Steveo with up to 25KW peak.
with latest firmware i got 17KW peaks with LOCKED firmware....like doc bass also get from his max-e