Instant Start 18 fet Infineon Boards are here...

[methods]

p.s. Methods... I just soldered in 3 pieces of your braid over the shunt with a crap load of solder .. (the most i've ever put on a shunt) ... I pulled 132amp @ 79.2v or 10454watts or 14hp! .. I'm about 7hp away from doc .. I'm on a 24" 5304 though

thanks
-steveo

Was that Peak current or continuous?
What current did it settle out to after a few seconds of hard acceleration?

The shunts are kind of funny... Some times I hardly have to coat them with solder to get 250uOhms and sometimes I have to drown them in it.

-methods

 

[steveo]

p.s. Methods... I just soldered in 3 pieces of your braid over the shunt with a crap load of solder .. (the most i've ever put on a shunt) ... I pulled 132amp @ 79.2v or 10454watts or 14hp! .. I'm about 7hp away from doc .. I'm on a 24" 5304 though

thanks
-steveo

Was that Peak current or continuous?
What current did it settle out to after a few seconds of hard acceleration?

The shunts are kind of funny... Some times I hardly have to coat them with solder to get 250uOhms and sometimes I have to drown them in it.

-methods

Hey methods

I will give you that info once i get my c/a back from ebikes!!...

I have a watts up metter wired between my legs.. I don't wana look down and give her full throttle... lol

I got the ebs working guys .. thanks for all your info....

I basically sorted the bk & ground (near the cr pin) .. Then i put a momentary switch between ebs- & ground ! and it worked !!

-steveo

 

[methods]

Ok, I will eat my words.

Even if precautions are taken it is still retardedly easy to short the +5V line to the +100V line.
Please do as Steveo did and cut back the +5V via and the 1" stretch of high voltage trace that fit in the Groove.
Depending on how you soldered your heat sink it is possible that it will rub on the bottom, top, or both.

Even though I conformal coated the bottom, top, and the groove of the controller I still got bit.
Blew my uC.

SAM-Pilot was nice enough to help me troubleshoot.
He pointed out that the components most likely to blow due to overvoltage for a short period on the +5V buss are:

The Polarized 25V Cap C4 (mine was fine)
The 7805 regulator (mine blew - easy part to find though)
and the uC (errr.... :? )

I think this will be a common enough occurrence that I can justify buying 50 of the micro controllers to send to people when they make mistakes.
For me it is no big deal since I can pirate parts off of other boards but for you guys it would be a show stopper.

Best part is - usually does not happen until you are completely done

-methods

 

[methods]

Oh yea, I forgot:

Here are some things you can look for:

Check the voltage between ground and the top of the 3K base resistor.
Should be 12 - 14ish volts. If it is not, check the 7805 regulator

The resistance between the +5V buss and ground should be something in the Kilo ohms. Perhaps 4.2K.
My chip blew and I was seeing 120 ohms to ground.

You dont have to remove the chip to see if it is dead. All you have to do is cut the single ground trace going to the chip.
The trace is easy to find. Just map it from the programming header. It is a slightly wider trace.

-methods

 

[Doctorbass]

I just ordered two uC from Kenny from crystalyte to replace the one of the blown infineon but they dont have any that are programmed.. he said they program them once they are soldered on teh board... in other words they need the board to be programmed he said to me.

so i'll send him my board... (the infineon 18 fets) they will install it and program it.

so i'll now have 3 fully working modified 18 fets controller 150A 100V... having two spare unit is a must for peole like us :wink:

Doc

 

[methods]

Shipping to and from China for a "built board" will be more expensive than just ordering a new blank I think.
Let us know.

-methods

 

[methods]

Important Note - Discharge all capacitors before assembly into extruded case


Having assembled 20 of these 18 fet units I have had the chance to look at quite a few.
As it turns out, if you use the supplied heat sink, when securely fastened, there are a few thousandths clearance between the +V track and the case and the +5V track and the case.

BUT

During assembly, before the heatsink is pulled back against the side of the case there is a point in time where both the +V and +5V are likely to come into contact with the conductive case and short together.

IF

You assemble your unit with the 100V capacitors charged it is highly likely that you could discharge one of those caps into the +5V line and blow out any number of components including, but not limited to you uC, Comparitor, etc. This would be fatal.

EVEN IF

You cut back the traces, conformal coat them, and be very careful it is still possible to screw the pooch.


PLEASE

Take the time to properly discharge the 100V capacitors before assembly.
Test your board, discharge the caps, assemble into the case, test for isolation, then test your board again.
This is absolutely mandatory

IF YOU FOLLOW THESE INSTRUCTIONS YOU WILL NOT HAVE ISOLATION ISSUES.

I dont want to hear about one more single instance of a low voltage section being blown out.
Doc has done it and I have done it.
No one else has to. Please heed my warning and proceed with due diligence.



oh yea - while you are in there, go ahead and cut back the traces a little to be sure.
I cut back the +5V trace at the corner on the top and bottom
I also cut back the VCC trace next to the big caps on the top left corner
Make sure you put some glue over the cut traces so as not to leave exposed copper

-methods

 

[steveo]

Important Note - Discharge all capacitors before assembly into extruded case


Having assembled 20 of these 18 fet units I have had the chance to look at quite a few.
As it turns out, if you use the supplied heat sink, when securely fastened, there are a few thousandths clearance between the +V track and the case and the +5V track and the case.

BUT

During assembly, before the heatsink is pulled back against the side of the case there is a point in time where both the +V and +5V are likely to come into contact with the conductive case and short together.

IF

You assemble your unit with the 100V capacitors charged it is highly likely that you could discharge one of those caps into the +5V line and blow out any number of components including, but not limited to you uC, Comparitor, etc. This would be fatal.

EVEN IF

You cut back the traces, conformal coat them, and be very careful it is still possible to screw the pooch.


PLEASE

Take the time to properly discharge the 100V capacitors before assembly.
Test your board, discharge the caps, assemble into the case, test for isolation, then test your board again.
This is absolutely mandatory

IF YOU FOLLOW THESE INSTRUCTIONS YOU WILL NOT HAVE ISOLATION ISSUES.

I dont want to hear about one more single instance of a low voltage section being blown out.
Doc has done it and I have done it.
No one else has to. Please heed my warning and proceed with due diligence.



oh yea - while you are in there, go ahead and cut back the traces a little to be sure.
I cut back the +5V trace at the corner on the top and bottom
I also cut back the VCC trace next to the big caps on the top left corner
Make sure you put some glue over the cut traces so as not to leave exposed copper

-methods

This is a reason i added that led light in there on the 5v rail with i think a 300ohms resistor is what doc roughly recommended....Before i disassemble everything.. I leave the watt up metter plugged in (mine has its own battery for power)... i flip on the controller .. watch the led burn out the excess power left in the caps .. .. and when the voltage is under 10v .. I short the neg and positive leads togther to get everything out of them!

then i disassemble!

-steveo

 

[geoff57]

hi
both me and knuckles have had the problem of caps staying charged with the 12 fet controllers Not to such a bad affect though.
I am using a stand alone CA on one of my bikes and when i pull the plug between the batt and the shunt it takes a good 5 seconds for the CA to shut down thoes caps must have one hell of a capasity.

For thoese of you putting controllers together this is not so good but how about this, disconnect the battery with the CA still connected then switch on the "ignition"switch (from Vcc to Vcc-L) the CA should come on discharging the caps.
This is a theory, I have not tried it not even on a 12 FET

Geoff

 

[methods]

That's how I discharge mine.

-methods

 

[liveforphysics]

I discharged the board you sent me into my thumb. It seemed to work fine.


Methods- I've never had something shipped to me with such a professional packing job! The USPS mail guy could have played soccer with your package all the way from California to WA, and nothing inside would have been damaged. You missed your second calling in life as a pro-packer Maybe a "power-bottom". Very creative packing materials as well. It was great opening the box in front of my mom.


Best Wishes,
-Luke

 

[ZapPat]

[...] Very creative packing materials as well. It was great opening the box in front of my mom.

Are you using your old playdude mags as packaging materials, Methods?! Maybe I should order one of those 18 FET monsters from you after all... :lol:

 

[methods]

Actually... If you must know....

I padded the 2 controllers in lukes box with about 300 empty condom wrappers.

-methods

 

[ZapPat]

Actually... If you must know....
I padded the 2 controllers in lukes box with about 300 empty condom wrappers.
-methods

I guess that proves that you listen to your mom's advice after all... at least I think I remember it was you saying something like "my mom always told me to double-bag everything", right?

 

[methods]

You have got a good memory :wink:


-methods

 

[Doctorbass]

Ecrazyman 12 faster than Infineon 18...

any idea?

http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=11366

Doc

 

[methods]

You need to program your controller.

-methods

 

[methods]

I built up a couple of bikes for the wife and I

Mine:
Beach Cruiser
9x7 rear 9C
18 fet controller
24S 2P 10Ah 88.8V Lipo pack
No current limit

Hers:
Walmart Special
10x6 front 9C
18 fet controller
12S 4P 18Ah 44.4V Lipo pack
No current limit

We took a long ride up to the top of a hill near our home in Livermore.


Here is a google map zoom.
Dead center in this zoomed in picture is where the bikes are parked in the pictures:
http://maps.google.com/maps?f=q&sou...015,-121.743523&spn=0.00126,0.003862&t=h&z=19

The wifes FWD actually did better than mine... Since she was doing a continuous burnout all the way up the gravel trail she was running at a more efficient speed. My RWD took the full brunt of my 18fet and punched through 140A bursts. My motor was definitely overheated - I could smell the insulation.

$150 hub motor
$145 100V 100A controller kit
$650 888W battery
$140 CA
$15 throttle

Riding your bike to the top of the world without breaking a sweat?

$priceless

-methods

 

[deecanio]

beautiful shots Methods, thanks for posting them

D

 

[Doctorbass]

Methods.. did you had this 140A peak in these little 45A anderson connectors?

Doc

 

[steveo]

Methods.. did you had this 140A peak in these little 45A anderson connectors?

Doc

I've done 132 amps with my 45amp andersons lol

I want to change the connectors ... but i only use 10 awg from controller to battery.. lol

-steveo

 

[methods]

Methods.. did you had this 140A peak in these little 45A anderson connectors?

Doc

Yea, why not?

The rating of a connector is simply a factor of its inline resistance and the amount of heat it will generate per amp.
They are rated to increase by a certain number of degrees at a certain ambient temperature.
You could run 400A through the connectors so long as you did not melt them together :wink:

The connectors were not the weak link....
The weak link was the 6" of 16AWG wire that was still on phase wires motor.
At 88V those motors are only rated for 10A too btw

I dont remember the inline resistance of the 45A Anderson connectors but it is measured in micro ohms.
The inline resistance of the fets is mili ohms.

So... I dont worry about such things.

-methods

 

[methods]

Here is the datasheet on the 45A anderson connectors

~ 500uOhms inline resistance

http://www.andersonpower.com/files.php?file=DS-PP45(6).pdf


-methods

EDIT: They are rated for a 220 degree F temperature rise.

 

[Mike1]

On the subject of shunt calibration. I plan to use Justin's reverse current method using a 12V SLA with a 200W 1R1 resistor in series to give a little over 10 amps. Do any of you gurus see any potential problems with this?

 

[methods]

Just remember to get the polarity correct.
I remember it being counter intuitive.
You have to drive the positive current into the controller ground and out of any one of the phase wires correct?
Do you have a link that you could post by chance? I would like to see if it is the same post that I am thinking of where Justin suggests using a large C battery.

I still have not tried his method as I really like using my method with the CA running.
I guess there is no reason that I could not use his method and my method combined.

One problem you will find using a 12V SLA is that as the battery discharges the current will slowly drop.
If you work fast you wont notice. I often times program 3 controllers at once (since I am set up) so I like to use a CV supply to keep things consistent.
Should work great though. Please let us know about your methods and results.

SUBJECT CHANGE

On the subject of the shunt - I will be adding new lines to the controllers for charging.
On my old bike I used to charge through the shunt so that I could measure negative AH.
It is a nice feature... Just turn on the CA while charging and you can monitor Charge Voltage, Charge Current, etc.

I will be running a pair of charge wires out of all future controllers.
These will be tied to V+ and PHASE(-)
This will force the current to run backwards through the shunt and roll back AH while charging

You guys will love it.
I will size the wires to about 20A charging.

-methods