You need three .ocx files. Here they areView attachment ocxfiles.zip
Copy them to your windows\system32 folder.
Copy them to your windows\system32 folder.
Instant Start 18 fet Infineon Boards are here...
- Thread starter methods
- Start date
methods
1 GW
That controller looks beefy! 
Please dont forget that the pigtail colors do not match up with standard colors
Especially not the CA.
Also -> if you plan to run >72V you should use the 3W 2Kohm resistor I gave you to bypass the surface mount "base resistors"
Looks good, glad you are having fun.
That is what the kit is all about.
-methods
Since I am on my soap box: Once again -> Everybody please remember to discharge the main caps before assembly (lol - how many times have I said that now ?)
Please dont forget that the pigtail colors do not match up with standard colors
Especially not the CA.
Also -> if you plan to run >72V you should use the 3W 2Kohm resistor I gave you to bypass the surface mount "base resistors"
Looks good, glad you are having fun.
That is what the kit is all about.
-methods
Since I am on my soap box: Once again -> Everybody please remember to discharge the main caps before assembly (lol - how many times have I said that now ?)
methods
1 GW
hmmm... I am sure you probably know this, but just for people looking at the pictures:
where the programming cable plugs in you have to short the +5V pin to its neighbor (pin #1) to enable programming.
-methods
where the programming cable plugs in you have to short the +5V pin to its neighbor (pin #1) to enable programming.
-methods
John in CR
100 TW
A quick question about beefing up the traces. Is the anything inherently wrong with using some solid copper wire, and just tack soldering it near each junction where the trace splits off? I ask because thickening up the trace the whole way for a caveman with a soldering iron like myself is just begging for solder to go running to where it doesn't belong.
John
John
methods
1 GW
Nothing wrong with that at all.
The idea is to lower the resistance as much as possible and tacking a solid copper wire at 5 or 6 places will work fine.
That said - of course if I dont point this out someone else will
- many small strands of copper will conduct better than one large chunk (but I personally use a pair of 14AWG solid copper house wires)
- having the copper soldered along the entire track will help protect the board traces from overheating
- more is better in this case so two controllers being equal - yours will run hotter since it will have slightly more inline resistance (due to less conductive solder)
it is fine though - all of this is mostly just overkill and heat management since we have not properly engineered these controllers.
I was actually considering doing what you suggested but when I tried my OCD kicked in and I HAD to solder the rest
-methods
The idea is to lower the resistance as much as possible and tacking a solid copper wire at 5 or 6 places will work fine.
That said - of course if I dont point this out someone else will
- many small strands of copper will conduct better than one large chunk (but I personally use a pair of 14AWG solid copper house wires)
- having the copper soldered along the entire track will help protect the board traces from overheating
- more is better in this case so two controllers being equal - yours will run hotter since it will have slightly more inline resistance (due to less conductive solder)
it is fine though - all of this is mostly just overkill and heat management since we have not properly engineered these controllers.
I was actually considering doing what you suggested but when I tried my OCD kicked in and I HAD to solder the rest
-methods
liveforphysics
100 TW
John in CR said:
I will bring you my favorite type of soldering tools when I come to visit. Make a pile of everything and anything you want soldered or tweaked. When I finish, I will give you the tools I brought, and teach you a few tricks of the trade if you wish.
methods
1 GW
Sounds good - test it out and let us know.
Sure, shorting back at the USB connector will work.
I have good news
Phase 2 of the 18 fet saga has started.
In about 2 weeks I will be opening a new box
It weighs over 47kg (100 lbs)
-methods
Sure, shorting back at the USB connector will work.
I have good news
Phase 2 of the 18 fet saga has started.
In about 2 weeks I will be opening a new box
It weighs over 47kg (100 lbs)
-methods
methods
1 GW
As soon as the box arrives I will start to take pictures.
This will be a fresh start.
I will document step by step the modifications for those who have more time than money.
I will be selling the completed controllers for those that have more money than time.
Projected price will be $300 for 100V 100A Programmable 18 fet Crystalyte controller.
USB programming cable / software
CA Ready
Throttle / cruise Control
Ebrake / regen (maybe)
3 speed
reverse
Phat teflon cabling
This controller will be significantly better than a standard Crystalyte and significantly cheaper than a Kelly
This time I am having Kenny do the bulk of the work for me so I can focus on finish work.
Every board will be inspected and modified.
* 100V base mod
* 1.8k Power resistors
* 4W 250u Ohm precision shunt
* USB adapter added
* Regen enabled (maybe - we have to talk about this)
* Soldering checked / re-worked
* Board conformal coated
* Anderson connectors (maybe, but these may be 8AWG wires - so I may leave it up to the end user)
* Tested
Crystalyte is a trusted brand name. After I do my work they will be even more reliable
Those interested in pure reliability can run the boards at 72V 50A with total confidence
Those interested in pushing the envelope can run 24S lipo 100.8V with a 150A current limit
I will be charging a premium for the modified controllers but I will also be freely sharing every modification I make so it will be up to you guys - your time or your money
I took a loss on the 18 fet fiasco due to the slow start boards, my low pricing, and a $600 shipment I sent to a member that got lost in the mail.
I have $3k wrapped up in this new shipment so until I recover my investment - I will be a little tense :|
I will be offering a special discount for people who want to buy other stuff I have:
CA V2.11
26" 9x7 Rear 9C
ebrakes, throttle, 3-spd switch
torque arms, mosfets, etc.
I will start the new thread when the box arrives.
-methods
This will be a fresh start.
I will document step by step the modifications for those who have more time than money.
I will be selling the completed controllers for those that have more money than time.
Projected price will be $300 for 100V 100A Programmable 18 fet Crystalyte controller.
USB programming cable / software
CA Ready
Throttle / cruise Control
Ebrake / regen (maybe)
3 speed
reverse
Phat teflon cabling
This controller will be significantly better than a standard Crystalyte and significantly cheaper than a Kelly
This time I am having Kenny do the bulk of the work for me so I can focus on finish work.
Every board will be inspected and modified.
* 100V base mod
* 1.8k Power resistors
* 4W 250u Ohm precision shunt
* USB adapter added
* Regen enabled (maybe - we have to talk about this)
* Soldering checked / re-worked
* Board conformal coated
* Anderson connectors (maybe, but these may be 8AWG wires - so I may leave it up to the end user)
* Tested
Crystalyte is a trusted brand name. After I do my work they will be even more reliable
Those interested in pure reliability can run the boards at 72V 50A with total confidence
Those interested in pushing the envelope can run 24S lipo 100.8V with a 150A current limit
I will be charging a premium for the modified controllers but I will also be freely sharing every modification I make so it will be up to you guys - your time or your money
I took a loss on the 18 fet fiasco due to the slow start boards, my low pricing, and a $600 shipment I sent to a member that got lost in the mail.
I have $3k wrapped up in this new shipment so until I recover my investment - I will be a little tense :|
I will be offering a special discount for people who want to buy other stuff I have:
CA V2.11
26" 9x7 Rear 9C
ebrakes, throttle, 3-spd switch
torque arms, mosfets, etc.
I will start the new thread when the box arrives.
-methods
methods
1 GW
WOH! STOP!
It looks like you dont have an insulator between the fets and the heatsink
Is this true?
Do not apply power to that controller!
The fets must be insulated from each other (groups of 3) and the case.
-methods
It looks like you dont have an insulator between the fets and the heatsink
Is this true?
Do not apply power to that controller!
The fets must be insulated from each other (groups of 3) and the case.
-methods
methods
1 GW
Everything else looks good, assuming you have worked out the pin-pin connections externally.
I cant see any additional glue, if it is not there, you will want to re-glue the capacitors as the glue supplied was not good.
So just to repeat for everyone else watching from the peanut gallery, the fet spacing needs to be like this:
Phase 1 ____________ Phase 2 _________ Phase 3
+++ space --- space +++ space --- space +++ space ---
Groups of 3 can touch.
+ groups must not touch - groups
No fets can touch the heatsink.
Either the supplied insulator must be used or (alternatively) a strip of Kapton tape can be used as well (like Crystalyte does)
Don't use any sort of heat conducting epoxy or grease on the front (fet) side of the heatsink, only between the case and the heatsink.
-methods
I cant see any additional glue, if it is not there, you will want to re-glue the capacitors as the glue supplied was not good.
So just to repeat for everyone else watching from the peanut gallery, the fet spacing needs to be like this:
Phase 1 ____________ Phase 2 _________ Phase 3
+++ space --- space +++ space --- space +++ space ---
Groups of 3 can touch.
+ groups must not touch - groups
No fets can touch the heatsink.
Either the supplied insulator must be used or (alternatively) a strip of Kapton tape can be used as well (like Crystalyte does)
Don't use any sort of heat conducting epoxy or grease on the front (fet) side of the heatsink, only between the case and the heatsink.
-methods
gwhy!
100 kW
bikeraider said:
Yes you need to insulate the fets from the heatsink ( the fet tabs must not conduct to each other, test with a meter on ohms setting when they are all mounted ). You will need to clean off the compound and use the grey sort of tape yes.
liveforphysics
100 TW
Yep, that silicone fiber insulation tape is an absolutely critical part. Also, if you are using heatsink paste contacting the FETs anywhere, you know you are doing it wrong. The thermal paste is for the connection to the FET mounting bar to the case only, and the silicone fiber tape is for the connection between the FETs to the bar.
If you applied power to your controller, it would be just like directly shorting your battery.
If you applied power to your controller, it would be just like directly shorting your battery.
methods
1 GW
Remove all the screws.
clean off the thermal compound.
Align the insulator along the face of the heatsink. You can add a couple of pieces of tape on the back to hold it still if you want.
Line up the first mosfet - and you can just screw right through the insulator, you dont need pilot holes. Pilot holes can actually be dangerous if you cut them too large. If you find that you need pilot holes for whatever reason keep them small and you *must not* poke them in the wrong spot as this will open up the chance for an isolation breach.
** IMPORTANT ** the heatsink is not symmetrical so you must mark it before removal. Put it back on in the same orientation.
** IMPORTANT ** Before you remove the heatsink look along the fets. You will notice that they are grouped in "threes". No group of 3 can touch another group. If they are you must adjust the mosfets to achieve isolation. The easiest way is to ream out the hole of the two mosfets closest to each other. Lean them away. It is ok if they touch the other 2 in their own group.
This is no big deal - will probably only take you 10 minutes
-methods
clean off the thermal compound.
Align the insulator along the face of the heatsink. You can add a couple of pieces of tape on the back to hold it still if you want.
Line up the first mosfet - and you can just screw right through the insulator, you dont need pilot holes. Pilot holes can actually be dangerous if you cut them too large. If you find that you need pilot holes for whatever reason keep them small and you *must not* poke them in the wrong spot as this will open up the chance for an isolation breach.
** IMPORTANT ** the heatsink is not symmetrical so you must mark it before removal. Put it back on in the same orientation.
** IMPORTANT ** Before you remove the heatsink look along the fets. You will notice that they are grouped in "threes". No group of 3 can touch another group. If they are you must adjust the mosfets to achieve isolation. The easiest way is to ream out the hole of the two mosfets closest to each other. Lean them away. It is ok if they touch the other 2 in their own group.
This is no big deal - will probably only take you 10 minutes
-methods
methods
1 GW
When you are done take these measurements at a minimum:
Resistance: turn on the controller (with no battery attached) to drain the caps before test. If there is voltage in the system (even a little) it can give you screwy measurements.
+V to GND
(Should *not* read low resistance, but will read some arbitrary value in the Kohms - maybe 10K ohm)
GND to +V (habit, stray voltages can fool you, always reverse the leads)
(Should not read low resistance, but will read some arbitrary value in the Kohms)
Heatsink to +V
(Must be totally isolated - completely as in no change what so ever on the meter - not even a blip - OL)
Heatsink to GND
(Must be totally isolated - completely as in no change what so ever on the meter)
-methods
Resistance: turn on the controller (with no battery attached) to drain the caps before test. If there is voltage in the system (even a little) it can give you screwy measurements.
+V to GND
(Should *not* read low resistance, but will read some arbitrary value in the Kohms - maybe 10K ohm)
GND to +V (habit, stray voltages can fool you, always reverse the leads)
(Should not read low resistance, but will read some arbitrary value in the Kohms)
Heatsink to +V
(Must be totally isolated - completely as in no change what so ever on the meter - not even a blip - OL)
Heatsink to GND
(Must be totally isolated - completely as in no change what so ever on the meter)
-methods
John in CR
100 TW
liveforphysics said:
LFP,
Great offer, but once you add in my unsteady hands and poor near eyesight, it's an uphill battle that is probably best left as me paying someone else to make mods to electronics, despite the fact that I tried that already and just ended up with a blown controller and 15 wasted 4110's.
John
methods
1 GW
John in CR said:
You wasting my fets
-methods
Ok so I got most of it figured out. I used some house multi strand 12 awg wire and bilt up the traces then install the shunt and lifted the old shunts. I installed the diode and I think all the wires to install should be strait forward. But you sent me 9 extra resistors methods and I am not sure what they are for. I see they are in groops of three. But I should ask now before getting to far into this. Thanks again.
Arlin.
Oh an PS what watt soldering irons are you guys using I am using a 25 and it is a little weak to build up the traces with!
Arlin.
Oh an PS what watt soldering irons are you guys using I am using a 25 and it is a little weak to build up the traces with!
liveforphysics
100 TW
My favorite iron for this sort of work is a 10w-150w iron. I have 3 super fancy pants iron with digital temp control with DC PWM heating control, and they mostly collect dust since I got the 10-150w.
John, I'm going to buy you a brand new 10-150w iron, and a lifetime supply of my favorite solder type, favorite flux type, and my favorite type of brass tip cleaner, and I'm gona fix all your stuff with them, and give them to you.
The 10w-150w iron is so slick. When it's sitting idle, it's only got 10w going to the tip, so it just hovers roughly around 500degF, but if you try to use it for anything more than surface mount stuff, it looses temp rapidly. But! It has this button on the side, which shorts across the current limiting resistor inside of it, and it dumps 150w when you hit the button.
Irons that are high power always end up eating the tips much too fast from oxidization while just sitting not being used. They also can't be 150w, or they would melt down. With a 10w-150w iron, it can sit for hours and keep a clean shiny ready to use tip, and yet when you want to connect some 4awg cable, it's more than capable when you touch the button. Very slick little iron
John, I'm going to buy you a brand new 10-150w iron, and a lifetime supply of my favorite solder type, favorite flux type, and my favorite type of brass tip cleaner, and I'm gona fix all your stuff with them, and give them to you.
The 10w-150w iron is so slick. When it's sitting idle, it's only got 10w going to the tip, so it just hovers roughly around 500degF, but if you try to use it for anything more than surface mount stuff, it looses temp rapidly. But! It has this button on the side, which shorts across the current limiting resistor inside of it, and it dumps 150w when you hit the button
. Irons that are high power always end up eating the tips much too fast from oxidization while just sitting not being used. They also can't be 150w, or they would melt down. With a 10w-150w iron, it can sit for hours and keep a clean shiny ready to use tip, and yet when you want to connect some 4awg cable, it's more than capable when you touch the button. Very slick little iron
John in CR
100 TW
methods said:John in CR said:
You wasting my fets
-methods
Not me, an electronics repair shop. My fault though, since I already knew this was a "Do it yourself if you want it done right" area of the world.
LFP, Thanks, you're the man.
John
methods
1 GW
Arlo1 said:
Choose the set of 3 that meets your voltage requirements:
You have 1.8K's, 1.5K's, and 680's
1.8 || 1.8 || 1.8
input 70v~ 105v
1.5 || 1.5 || 1.5
input 60v~ 86v
1.8 || 1.8 || 0.68
input 50v~64v
1.5 || 1.5 || 0.68
input 30v~54v
I am using a 60W Weller WESD51
I set it to 850 degrees F and that allows me to solder those high power tracks like butter.
Heat will sag to around 650F if I stay on too long.
listen - dont risk cold solder joints. Make sure you properly heat-soak the parts and make a good bond.
Dont worry too much about over-heating the parts - you probably dont have enough iron to do that.
I have heated the traces up so much that the heatsink were hot to the touch
All this heat had to of traveled up the tiny legs of the fets.... that is A LOT of heat (too much)
I dont think you will get there.
-methods
Doctorbass
100 GW
The Best soldering iron to make beffy trace and keep stain melted to let you make a good job is one that have a BIG TIPS and moderate power.
A simple 80W with heavy tip.. like this will do the job perfectly!
http://www.canadiantire.ca/AST/brow...D~0586306P/Mastercraft+80W+Soldering+Iron.jsp
Sku: Product #58-6306-2
.. it will beat any 325W weller gun because other weller will cool down too fast.. and they can store less heat!..
That mastercraft tip can store more heat.. that's the key.
Doc
A simple 80W with heavy tip.. like this will do the job perfectly!
http://www.canadiantire.ca/AST/brow...D~0586306P/Mastercraft+80W+Soldering+Iron.jsp
Sku: Product #58-6306-2
.. it will beat any 325W weller gun because other weller will cool down too fast.. and they can store less heat!..
That mastercraft tip can store more heat.. that's the key.
Doc
Thanks Methods. I did see you post that and it may be in here even but I read alot of the first few pages again last night and could not find it. How about the funny ribbon wire? for the traces? And Doc thanks to you to I knew it was a curse to pay off the canadian tire card! lol I wont work any more on the traces till I get the stronger Iron.
methods
1 GW
That was solder wick.
I use that to build up the phase pads and parts of the main traces.
It is nice because it is so flexible.
-methods
I use that to build up the phase pads and parts of the main traces.
It is nice because it is so flexible.
-methods
