Instant Start 18 fet Infineon Boards are here...

[methods]

I like to use contact adhesive jell but any flexible glue will work.
Dont use CA (superglue) use a glue that is thick and rubbery.
I suppose you could use RTV (calk) or something like that.

I like craft glues because they stick to anything (including foam), they can be used as a contact adhesive (instant set), they are rubbery, inexpensive, etc.
The one drawback is that sometimes they "foam up" or "bubble up" and dont look very nice.
UHU Creative makes a nice line of glue that is safe for any material, stick to anything, dries rubbery. It is their Styrofoam glue. Popular in RC.

-methods

 

[Doctorbass]

Methods,

What can of glues you uses for the capacitor, to fix them in place ?

Thanks, good day!
Bikeraider

General purpouse silicone ( no odor or vinegar odor is the best).. the kind of silicone that smel near nothing dont have any acid vapor and will not be corrosive for the electronic components..

Otherwise the good Dow Corning RTV like these one are the right thing for that need!:

DC 748 Noncorrosive Nonflowable Low Odor— Use with corrosion-sensitive electrical and electronic equipment. Plus, it doesn't have the harsh vinegar smell of other silicones. Bonds metal, plastic, glass, and rigid PVC. Resists moisture and weather. Begins to harden in 30 minutes. Complies with FDA CFR21 177.2600 covering indirect food contact. UL 94HB listed.
748 3.0-oz. Tube -67° to +350° F Off-White 74915A654
8.02$

748 10.1-oz. Cartridge -67° to +350° F Off-White 74915A655
15.34$


ALSO AVAILLABLE AT LOW PRICE ON EBAY..

Doc

 

[Arlo1]

Good week and hope you have fun riding tonight, is summer here!
Bikeraider

Here too mid 20s and I was swetting in my shop fixing a motorcycle for a customer all day :-( When will these guys learn that there is almost no maintenance in an electric bike? Oh wait then I'd have less income......

 

[methods]

It was 100F here today and we rode our bikes to dinner, then again till 9pm

-methods

 

[Arlo1]

Hi,

The controllers is working but I need to connect the USB cable on my computer (with the computer on) otherwise I having no throttle. And the other thing I'm experiencing is turing in the wrong direction.

Can you tell how I can correct this (2) strange things or what is going wrong, thank you!
Bikeraider

Methods needs to correct me, but I belive that the throtle may be working when usb is connected because the usb is suppling power so you need to test power in the throtle cuircit (with the usb disconected.) And I have seen sensorless motors spin backwards and you just need to switch 2 of the fase wires any two BUT THAT IS ON A SENSORLESS I am not sure about the sensored motors. Deffinatly inspect the phase wires.

 

[methods]

Yea - the fact that the motor even turns is just luck.
You need to figure out the combination of phase wires and hall wires that match your motor with this controller
They are all different.
The correct configuration will make the motor turn forward and very quiet. Free wheel should be around 2A, if it is noisy or draws 3A or more this is not the ideal configuration.

You can swap any two phase lines but you must also swap the two hall lines.
This may or may not garner the desired result :|

As far as the USB thing -> that is a problem.

Questions:

Did you hook up the switch? Between where and where?
What voltage are you running? Which resistors did you install?
When you turn the switch on, what voltage do you read at the RIGHT side of the big resistors and on the LEFT side of the big resistors?
What voltage do you read on the +5V rail?
On the bottom of the board is a TO-220 package. That is a 7805 5V regulator. Read the voltage on each pin with respect to GROUND.

PLEASE!!! BE CAREFUL WHEN PLAYING WITH THE BOARD WITH POWER APPLIED!!!
If you roll the board onto a tiny piece of wire or solder you can short the 100V lines to the +5V lines and blow it out.

Also - dont test this anymore with the USB applied.
Something is wrong and you should find it.

-methods

 

[methods]

If you already assembled the controller you MUST discharge the main capacitors before disassembly!!!
Do this by turning on the circuit. Measure the voltage at the main power leads till it is low then apply a short to get the rest of the voltage out.
Must be a 0V before moving forward to avoid any type of possible mistake.

-methods

 

[Mike1]

I finished building my controller this week. I've been running it for a couple of days, lots of torque, great acceleration and speed. But that was before I programmed it. Now that it is programmed it feels like my old 12 fet soft start controller, poor torque, slow acceleration.

I've set the current limit to max (I'm still using the original shunt).

I'm confused. Is this why Doc was getting such good performance with the stock shunt, because he hadn't programmed it?

Could this be a block time setting effect?

Here are the settings.

 

[Mike1]

I've now tried changing the block time to 5 and then 10. Didn't make any difference. I can only conclude that before it was programmed current limiting was not active and now it is. The only other thing I did beside programming was to change the 3K base resistor and that should not have had any effect.
I'll have to wait for the 250u shunt to arrive now to get it back up to decent power.

 

[methods]

Hi Methods,

i'm continuing troublshooting it tomorrow but i send you a pic of the swith and the resistor for now.

Good day!
Bikeraider

Can we see a picture of the right side of the board?
Any chance you cut the +5V line while you were cutting back the traces?

-methods

 

[methods]

I finished building my controller this week. I've been running it for a couple of days, lots of torque, great acceleration and speed. But that was before I programmed it. Now that it is programmed it feels like my old 12 fet soft start controller, poor torque, slow acceleration.

With the stock shunt and the settings on MAX you should get about 60A of current - that is hardly weak.
Something else must have changed in your setup.

*** Please list every single thing that was done between when it worked well and when the current limit came down.
Sounds like you did some soldering?


Regardless - next time you have the controller open just solder the shunt down to around 250u Ohm
You can then vary the current between 5A and 120A

There is no value in keeping the original shunts in the original configuration.
They are too large and too sensitive to temperature - at any current.
The lower your shunt value the lower the power it will dissipate so the less it will swing with temperature.
The loss of resolution is tiny compared to the temperature swing.

-methods

 

[Mike1]

Methods,
I thought the stock shunt would only give about 30 amps on maximum? I'm waiting for Geoff to send me the 250u shunt from Digikey.
I don't think it was current limiting before, it was violently powerful, in traffic I had give the car in front a good head start to avoid running into it. The bike weighs about 260 pounds and I weigh 170 and it was reaching 30 mph in under 4 seconds. It was scary fast. I suspect it is now performing like it should with the standard shunt and there may be an intermittant connection on the VSense from the shunt that is now making contact. I'll open it up and check it out.

All I did was change the 3K base resistors for the high power one and program it.

I've now increased throttle to 110 percent and that has helped a little.

 

[methods]

I've now increased throttle to 110 percent and that has helped a little.

This should make no difference...
The speed setting at >100 will just allow you to go a little faster - it should not affect acceleration.
Are you saying that you are both speed and acceleration limited?
Are you saying that acceleration went up when you turned the speed limit up?

What this sounds like to me now is you are not getting 100% of your throttle signal.
How did you hook up the CA? Is there a resistor inline with your throttle signal?

Please measure the throttle signal where the wire solders into the controller.
I suspect that it is not getting to 3V

-methods

 

[methods]

we find it this morning with Mike1, he helping me allot

Ha! I got it right.

Where were you guys working on it? Did you start a new thread?
Oh well, problem solved.

-methods

 

[Mike1]

I didn't expect the 110 percent to make any difference either. I haven't tested it properly on the road after the programming so I don't know if top speed has been affected.

My CA hasn't arrived yet so no wires attached yet, I soldered in the diode ready for the throttle override but that's all.

I'm not really treating this as a problem, it is behaving now exactly as I would have expected with the stock shunt. I was just a little surprised that it performed so well before the programming and thought that maybe these boards came totally un-programmed from the factory and so the current limiting was inactive cos that's what it felt like. I'm sure once the 250u shunt goes in I will set it to 30/80 and it will be plenty powerful enough.

I'll do some voltage measurements tomorrow (it's dark here now) but I don't expect to find anything wrong.

Bikeraider PMed me this morning asking for help with his controller, not sure why he chose me, maybe cos I was online. He gave me the details of the problem and once we'd established that it was getting 5V from his computer USB port it wasn't hard to work out where the problem was. I asked for the same picture that you asked for. I'd already worked out that you had to keep one of the 5V vias intact when cutting back when I did my board.

 

[HAL9000v2.0]

As a shunt gets smaller (250u Ohms in this case) the S/N gets lower....
We will see how much the solder tabs vary the resistance value after I do a few.
My guess is that it will be negligible when compared to the shift with temperature.

In my perfect world it would be a slighter tighter fit (and on top of the board) but this is damn good for $1 and 5 minutes.

I am going to (at first) continue to tap in the same spots: On the GND side it will be the closest hole. On the PHASE side it will be the the trace leading to the chip. Even if this is not the idea tap point it will be consistent from board to board. I dont like soldering wires right to the shunt - too janky.

-methods

P.S. Anyone who has bought a kit or CA from me is entitled to 1 or 2 of these for the price of shipping.

Do I leave original shunt wires? I guess they will not have much influence.

 

[methods]

Do I leave original shunt wires? I guess they will not have much influence.

I have been cutting mine out but that was when I was under the delusion that I would end up with something like 245 - 255 uOhms
It turns out that I have been landing at more like 270u ohms.
I think the stock shunt is something around 1 - 2 mOhms right?
You can leave them.

The one argument I would have is that the 250u Ohm precision shunt is very stable across temperature.
I cant speak to the 'M' shunts - my guess is that they are somewhat thermally unstable.
This could make your measurement less accurate at higher currents.

So - if it is easy I would cut them out.

-methods

 

[methods]

Hi Guys,

Ok i'm sending a picture about the fix for my 5 volt trace what do think about it ?
Thanks
Bikeraider

I dont have the circuit in front of me but IIRC the +5V traces were pretty thick over on that side of the board
There are a lot of fast transients and it is critical that the +5V section remain as stable as possible
This fix you made will certainly get the board working - though it is not clear if this will lower your noise immunity or not.

My guess is that it will be fine

-methods

 

[Mike1]

I would have reconnected it in the corner where it was cut and used the nearby hole to pass the wire to other side of the board, but I really can't imagine that there will be a problem with the way you have done it.

Did it work?

 

[Arlo1]

OK so time to ask for some info again.
I just finished soldering all the small wires that I think I should need
Methods are these more or less a user defined set?
#1. I installed the 6 wires for the CA.
#2. 3 wires for a throtle
#2. Usb conector
#4. 6 wires for the hall sensors and are they suposed to have 2 5v positives and why?
#5. used the 4 wire to run regen (bk-gnd) and ebrake (ebrake- & ebrake+) out of the case for brake switches
I have one 6 wire conector left over but I think I can use it in the future for something like the 3 speed or somthing unless I am missing something important?
I am also looking for the trace stevo cut back and can't find a picture of it is there one up?

 

[Mike1]

Arlo1,

You only need 5 wires for the halls, not sure why you have 6.

Ebrake- is for brake switches, you will need to add a diode to the board for it to work.
Ebrake+ is for a 12V brake signal, for use on a motorbike where the signal would be taken from the brake light feed. You need to add a transistor if you want that to work.

 

[Arlo1]

Arlo1,

You only need 5 wires for the halls, not sure why you have 6.

Thats what I thought I think it muse be just a universal wire but I was trying to look at all the pictures to see where the second red wire went.

Ebrake- is for brake switches, you will need to add a diode to the board for it to work.
Ebrake+ is for a 12V brake signal, for use on a motorbike where the signal would be taken from the brake light feed. You need to add a transistor if you want that to work.

I will remove the wire to ebrake+ and run it to ground so I can use a switch thanks alot.

 

[Mike1]

You'll need to install diode D32 and then you switch EBS- to GND for brake signal.

 

[Arlo1]

You'll need to install diode D32 and then you switch EBS- to GND for brake signal.

Yea I got that in there as well. thanks.

 

[Arlo1]

Mike1,

You the man I have try the configuration your suggest with my X5304 and sending here at ES by Dirtdad. The motor is now spinning in the right direction and it turn smoothly 8) I' m copying those wiring connection here, for other people with a X5304 like me.

Phase Wires

Blue Yellow
Green Green
Yellow Blue

Hall Sensor Harness

Blue Green
Green Blue
Yellow Yellow

P.S like frankstein says It's alive, it's alive, i'm very happy sorry :lol:

Thank's again Mike, nice find cheers!
Bikeraider

So is it fast? 20 inch wheel right? I am buildng the same thing for now, I can't wait till the self build motor is done.