Instant Start 18 fet Infineon Boards are here...

[docnjoj]

Welcome back Knuckles!
otherDoc

 

[methods]

hi
I did not find the fault with the sensorless boards i............ <review of sensorless boards>

Geoff

Hey, thanks for the info Geoff!
Sounds like I need to distribute these and get more test info.

-methods

 

[methods]

I always wanted to play with one of those sensorless boards.

Do you want 1 or 2?
Last night I bagged them all up and I think I saw your address scratched into the bathroom wall here at work.
I can get them sent out today for Free-99.

-methods

 

[methods]

I ran the 100V 100A test.
Both last night (@ 90V) and again on the way to work (@100V)

Results were inconclusive. Turns out I have too much inline resistance between the controller and the motor to develop much more than 150A peak.
That and the fact that I cant get the front tire to stick Hard to load a motor to 100A when it is just spinning.
I took video but it is stuck on an SD card in my pocket.
I will develop it at lunch and post it along with a review.

To do a proper review I am going to need to modify my motor with larger gauge wire and stickier tires.

I think it may be possible that the 500uOhm shunt setting was actually more like 100A and I just did not have enough load.

-methods

 

[Doctorbass]

Methods.. I actually HAVE THE SETUP FOR THE LOAD YOU NEED

just in case.... :wink:

5305 at 100V or 5303 at 100V would work!

Doc

 

[methods]

100.8V 100A Test

Motor: 9C 9x7 with 5' long tiny phase wires presenting substantial constant resistance
Controller: Shunt set to 230uOhms for something >120A current limit

I figured out that if I face the CA forward, lean all the way over the front wheel, put my chin on my chest (looking down at the CA) I can wrap the throttle around and get it to pull ~100A all the way up to 25mph. After that it starts dropping off rapidly to 60A then 30A. After I did this test the phase wires were so hot that I could not hold onto them. The motor was warm and the controller was luke warm.

Obviously I am no longer fit to continue this testing so I am going to ship the controller to Gary.
I am in the process of building up 2 more controllers and I will use those on the Project Mayhem bike to do a proper test. Project Mayhem has 3" wide soft rubber tires that wont let go. This Walfart bike I am testing on is about to give it up anyway... The front fork is wobbling back and forth about an inch :?

Gary - it is probable that the current limit is set >>120A so I want you to go easy on the poop-stick ok?
I will send the new Parameter software along as soon as I get it.
I will solder in your programing header and attach the USB cable for you before sending.

I have received a few requests for kits. I will release kits as soon as I have verified the programming software works. I want to get this shunt/current limit/software clear before distributing. IR shipped my fets already so I will have them very soon. Kits will be available with the 4110's at a discount

Unfortunately guys, I wont be able to sell the 4110's as cheap as I was before. We have lost that deal. They will still be half what they would cost you on Digikey though

-methods

 

[methods]

Video showing that the Instant Start can run at about 1MPH


Can someone please do whatever magic is required to make this video small and nice?
I just cut it down to size and left it in a high quality format. It is about 16M

http://www.deviantmethods.com/infineon/video/Infineon_9C_Test_001.wmv

When the wife comes home for lunch I am going to get her to take a movie of me doing my (now infamous on my street) 0 - 30MPH burnout
The on board vid had so much wind noise that it was pointless.

-methods


Edit:
PLEASE DO NOT STREAM FROM MY SERVER. Right Hand Cliick - Save As

 

[geoff57]

Video showing that the Instant Start can run at about 1MPH


Can someone please do whatever magic is required to make this video small and nice?
I just cut it down to size and left it in a high quality format. It is about 16M

http://www.deviantmethods.com/infineon/video/Infineon_9C_Test_001.wmv

When the wife comes home for lunch I am going to get her to take a movie of me doing my (now infamous on my street) 0 - 30MPH burnout
The on board vid had so much wind noise that it was pointless.

-methods

hi
what size do you want and have you got a pic of the bike this was done on.

I have just finised a 480 x 360 medium quality WMV this can be uploaded to youtube as is if wanted i have sent other videos to youtube in this format size of file is 3.1Mb will that do you, a picture of the bike would finish the video off nicely IMHO.

Geoff

 

[methods]

Hi Geoff,

Whatever size you think is appropriate.
In the past people have complained about my WMV format and large size.

I am sorry, no pictures of the bike as it is a somewhat shameful Walmart bike.

Thanks,
-methods

 

[geoff57]

Hi Geoff,

Whatever size you think is appropriate.
In the past people have complained about my WMV format and large size.

I am sorry, no pictures of the bike as it is a somewhat shameful Walmart bike.

Thanks,
-methods

hi
how do you want the vid i just finished then?
on youtube or sent to you direct the second will take longer.

Geoff

 

[methods]

YouTube is great. Thanks Geoff.

100V 100A 9C Burnouts

The phase wires were so hot after I did this that I burnt my hand
I am talking hot like I got real worried hot...
The 10AWG at the controller was not even warm though

44Meg
Pedal-pumping-burnout
A fly-by
More burnouts
A leaning over the handlebars launch (by this point the battery and phase wires were heat soaked so the power was a little more managable)

http://www.deviantmethods.com/infineon/video/100V_100A_Infineon_9C_Burnouts.AVI

And yes, before anyone asks, I am a Pastafarian.

So how do you like them apples for a $150 hub motor, a very affordable controller, and Hobby City Lipo?
Ebike power is coming into reach for the common man.
What common man could want more power than is shown in this video?!?!

-methods

PLEASE DO NOT STREAM FROM MY SERVER. Right Hand Cliick - Save As

 

[geoff57]

hi
I'm going to try and incorprare that last video in with the first to get a better effect should get it in for between 5 and 6 Mb
Geoff

 

[julesa]

AWESOME Patrick! That first video really shows off the bike's excellent cornering characteristics.

The phase wires were so hot after I did this that I burnt my hand

Are the phase wires the stock 16g?

 

[methods]

AWESOME Patrick! That first video really shows off the bike's excellent cornering characteristics.

The phase wires were so hot after I did this that I burnt my hand

Are the phase wires the stock 16g?

yea, whatever the wires are that come stock with an Amped Bike setup. Very long too, they go from the front hub all the way to the back and then I have it wound up with a zip-tie.
I just took the motor out of the box, changed the connectors to Anderson and Clyte then hooked it up
They are Teflon coated wires so I am not too worried about them shorting out.
I am more worried about the phase wires melting the insulation off the hall wires and having those short out (which I have seen)

Dude, all of my bikes are in ruins. I just keep juicing them up till they explode :|
This is the only setup I have going right now.

-methods

 

[Doctorbass]

Thanks for the vid Methods

I personally think that the real power of an ebike can be seen usually with rear wheel... The front tire that are too skidding on the ground dont really show power.. it show missing traction and lost of rubbers...

I would love to see an ebike doing the same with the rear wheel in the same conditions !! noh?

Now double these phase wires and go reverse !! lol.. i'm kidding!

anyway.. great vid man!

Doc

 

[PaulM]

When the wife comes home for lunch I am going to get her to take a movie of me doing my (now infamous on my street) 0 - 30MPH burnout
The on board vid had so much wind noise that it was pointless.

You should do more of those burnouts

Something all men would like to hear from their wives . . . :lol:

Awesome video!!!

 

[geoff57]

methods
check your PM inbox I have something for you
Geoff

 

[swbluto]

Some kind of indicator would be nice, for errors and stuff. Audio indicators would be nice so that a person can get an idea what's going on while riding. I just purchased a few SMD mylar speakers, and the can-be-loud kind only cost something like $2 and less.

Also, having the ability for the end user to add on a sensor (as simple as one extra plated through-hole in the pcb) and being to tell the controller what to do if the sensor goes below or above a programmed value(as simple as updating the programmable software) would be helpful. Example, if the voltage level goes above 4.5 volts from a 5 volt temperature sensor corresponding to 300 degrees fahrenheit, the user could select a) pull down the throttle line b) sound the audio indicator with a specific choosable audio pattern c) whatever else seems like a reasonable response sequence. Then the user could add temperature protection to the controller or motor or whatever, really if they feel like they might risk burning out something(Such as for the power freaks). And, it's just that, one extra through hole and the requisite programming. There'd really be no extra pcb costs since it's just one more hole, and this presumes the user buys the sensor and installs it themselves. Of course, there should be support documentation of some sort, which I'm sure people would be happy to write up on the forums. Also, it'd be cool if there was some "open source" playground for programmable software features so people could make the controller more feature rich if they wanted to on their side, but that'd take some development which it seems doubtful keywin or whomever would invest in.


Edit: Well, if methods doesn't want this message to be responded to and that's why he didn't keep it on here, and that's why my post might seem a little tangential, then don't respond to the following copy I made. This is just for justifying my post's raison d'etre, that I so don't want to delete as I put hard work into it!

Geoff was kind enough to fix the video up.
He did a good job. i will upload it to my server a little latter.

I also have some more pron pictures.... Something showed up in the mail

So I am going to start an open debate in a few posts here...
I need you guys to put away your extremest hats and put on your "normal" hats.

We are going to debate how to configure the controller

* Phase wire length
* Power wire length
* Wire gauge
* Connectors for phase and power

* Conventions
* Compatibility
* Sticking to Crystalyte connectors
* Cost vs. reliability vs. ease of use

* Which wires to bring out (Cruise control, brakes, regen, external power switch, etc)
* Keeping it simple (like ebikes.ca)

We need to talk about all the tradeoffs.

Features vs. reliability
Overkill vs. "enough"
Wants vs. needs

This is your chance to literally build the controllers how you want them... but ....

DONT be too fast to spew out every feature you ever wanted because you would be surprised at just how much it can increase the cost
Every connector we add decreases reliablity, increases parts cost, increases congestion inside, increases build time... Just adding one cable could add $10 to the price of the controller.

I really like Justin's idea of keeping it simple.
I also like the idea of bringing out one master connector that has all the bells and whistles in it that most people wont use

anyhow. . .

Start thinking about that.

-methods

 

[Doctorbass]

Some kind of indicator would be nice, for errors and stuff. Audio indicators would be nice so that a person can get an idea what's going on while riding. I just purchased a few SMD mylar speakers, and the can-be-loud kind only cost something like $2 and less.

Also, having the ability for the end user to add on a sensor (as simple as one extra plated through-hole in the pcb) and being to tell the controller what to do if the sensor goes below or above a programmed value(as simple as updating the programmable software) would be helpful. Example, if the voltage level goes above 4.5 volts from a 5 volt temperature sensor corresponding to 300 degrees fahrenheit, the user could select a) pull down the throttle line b) sound the audio indicator with a specific choosable audio pattern c) whatever else seems like a reasonable response sequence. Then the user could add temperature protection to the controller or motor or whatever, really if they feel like they might risk burning out something(Such as for the power freaks). And, it's just that, one extra through hole and the requisite programming. There'd really be no extra pcb costs since it's just one more hole, and this presumes the user buys the sensor and installs it themselves. Of course, there should be support documentation of some sort, which I'm sure people would be happy to write up on the forums. Also, it'd be cool if there was some "open source" playground for programmable software features so people could make the controller more feature rich if they wanted to on their side, but that'd take some development which it seems doubtful keywin or whomever would invest in.


Edit: Well, if methods doesn't want this message to be responded to and that's why he didn't keep it on here, and that's why my post might seem a little tangential, then don't respond to the following copy I made. This is just for justifying my post's raison d'etre, that I so don't want to delete as I put hard work into it!

Geoff was kind enough to fix the video up.
He did a good job. i will upload it to my server a little latter.

I also have some more pron pictures.... Something showed up in the mail

So I am going to start an open debate in a few posts here...
I need you guys to put away your extremest hats and put on your "normal" hats.

We are going to debate how to configure the controller

* Phase wire length
* Power wire length
* Wire gauge
* Connectors for phase and power

* Conventions
* Compatibility
* Sticking to Crystalyte connectors
* Cost vs. reliability vs. ease of use

* Which wires to bring out (Cruise control, brakes, regen, external power switch, etc)
* Keeping it simple (like ebikes.ca)

We need to talk about all the tradeoffs.

Features vs. reliability
Overkill vs. "enough"
Wants vs. needs

This is your chance to literally build the controllers how you want them... but ....

DONT be too fast to spew out every feature you ever wanted because you would be surprised at just how much it can increase the cost
Every connector we add decreases reliablity, increases parts cost, increases congestion inside, increases build time... Just adding one cable could add $10 to the price of the controller.

I really like Justin's idea of keeping it simple.
I also like the idea of bringing out one master connector that has all the bells and whistles in it that most people wont use

anyhow. . .

Start thinking about that.

-methods

Instead of speakers.. Why not use the motor coil.... like the RC stuff do :wink: Magnet + coil have allways been able to make sound!

Doc

 

[swbluto]

Good idea! Can sound be played through the motor while it's moving?

 

[julesa]

This is what mine sounds like:
http://www.youtube.com/watch?v=FjMWOzeuRiw&feature=related

 

[methods]

Edit: Well, if methods doesn't want this message to be responded to and that's why he didn't keep it on here, and that's why my post might seem a little tangential, then don't respond to the following copy I made. This is just for justifying my post's raison d'etre, that I so don't want to delete as I put hard work into it!

Sorry about that!
Did not mean to leave you hanging.
Right after posting I considered thinking it through some more - maybe setting up some kind of focused hit-like this:

Phase wires - All the way to the motor or short
Power wires - 45A anderson, 75A anderson, bare coper
Phase and power wires - 10 AWG or 8AWG
Brake connector - Yes or no
Cruise Control - Yes or no
3 speed - Yes or no

A kind of focused questionare with a comments section at the end.

Your input is valued. Thanks for taking the time to post.
If I had seen your response I would not have deleted my original post.

Your idea of adding an optional sensor is actually fairly easy.
In that case, I would add a micro controller to the mix with an opto-isolator
For instance, the uC could limit the throttle setting to 50% (like the CA does) if the input voltage (say and RTD) crosses a certain threshold.

Something like this is an investment up front but nearly free on the back end.
I work with a specific uC (the MSP430) that comes on a board the size of your thumbnail with everything ready to go - A/D's, D/A's, Capture Compare, Timers, Counters, Digital I/O, a 16Mhz clock, Ultra Low (sub 1uA) power consumption, ...

Anyone can feel free to talk about anything they want in my threads - so fire away!
I like free thinking tangential threads.

-methods

 

[swbluto]

Oh, ok. I was thinking that the controller already had some kind of microcontroller with unused A/D ports or something, and that it could be used for possibly more. Anyways, it seems like the per-unit additional cost might not be worth it if it were some "general model" meant for the masses as price is one of the infineon's main attractions, but if "bells and whistles" could be easily added on with the microcontroller, then the expansion of the feature set might justify it.

Anyways, I was just looking up digikey for those incredible specs and I noticed it has pretty decent A/D resolution and quite a few A/D ports and it's nearly as cheap as the 8-bit atmega168 AVR's that I currently program! When you say micro-amps, do you mean during sleep mode or during actual operation?

 

[methods]

This is what mine sounds like:
http://www.youtube.com/watch?v=FjMWOzeuRiw&feature=related

Nice - Queue that song up and press play before viewing these.





Tomorrow morning I am going to solder up some proper phase wires for the 9C.
I will set up the camera on a tripod and get a good acceleration video with no tire scrub for Doc. I can tell he is getting agitated
I actually have a rear 9C but it is crooked as hell---- Amped bikes should not advertise that as "dished correctly".
I see why Justin does not sell them.
They have promise, but they need work.

Here is the Geoff remix - 4MB
Slow speed demo followed by some angry burnouts
http://www.deviantmethods.com/infineon/video/methods_combi_1.wmv

-methods

 

[methods]

Anyways, I was just looking up digikey for those incredible specs and I noticed it has pretty decent A/D resolution and quite a few A/D ports and it's nearly as cheap as the 8-bit atmega168 AVR's that I currently program! When you say micro-amps, do you mean during sleep mode or during actual operation?

This thing is only $20
You get the emulator/programmer + 3 target boards.
The target board is the tiny rounded part at the end.

For those of you who dont know what you are looking at - THIS THING IS AMAZING. The thumb drive part allows you to not only program and power the chips but you can also do in-circuit real time emulation... i.e. you can actually go into the code and set breakpoints, monitor values, pause the code and change values... right on the real hardware. Totally changes the game. You can code it in C++ to with dynamic memory allocation, OOP and the works.

After you do you magic just pop the little booger off the end and you are set! Solder in a single cell Lithium and it will run for 30 years.

These feature either an MSP430F2013 or an MSP430F2012
Additional target boards are $4 each

The A/D is 16 bits and you can do all the fancy stuff.... Auto scan and all.
The sleep current is sub 1uA, but this thing can sleep like 90% of the time
I develop 100pin devices that use only 250uA on average

Here are the chip specifications:
http://focus.ti.com/docs/prod/folders/print/msp430f2013.html

Here is the order page with prices:
http://focus.ti.com/mcu/docs/mcuprodmsptoolsw.tsp?sectionId=95&tabId=1203&familyId=342&toolTypeId=1

Here is the manual. Check out the end of the manual for the pinouts etc.
http://www.ti.com/lit/pdf/slau176

I built one of these up for Luke for a project but I think he fed it to his dog instead of using it
(thats right Luke - you are ish-listed buddy! Like the "Soup Nazi" from Seinfeld - "NO INFINEON FOR YOU!"

-methods