Methods 100V 100A Programmable Regen Controller

methods

1 GW
Joined
Aug 8, 2008
Messages
5,555
Location
Santa Cruz CA
There is no greater value in a high power full featured ebike controller on the market today

This controller is the awesome result of a group effort to explore the new 116 based 18 fet controller boards. Everything you see here is what happens when we develop in a pubic forum.

This is an Open Source Mod Thread and anyone is free to duplicate any of the mods you see here for profit or personal use. Our number one goal is the advancement of the Electric Revolution and screw anybody who just wants to make a quick buck :p


022_100V_100A_Programmable_Regen_Controller_Finished.jpg


Electrical
Voltage range: 36V - 100V
Peak Current: >300A
Current Limit: 5A - 120A
Regen Voltage: < 90V


Features
Fully Programmable via supplied USB cable
Regenerative Braking
Cruise Control
3 Speed control
Pedal Sense
Cycle Analyst Ready
Power Switch
Power Indicator
Reverse Switch Grip
60 degree or 120 degree compatible


Whats Inside
Genuine IRFB4110 Mosfets
Precision Calibrated 4W shunt
10AWG Teflon coated tinned copper cabling
6AWG equivalent traces
Precision 1% Reference

Reliability
100% Testing
Cleaned and inspected
Conformal coated
Glued Capacitors
Large Heat Dissipating Case
High Pedal Lockout
Reverse Lockout


Modifications
90V Regen Mod
100V Power Resistor Mod
3K Base Mod
250uOhm Shunt Mod
50V / 100V Switch Mod
Programming Mod
10AWG Wire Mod
4110 Fet Mod


Programmable Features
DC Current Limit
Phase Current Limit
Low Voltage Cutoff (LVC)
Max Regen Voltage
Regen Strength
Global Speed Limit
3 Switch Selectable Speed Limits
Reverse Speed limit
Throttle Over-voltage Protection

10 Example Programs supplied - Simply import and download!


Pricing
$300.00 USD
Controller, USB Programmer, Reverse Switch Grip, Mating Connectors

Also Available: Cycle Analyst V2.11, eBrakes, Cruise Control, 3-Speed switch, Mating Connectors, CA Extension cables, etc.

Buy Here

017_100V_100A_Programmable_Regen_Controller_MarkedUp.jpg


Included Revers Grip!

View attachment 8

044_100V_100A_Programmable_Regen_Controller.jpg
019_100V_100A_Programmable_Regen_Controller_Finished.jpg
(above) This button allows you to switch the controller between 100V range to 50V range.

009_100V_100A_Programmable_Regen_Controller_Finished.jpg
(above) The traces end up as a 6AWG equivalent. Your input connectors WILL melt before these traces peel up

047_100V_100A_Programmable_Regen_Controller.jpg
008_100V_100A_Programmable_Regen_Controller.jpg
010_100V_100A_Programmable_Regen_Controller.jpg

Above:
Dont underestimate this "3 speed switch". It is actually much better than a speed governor - it is a throttle governor. The difference is that a true speed governor would allow you to accelerate at your full current limit up to the new maximum speed. Useless really. A Throttle governor on the other had governs the throttle input signal which in turn governs the current limit in a sense. If you set the speed limit to 50% it will be the same as applying only 1/2 throttle so you dont pull as many amps.

Example: Using these 3 settings in the software with a current limit of 100A
100% - Applying full throttle results in huge bursts of current and acceleration up to top speed
50% - Applying full throttle results in moderate acceleration up to half speed
25% - Applying full throttle results in the bike just creeping up to 1/4th max speed

Anyone who has ever had a 100A bike will understand - this is hugely useful. When you want to get crazy you want a very responsive throttle but when you are just cruising with the wife this is wasteful and uncomfortable. Setting the speed to medium or low will make the throttle much less responsive and will make your pack last a lot longer.

Awesome feature!

015_100V_100A_Programmable_Regen_Controller.jpg


This thread shall serve as the Technical Discussion Area for the
Methods 100V 100A Programmable Regen Controller


* Questions
* Requests
* Suggestions
* Development

There will be a separate "For Sale" thread where you can ask questions about pricing, shipping, combo's, etc.
Please - Lets try to keep it Technical :D


As you can tell this is a Crystalyte spin-off of the "18 Fet Instant Start Infineon" boards. The Chip is a 116 and it is NOT an infineon but you will find that much of the information we have learned about the Infineon controllers applies.

Please cross link this thread.

-methods
 
.
.

Buy Here <- <- <-



WORK IN PROGRESS


How To

This slot is where I will maintain a current list of how to perform each mod used
I will fill it out with pictures as I go.

Please understand - these are my personal build notes so they may be a little vague. I will flesh them out with time.

Assumption:
You are starting with a 72V 50A Crystalyte 18 fet Controller built to "Methods Specification"
Basics like building up the traces, running 10AWG wire, running all the pigtails, etc. are not addressed here.

If you are starting with a Standard 72V 50A 4110 (Like a Group Buy Controller) you must remember to swap all the cabling for a larger gauge.

If you are starting with a Bone Stock 72V 50A 18 fet controller you must swap all the fets for IRFB 4110's or better.


Prep

1) lay out paper towel
2) Set up small tip on Iron
3) Short power leads and toggle power switch
4) Remove 8 endcaps screws and 5 heatsink screws
5) Clean up white goop


Inspect board & touch up

1) Hit cold solder joints
2) Fill in anywhere this is light on solder


90V Regen Mod

1) Remove R12
2) Solder in 1K between via by R35_1 and GND
3) Short BK to GND


100V Power Resistor Mod

1) Solder 3K resistor in parallel with Power resistors anywhere


50V – 100V Switch mod

1) Cut out one 1.5K resistor
2) Solder new 1.5K resistor to switch lead
3) Cover with shrink
4) Trim resistor lead
5) Solder to bottom of board
6) Solder other switch lead
7) Tuck switch next to main caps on left
a. Long throw means open – 100V
b. Short throw means closed – 50V


3K Base Mod


1) Scrape RS13 & RS12
2) Solder left lead to main pad
3) Solder right lead to 3 open SMT pads
a. Hit all three right pads for stronger bond
b. Don’t waste time scraping!

Programming Harness

1) Don’t strip first
2) Run long 3/16ths shrink
3) Cover backshell with 1/2” shrink (maybe smaller)
4) Feed through PHASE grommet
a. Push out a hole with the small needle nose
b. Feed to the shrink point
c. Push the pliers through again
5) Strip only 1/8th inch
6) Solder into board
7) Short +5V to Prog. Enable


250uOhm shunt


1) SWAP TIPS TO LARGE TIP
2) solder on back


Build Up Traces

1) Solder 12AWG solid core wire in parallel with the existing 12AWG


Build USB adapter

1) Cut 3.3V pin
2) Build up top of 4 pin connector (or Polarize)
3) Heat shrink at some point


Spray clean board



Spray conformal coat


1) Cover power resistors
2) Cover back of heat sink
3) Cover 5V regulator
4) Spray top and bottom


Glue Capacitors


Reassemble

1) Apply heat grease
2) Slide together
3) Fasten heatsink
4) Position Button
5) Fasten endcaps


Set up Andersons

1) Cut back
2) Strip short
3) Do power and phase



WORK IN PROGRESS


-methods
 


WORK IN PROGRESS


Here I will keep awesome pictures that are relevant.

* Board closeups
* Schematics
* Pin Descriptions
* Booby Girls
* Data Sheets
* Etc.

!!! MOVIES !!!

This pictures shows that even with a pair of 12AWG wires on the main power buss there is still a huge gap between the controller board and the case. This Crystalyte case is MUCH larger than the Infineon case. (below)


(sorry for the quality - cell phone picture :roll: )
012_100V_100A_Programmable_Regen_Controller_Finished.jpg

.
.

Note the huge gap to the right of the circuit board. This is the location where shorting occurs on the Infineon boards. With the Crystalyte design there are standoffs to keep the board off the bottom of the case and large space between the edge of the board and the case. No Shorting between the +V and +5V buss. (below)

018_100V_100A_Programmable_Regen_Controller_Finished.jpg
.
.
.

These are the settings I use for testing all the features
View attachment 3
.
.
.
These are the settings I used to use for the Infineon Version
AlmostFullBlast.jpg
.
.
.
These settings will put you at about 34A depending on your shunt. Notice that I keep the ratio of Phase Amps to DC Amps the same. This was for a 280uOhm shunt. For a 230uOhm shunt (newer controllers) the resulting current will be higher.
View attachment 1
.
.
.

Same example but this time it works out to 65A with a 280uOhm shunt
DC65A.jpg


HOW TO PROGRAM

There are a few tricks to programming these things.

1) Get the software here in zip format
2) Get the drivers you may need for the software here and here.
3) Get the drivers for the USB dongle here in rar format
4) Install everything
5) Run the software and plug in the USB cable
6) "Point" the software to the correct com port. Must be 1-5. If your computer assigned a higher port number then you can manually change it. That can be dealt with later.
7) Upload settings - here is a zip file full of examples
8) Start the software BEFORE plugging into the controller
9) Insert the USB programmer into the controllers 4 pin programming harness. The button on the connector goes UP, the visible pins go down.

As soon as you plug in the connector the program should download and indicate success. Here is a troubleshooting area.


CANT GET SOFTWARE INSTALLED

* Please post your question in the thread below. This is the easy part :p


CANT GET SOFTWARE TO RUN

* Did you install the two drivers? If you are on Vista, try running as administrator


CANT RECOGNIZE PORT

* Go into Device Manager and look in the com port section. Look for the USB device. Make sure it has a com number of 5 or less. You can manually assign com numbers by going into the advanced properties and manually setting the com port to a lower number. Com's 1 and 2 are usually open. These were reserved on older computers for things like parallel ports that no longer exist on new computers. Use one of those. You may have to take something that is on 1-5, change it to 6 or 7, then change your USB dongle to one of those lower numbers.

* Make sure the USB dongle has lights on and is actually plugged into a USB port and not your network port or some other hole on the computer :roll: Don't laugh - happens all the time. Guys will put the dongle into any hole where it will fit.



EVERYTHING IS RUNNING - JUST WONT PROGRAM


* The controller must be powered off for this entire sequence.

* You must have the software running (i.e. pressed start) before plugging into the controller. This is because the controller only accepts a firmware update right at powerup. After that the controller multiplexes the programming pins out to other functions. So - if your controller is powered or the USB is plugged in before you hit program the microcontroller inside will be ignoring the programming ports. You must apply power to the controller 5V buss (via the USB) at the same moment you attempt to program. Dont worry - it is repeatable and reliable.

* If you get a message that says that the download FAILED then what is happening is the peripheral devices on your controller are sucking more current than your little USB dongle can source. Please disconnect a few items like the throttle, ebrake, cruise control, 3spd switch, etc. All those devices draw power from the +5V buss


SOFTWARE CRASHES, COMPUTER MAKES NOISE LIKE SOMETHING HAS BEEN UNPLUGGED AND THEN PLUGGED BACK IN


* In some cases (mostly with weak USB ports) the controller has so much capacitance on the board that when you plug in the USB dongle it crashes the little controller in the dongle. I remedied this by soldering a small inrush current limiting resistor of 100 ohms on your little USB dongle. Normally the dongle could handle the inrush but once it was put at the end of the long usb extension cable things got more sensitive. As a last resort you can always cut away the heat shrink I put on the dongle, unplug it from the extension cable, and plug it directly into the computer. This will fix the problem. Your new problem will be that you need to set your computer RIGHT ON the controller due to length issues... In this case you would want to procure a higher quality USB extension cable. Dont worry though - I wont leave you hanging. If you have problems with this uncle methods will take care of you.


OK, I ACTUALLY GOT IT TO PROGRAM - WHY ARE THE VOLTAGE AND CURRENT NUMBERS DIFFERENT THAN WHAT I PROGRAMMED :?

* CURRENT *


There is DC current and Phase current. The numbers you see in the software are based on a LARGER shunt resistance value. When you cut the shunt resistance value in half you are effectively tricking the controller. When you are running 100A the controller thinks it is only 50A. That being the case, we have understand the relationship between the actual current and the programmed current.

The shunt values range from around 220uOhms to 240uOhms. This will put some variance in the numbers. As a rule of thumb, the controller is going to run about double the current you set up. If you want a DC current of 60A I would suggest that you start by setting the DC current in the software to 30A.

The phase current is always a constant multiple of the DC current. As a general rule, if you cut the DC current in half, also cut the phase current in half. If you triple the DC current, triple the Phase current. Off the top of my head - I believe that the phase current is about 2.5X the DC current. This is not critical, but if you set the Phase current too low you will get results that may seem odd... For instance you may find that at low speeds you have a lower current limit than you do at high speeds. This is because the phase currents are greatest at low speeds and lowest at high speeds... we can talk about this off line if you need to.



* VOLTAGE *


If you want to utilize the LVC functionality of the controller then you will need to scale the voltages by 20%. I "fooled" the controller by changing the values of the voltage divider that it uses to read the V+ rail. The controller things that the voltage is 20% less than it actually is. If you tell the controller to set the LVC at 50V it will actually set the LVC at 60V

20% of 50V is 10V
50V + 10V = 60V

This was done to raise the maximum regen voltage and relieve an over voltage condition on the A/D input of the controller (credit to SAM-Pilot for finding that). The stock regen voltage is 75V. We wanted to have regen up to 90V

20% of 75V is 15V
75V + 15V = 90V

Here is a quick look-up table to help those who are not so math inclined.
(CHECK THESE NUMBERS - NOT TESTED)

Desired LVC Number to enter
20 V 24 V
21 V 25.2 V
22 V 26.4 V
23 V 27.6 V
24 V 28.8 V
25 V 30 V
26 V 31.2 V
27 V 32.4 V
28 V 33.6 V
29 V 34.8 V
30 V 36 V
31 V 37.2 V
32 V 38.4 V
33 V 39.6 V
34 V 40.8 V
35 V 42 V
36 V 43.2 V
37 V 44.4 V
38 V 45.6 V
39 V 46.8 V
40 V 48 V
41 V 49.2 V
42 V 50.4 V
43 V 51.6 V
44 V 52.8 V
45 V 54 V
46 V 55.2 V
47 V 56.4 V
48 V 57.6 V
49 V 58.8 V
50 V 60 V
51 V 61.2 V
52 V 62.4 V
53 V 63.6 V
54 V 64.8 V
55 V 66 V
56 V 67.2 V
57 V 68.4 V
58 V 69.6 V
59 V 70.8 V
60 V 72 V
61 V 73.2 V
62 V 74.4 V
63 V 75.6 V
64 V 76.8 V
65 V 78 V
66 V 79.2 V
67 V 80.4 V
68 V 81.6 V
69 V 82.8 V
70 V 84 V
71 V 85.2 V
72 V 86.4 V
73 V 87.6 V
74 V 88.8 V
75 V 90 V
76 V 91.2 V
77 V 92.4 V
78 V 93.6 V
79 V 94.8 V
80 V 96 V

Same rules apply for the REGEN voltages.

55V, 60V, 75V correspond to 66V, 72V, and 90V

-methods





WORK IN PROGRESS
 


WORK IN PROGRESS


FAQ


Q: Why did you limit the Regen voltage to 90V instead of setting it to 100V?

A: We dont understand how the controller handles the max regen voltage yet. We do know that there must be some degree of overshoot but we dont know how much that is. I set the max regen voltage to 90V to allow for a 10% overshoot. All the fets and caps are rated to 100V..... We have enough test data to be confident that this is safe for a non-regen 24S (100V off the charger) setup but we need more data on regen.



Q: Why did you only use 10AWG cable and not 8AWG or 4AWG?

A: Assembling and routing 8AWG cabling inside the controller is possible but presents too many difficulties for production. It is a tight fit inside the controller and 8AWG cable just makes it too hard to work with



Q: How beefy are the traces?

A: The traces come with what looks to be 13AWG stranded wire soaked in solder. I am going to call this 12AWG equivalent. I solder 12AWG solid core house wire on top of this wire. I dont soak it out, but rather I solder it in every inch or so. This acts to double the conductivity while making for good cooling. The new effective gauge is somewhere around 6AWG. This is overkill but there is room in the case and it will help to sink the heat from the fets.



Q: Is your hot wife really a Laser Physicist and did you really make her swear to "Support the Electric Revolution" in your wedding vows?

A: Hell yea she is and hell yea I did. I have it on video. I also have video of me taking her away from the church in an "Electric Ghetto Chariot". Pics added.



Q: Is this thing as "Rock Solid" as a Kelly Controller?

A: NO! This is a budget controller. Many of the parts in this controller are being pushed to the absolute max. This is the tradeoff you make to save $400. That said, I dont know anyone who has ever blown a properly modified 18 fet controller in a way that was not caused by stupidity or bad luck. For that matter - I dont know anyone who has ever blown one. We have been doing these mods on inferior 12 fet controllers for years and many (including myself and the Doc) successfully ran 100V 100A on 12 fet controllers. It is important to keep in mind that just because someone can mod a controller DOES NOT mean that they did a good job of it. There are a 1000 ways to screw something like this up and the devil is in the details. This is why I take the extra time to clean and inspect boards. Conformal coating goes a long way to protect from errant solder balls.



Q: What is the recommended operating temperature range for this unit? I've had older analog Crystalyte controllers die when operating them at both very low and very high ambient temps.

A: As of right now I will call out 0C - 38C (Freezing to 100F degrees). You get the back of the USPS truck guarantee on that. :D I am not intimately familiar with the Analog design but my gut feeling is that this new 116 based controller will be reliable to much lower temperatures.



-methods

notes: Rip DVD and post wedding video




WORK IN PROGRESS
 

Attachments

  • 002_Tania.jpg
    264.7 KB · Views: 32,116
  • 003_Tania.jpg
    262.7 KB · Views: 32,153
  • 004_Tania.jpg
    253.9 KB · Views: 32,035
  • IMG_8353.jpg
    288.3 KB · Views: 32,186
What is the recommended operating temperature range for this unit? I've had older analog Crystalyte controllers die when operating them at both very low and very high ambient temps.
 
Great Question.

We are going to have to figure that out.

I have done quite a bit of testing in 90F weather with the following equipment:

Good air flow - 45mph bike
24S Lipo 15Ah
120A current limit
Drilled 5305 in a 26" wheel
Very aggressive riding (165C winding temperatures on the motor - cant take any more)

I have ridden it hard on the hottest days and the case temp never got more than 10C over ambient.
More heat was due to "sun soak" than anything.

Now... Cold could be an issue but I dont think it will.
Most of the parts are rated down to at least 0 degrees and I am almost positive that the batteries will give out before this controller does.

Now I am in California you see... I am no Canadian!

In past threads I have provided test units to those who have the means to do extensive testing -
Do you have a Thermo-chamber and some nitrogen by any chance?

-methods

EDIT: Doc posted some hot temps back when he was trying to break the drag racing record. That was a smaller Infineon housing and he may have limited airflow due to his windscreen. When he comes around we will get him to post his numbers.

EDIT 2: Added to the FAQ. To be modified as data comes in.
 
No, thermo chamber, but I live in Chicago. Last year I was out at a party on the far south side of the city at 2 AM on the coldest night of the year, -18 F. On the ride home I had 25 mph headwinds gusting to 35. I cruised slow in the low 20s because the windchill was so fierce. I couldn't pedal to keep warm because my chain had frozen solid along with my gear shifters and brake levers, etc. Also the controller was acting funny, kind of like a bad hall sensor wire induced stutter. I've experienced this many times in really cold weather so I wasn't too alarmed. As often is the case for the cold stutter, it ran ok at full throttle so I was blipping it full on / full off in couple second pulses to maintain my desired speed.

I was wearing a full face motorcycle helmet and at about the 12 mile mark the visor had fogged/iced up enough that I could scarcely see. Stupidly, I cracked the visor open slightly to clear the fog and my eyelids both froze shut as I blinked. I got lucky and stopped in the middle of the road without crashing and thankfully there was no traffic out. It took me several minutes to thaw my eyelids/eyelashes open. And a couple more to get ready to ride. And when I finally started out again the bike hicupped and surged a couple times as I started to creep. I gave it full throttle and it died as if I had blown a fuse.

From previous inspection I knew all the ICs in the controller were the cheapest versions with the lowest temp ranges. No mil-spec parts to be found. Funny thing is each of the better spec'ed parts was only a few cents more expensive. Seemed like a really poor bargain compared to the unpleasant walk home.

Btw, when it's that cold out I take my batteries inside with me, but the bike is often parked outside.
 
Methods, just curious, how big a project would it be to build a reliable 144v 300a controller here in the USA with regen for a brushless motor (Like MarkCycles)

seems the US government is looking to give away some money for 2 seat trikes
 
todayican2 said:
Methods, just curious, how big a project would it be to build a reliable 144v 300a controller here in the USA with regen for a brushless motor (Like MarkCycles)

seems the US government is looking to give away some money for 2 seat trikes

I think the question isn't how big a project, but how expensive would it be. Have you looked at the Zilla controllers? I think they have a model that's a decent match for your purposes. But they make Methods $300 unit look like a bargain. I'm not sure of the pedigree of the Zillas though. I bet they are US designed and perhaps US assembled but the boards are likely stuffed in Asia.

Does the controller need to be built in the US? Methods is not building his here. He's modding them here.
 
Just thinking out loud. I am toying with the idea of entering the running for some of that federal money, and I thin every bit of "made, or engineered" in USA helps when a grant like this is being reviewed.

It has always been a dream (goal?) of mine to build a ridiculusly high mpg trike and be able to offer it to the public for a fair price.

This might be the way to put light e-vehicles in the hands of Joe Public
 
Hey guys,

I've seen my 18 mosfet peak at 200amps before, this controller is solid...

think about this.. with 4110's; doctorbass held 100v @ 100amps contious for one minuite pulling a school bus... and that was with a 12 fet old school crystalite.. with a excellent controller casing with great heat transfer in my opinion!

I'd like to get this controller at 150v with irfb4115 @ 75amps ... i remember running my x5304 at 133 v in a 24" and it was insanely fast!!!

-steveo
 
Ok.

I will make a prototype High Voltage version out of one of them.
Lets see...

- 160V Fets
- 160V Electrolytic
- I know nothing of the low ESR caps so they all have to be replaced
- Recalculate the Base resistor
- Much, much larger Power resistors

What else?

Say... $4 * 18 for the fets, $15 worth of caps, $10 worth of resistors, $10 odds and ends - thats $100 more in parts alone.
Then we have to talk about the labor to de-solder all old fets and solder in the new ones.
Desolder all the large and small caps and solder in the new ones
Desolder all the regulator resistors and solder in new ones

Then we have to address the bottom of the board. 1mm spacing wont cut it for 160V inductive loads. That means that ever crack and crevice needs to be painstakingly cleaned and carefully conformal coated - maybe under a vacuum. Some of those pads are VERY close after being built up with copper and solder.

I would not do that labor for less than $100

So - We are at $500 and an assload of work.

Well - maybe not :?
The reason I am modding now instead of assembling is because it is grueling work.
Building one single controller on a marathon when you are excited is one thing.
Building 30 to sell to other people (for about 1/5 what you make at your real job) is not so much fun.

But... Maybe I will try it anyway :mrgreen:

-methods
 
It does seem like a big hassle to make a 160v version. However, it might be your only chance for a weenie hub motor to ever go faster than my bike :) But then I would just re-gear my bike :)

A monster 160v controller does seem like something you're going to need to make Methy. Now that you've thought it through, you've pretty much got a moral obligation to do it. And yes, you know you're going to ruin at least 2-3 controllers before you get the bugs worked out of all the systems that can't handle 160v, or arc at 160v etc etc. Just accept the casualties of war in advance so you don't go crying and getting discouraged mid-way through and quit like a chick. ;) I'm just thump'n you're sack bro. :p


I wish you would have had these badboys available before I bought that monster Kelly that i've still never used for something like $600 if I remember right. These things are a lot lighter and more compact, and the FET setup kicks the Kelly's ass.
 
liveforphysics said:
It does seem like a big hassle to make a 160v version. However, it might be your only chance for a weenie hub motor to ever go faster than my bike :) But then I would just re-gear my bike :)

That sounds like racing words to me. Are you still at 62mph? My "weenie hub motors" have a no load speed of 960rpms at 60V, and I hit 58mph with 63V nominal, so 84V nominal with my new Methods controller should get me near 70 with enough running room.

John
 
Video please.

It is not that simple... Wind resistance goes up with cube of velocity.

It will probably take you 50% more power just to step up a few mph so if you are doing 58 mph with 60V plan on needing more like 160V to go 70mph
That means doubling or tripling your voltage - not bumping it by 20V.

Your money would be better spent on aerodynamics if what you truly want is to go fast.
There are guys PEDALING at 65mph.

After 30 mph an ebike is inefficient

After 50 mph an ebike is impracticable without aerodynamics

-methods
 
First things first. I want a real tire on the front. I haven't even tried for max speed since I went to 74v nominal (much less the 98v off the charger I'm about to try), because I really didn't like that 93kph on a cheap bike tire. I do have an aero advantage over Luke due to my much lower longer bike than a typical upright, but yeah I'd need to address aero to explore speed. I'm not really interested in speed though. It's more acceleration that I'm after with your controllers.

John
 
Reminder:


I am constantly adding pictures and information to the top of the thread.
If you like controller porn, please take the time to start from the top.

-methods
 
Just out of curiosity, what does a 160V-capable Kelly cost these days?

100V/100A will suit me just fine, for the foreseeable future. :) Any higher and I'd probably use it with one of Mark's x6 motorcycle motors.
 
If there's any way you can seat those main filter caps vertically on the board, you should try it. (yeah, I know they won't fit in the case if you do this, unless you drill holes for the caps to stick out)

Those cap leads are almost certanly only steel, plated with copper and tinned, so they don't conduct electricity *or* heat very well compared to the copper traces on the board. At high power draws in the controller, having the caps' leads a fraction of what they are now should make a difference to performance.

If you can get caps that have formed copper/alloy leads not made of steel, it'd be even better. Or at least ones with thicker plating on them, and thicker leads overall.


When I have tested my 2QD brushed controller with various cap configurations, it performs better under heavy loads with the cap vertically right on the board with the leads as short as possible than mounted as I typically see in photos of ebike controllers, sideways, parallel to the board with long leads down to the PCB.

The waveform at the controller output is smoother, a bit, using the same size cap with vertical mounting, at the higher current draws. AFAIK it's because the cap leads aren't creating as much resistance when they're shorter, so not preventing them from doing their jobs as much.

I first even looked into this because of info at the 4QD site about it, and finding that under some conditions it's actually possible to heat the cap leads enough to melt the solder holding them into the board, when they're that long (hasn't happened to me, but did at 4QD to the designer during early versions of his controllers).
 
amberwolf said:
If there's any way you can seat those main filter caps vertically on the board, you should try it. (yeah, I know they won't fit in the case if you do this, unless you drill holes for the caps to stick out)
...
I first even looked into this because of info at the 4QD site about it, and finding that under some conditions it's actually possible to heat the cap leads enough to melt the solder holding them into the board, when they're that long (hasn't happened to me, but did at 4QD to the designer during early versions of his controllers).

I suppose a quick hack solution leaving the caps in place would be to solder some fat copper wire along the leads. Kind of like how people sometimes build up the traces on the boards...
 
Doc has tried that. I dont like it for a few reasons. One reason is that it makes it so that you cant move the caps around anymore - makes them stiff which makes them brittle. It is also a lot of work and somewhat risky and difficult in the tight confines.

I suspect that it is the inductance of the long leads more than the resistance that affects performance.

Solutions include mounting smaller caps vertically...
I suppose I could mount 470uF caps vertically and then try to jam a few 1000uF in there to make up the total difference.

The reality is this is now I specified the controllers to come in so I probably wont be re-working them.

The way I see it there are ceramic caps in parallel and I would rather have 4,700uF of slightly higher ESR electrolytics than 2,000uF of lower ESR electrolitics.
I think the ceramics in parallel really help.

Good suggestion though. Maybe someone can do a test and if the results are astounding (like 5% or 10%) improvement then we can think harder about it.

-methods
 
GGoodrum said:
Just out of curiosity, what does a 160V-capable Kelly cost these days?

100V/100A will suit me just fine, for the foreseeable future. :) Any higher and I'd probably use it with one of Mark's x6 motorcycle motors.

Well - since Kelly is rated by burst phase amps (1 minute) the best equivalent to ours is probably the 250A model.
That controller will probably equate to around a 120A DC current limit. True continuous current would be around 60A

The 144V 120A DC version is $1,000

So we could still do it for about half the price of a Kelly.

It should be noted though that the Kelly requires a separate low voltage power source and it is limited to 40,000 ELECTRICAL RPM.

-methods
 
Another thing to consider is using smaller uF caps, but more of them, so that the current flow in and out of each one is less. They'll heat up less, and so will the leads. It will still matter for the leads to be as short as possible, but less than if all that current has to flow thru fewer sets of leads. :)

Of the few larger controllers I've looked at, such as my Curtis 1204, all use multiple smaller caps in parallel, mounted vertically to the board. Some glue the caps down to the board at the base of each cap, and also really fit them tight against the PCB so they can't flex from vibration. AFAICS the Curtis just has glue between caps, but there are so many that they probably damp the vibrations enough to not worry about it much, plus the whole board bottom is secured to a couple of bus bars that help prevent the PCB from flexing (along with the twin heatsinks the MOSFETs are on).

I'm fairly sure it's not a cost thing, because most of the time when I've looked for low-ESR caps to replace blown ones in various power supplies, bigger ones don't cost enough more than little ones to justify using a lot of little ones unless there is some engineering reason to do so. ;)


I'm not sure I have the capability to do the kind of testing needed to show the difference between both ways, plus my controller is just a brushed DC unit, and so would be significantly different in operation than these BLDC controllers (though I suspect not that much different in actual results between the two types for this issue).

If I do manage to be able to test this, and find anything different between the two methods, I'll post my results in a new thread specific to that. Actually, I'll start that thread now so I maybe have a reminder to myself for it. :)
 
I agree that multiple caps in parallel would lower the ESR significantly.
That would really need a new board spin though.
Too bad I dont speak Chinese or else I would suggest to them exactly that.

-methods
 
Yeah, I guess we'd need to make our own PCBs for that or something. :) It'd cost more for them to make them with multi-small caps, so it's unlikely they'd go that route since they don't actually *need* to, to deliver a functional controller.

FWIW, you could send the suggestion to them in English and there's probably *someone* there that reads it well enough to figure it out. :) Might require drawings, though.

After I get the rewound cieling-fan back from Karma (still waiting on enough money to ship it to him first), then I'll have incentive to build a BLDC controller to test it with. Maybe I should take those images posted previously of the bare board, and work on creating an Eagle layout for them that uses many smaller caps, vertically-placed. Catch with doing it that way is the board isn't probably going to fit into the same casing. I'll try to keep the same outline, but not sure if it's possible.

If I do get it done, I could test it out if I could get someone to send me a fried controller's IC chips (as long as they work) so I could transplant them to a test PCB. Well, first I'd have to succesfully make a correctly-registered-and-drilled double-sided PCB, which I'm not entirely sure I can do. :) I'm sure I have 18 FETs of the same type around here, although they're not as good as the 4110s, and I don't remember what they actually are.

This is the thread I just made for the purpose of discussing the results and testing, if any of us gets around to trying it:
http://endless-sphere.com/forums/viewtopic.php?f=2&t=13656&start=0
I also added this latest info to it.
 
Back
Top