Mongoose Blackcomb

[Barncat]

motor detection results with the updated tool console were identical to the old version, so that's very disappointing. may have something to do with this HI200 knock-off vesc running on old firmware or whatever... [edit- yes, apparently it's not the tool but getting the latest fimware on the vesc itself. still trying to figure that out...]

glad i wrote down my settings from the other day. will just reset all manually and try to optimize as best i can.

 

[owhite]

glad i wrote down my settings from the other day.

First I'd like to salute your heroic efforts to get a vesc to work with a new BLDC motor -- I know from personal experience how many years that can take off of one's life span.

Second, let me quickly note that in the version of VESC-tool that I run there's a menu item File-->Save App/Motor configuration that allows you to save all your motor and app settings to disk, and then you can reload them during another session. This has worked flawlessly when I've blown one of my BESCs, but was still able to power it up to save the configs for use with another BESC.

Also, I apologize if you know this already but use this tab to get to the menu to load up your new firmware.

cheers.

 

[Barncat]

hey owhite- thanks for the tips. yes i was on the correct tab in the newer tool version but couldn't see how to pull up firmware version 5.3. i winged it and clicked on a tab that apparently loaded 5.3 and then ran motor detection. screen said it was successful but it spit out some weird numbers, and even after manually overriding some values the motor is WAY hosed up. not damaged or anything just not responding normally. at the moment i'm in way over my head and am very sorry i attempted this "upgrade".

so i don't know if:

1) the HI200 can even handle the 5.3 firmware
2) whether i've damaged the vesc by loading it
3) whether i can backtrack and load the original firmware

there does not appear to be any menu to select from in the "firmware" tab, and i can't see how to find what i believe was v5.2 that it came with.......................... i can reset everything from my notes if i can reload 5.2

 

[owhite]

I'm really sorry it's so annoying.

So I have an outdated version of VESC-tool -- ignore that -- but what happens when you go here in the menu?

DEVELOPER --> Load firmware configs

 

[Barncat]

well, long story short, i was saved by my laptop. i figured hey, maybe the original vesc tool with the v5.2 firmware is still in the computer files and not erased by the new downloads. lo and behold i found and fired that back up and loaded the 5.2 back on the HI200. i manually reset the detection values from a few posts ago and it runs great again. no damage. still need to figure out why it's clipping a bit at max throttle, but am only going to tweak the current setup with no major changes. it seems that the HI200 doesn't like v5.3 firmware. i'll leave that to the experts.

so regen, better forks, and a couple other details to go. will post some final pics and observations soon.

 

[Barncat]

OH. so that's how you get to them, from the "developer" tab on top then "load firmware configs". that wasn't obvious to the uninitiated- many thanks for future reference. as you can see i managed to extricate myself from my predicament and i'm glad the older stuff reloaded without incident.

 

[owhite]

so that's how you get to them, from the "developer" tab on top then "load firmware configs".

yeah, sadly I was screwing with VESC-tool for months before I even discovered there was a menu.

 

[Barncat]

so close, yet so far... two problems left.

1) the vesc tool wouldn't even recognize the thumb throttle i put on the ADC2 wire for regen control. the +5v and GND were paralleled with the throttle wires on the COMM port. main throttle is working fine but it won't map the thumb. brand new- may be defective. will cut into the wires tomorrow and test for signal and voltage.................

2) the system cutout i'm getting is throwing red lights for about one second. it's actually happening on deceleration. on the test stand under no load, motor instantly winds up to max no problem but then lights up about half way or so back to zero, with a minor but audible clunk coming out of the motor- and it's related to the RATE of decel- if i back off slowly it doesn't throw the light. any ideas from the pros on this one?

all current and voltage settings are within spec for the HI200 and battery pack, and testing wider A and V ranges had no effect on it. [EDIT- maybe, i'll be trying more of that today...]

 

[owhite]

I dont know about item 2, but for the first one I can offer that:

• I've gotten screwed up on wiring of throttles. I think the thumb throttles are based on hall sensors. I had one that was based on a hall sensor which meant I could not figure out the correct connections using a voltmeter. I connected it wrong and it blew the sensor.

• What I find to be more reliable to test the VESC is to use a potentiometer. If have one or want help with wiring let me know. To order on amazon search for: "10k potentiometer".

• TBH I would not worry about regen until you're happy with the rest of the system. (He said, rendering his other points to be moot).

 

[Barncat]

these common 3 wire analog throttles just need +5v, GND, and someplace to attach the third signal wire. main throttle goes to ADC1 vesc wire, and regen goes to ADC2 wire. was shocked, pun intended, that it didn't work, which leads me to believe it's defective right out of the box. testing shortly... other program settings must of course be done during that process but vesc must see and map the throttle first.

will experiment with other settings today re the red light.

 

[owhite]

hese common 3 wire analog throttles just need +5v, GND, and someplace to attach the third signal wire. main throttle goes to ADC1 vesc wire

agree but in my case the wires werent the traditional white, black, red, and I couldnt figure distinguish which was which.

 

[Barncat]

the several i've seen thus far are either red/black/white signal, or red/black/green signal. the brand new thumb throttle from yesterday tested as defective, so that's rare. had 5v in the circuit but no voltage control or change on the white wire.

the very good news however is that the vesc red light on decel was due to Switching Frequency, found in the FOC Advanced tab. doubled it from 30khz to 60khz, and below that on the same page at Current Controller Decoupling i selected "FOC CC Decoupling Disabled".

this has finally unlocked the performance and torque this build is capable of. no joke. once i get all the amp settings safely maxed out i'll be hitting close to 60mph running downwind. on a 47 lb bike.

updates to follow.

 

[owhite]

Congratulations on your progress. When you have a chance could you shout out: number of teeth on front gear, back gear, and size of back tire? Also, you ended up with the Tiger Motor U15 80kv, correct? it looks like it is either backordered or no longer available at a lot of sites, but still seems to be available on aliexpress if you search under "U15 KV80 T-MOTOR". Same product?

 

[Barncat]

16T, 80T, 20 inch, U15 80kv bought last year on ebay. they're very pricey new... there's almost nothing to them so i'm surprised they retail for what they do. seems like you'd have to really beat on it to overheat it- it's ALL ventilation.

so my 15s battery will safely do 225A bursts, the HI200 is rated for 300A bursts, and the motor up to 24s 180A continuous.

in the vesc tool, i believe Motor Current Max = nominal phase current. can anyone confirm this?

so i'm thinking i can 'safely' boost the motor current max to 300A, and set battery max at 225A. what then, based on those with considerable vesc experience, would be the recommended Absolute Max Current setting?

and i assume the HI200 is the weak link in the chain. it has a built in heat sink and is bolted to a roughly 5" x 5" x 1/8" aluminum plate- which i may swap for 1/4" plate. also- i would never go WOT for more than 10 or at most 15 seconds.

 

[owhite]

what then, based on those with considerable vesc experience, would be the recommended Absolute Max Current setting?

I wouldnt call my experience "considerable", but based on recommendations I've read on other blogs I've got motor max current at 200A, and absolute max current at 400A. I NEVER get close to that amount of absolute max current. It's more like 80 on a hardcore hillclimb, and I have a 150A fuse on the plus side of input to my controller.

 

[mxlemming]

16T, 80T, 20 inch, U15 80kv bought last year on ebay. they're very pricey new... there's almost nothing to them so i'm surprised they retail for what they do. seems like you'd have to really beat on it to overheat it- it's ALL ventilation.

Could you measure the stator in it?

Diameter and thickness of the lamination stack.

Max torque is given by diameter squared x stack height.

I have a motor i think is similar size, and a battery of similar size, but due to months of rain and dark not really had a chance to properly ride it. Wondering what to expect.

Regarding the currents.

1) there's no gains to be had beyond saturation of the motor. It simply won't produce more torque.

2) the abs max current is there to protect against loss of control, it's not a parameter that is "used" it's there as an absolute limit where you say "if it's exceeded, just turn off to avoid fire". So it should be set at a level enough above the motor current setpoint that it doesn't accidentally trip but well within the limits that would burn the controller, motor, wires or battery (or fuse?).

If you're running 200A and the controller sees 300A, unless it shuts down, the next measurement will probably be 500A. At this point it's lost control.

I'd set it at 1.5x the running current, and definitely no more than what the hardware can read (about 480A) under any circumstances.

 

[Barncat]

hey guys- my last test ride settings (it's been too cold even in FL to ride this weekend...) are 200A motor, 300A absolute (yes- a safety valve), and just 120A battery max. i'll likely edge all of those up a bit, and i've noted your respective recommendations, thanks.

running real time graphics in vesc tool wide open with no load it was pulling just 30A or so battery and 90ish phase as i recall, so without a dyno of some sort to load the rear wheel you can't tune it accurately on the stand. probably not worth bothering with trying to make one.

the diameter of the stator is about 135-137mm. hard to be exact. stator teeth appear to be about 17mm tall x 20mm wide. i'll go look online for a drawing. there are some published torque #'s out there somewhere too, and i'm at 5:1 reduction. AFAIK, i'm the only guy to put this motor on a bike; they're always used in drones...

 

[owhite]

Oh that reminds me -- you'll probably want a speedometer. I built one that uses the serial out of my BESC that measures mph, volts of the battery, amps going to the controller, and temperature. It should connect to your HI200 with a different connector no problem.

It's made from a custom PCB, a microcontroller called a teensy, a adafruit 14 segment display, a 3d printer, and some laser cut stainless steel.

If youre interested I can get you one at cost, or if you want to make your own from scratch I can walk you through it.

 

[Barncat]

nice job on your gauge unit, and thanks for the offer. i see you know your way around electronics and computers based on your BESC thread. all i'm going to put on the bike is a V indicator "gas gauge".

i have a set of new 120mm Chinese air forks in the mail, so will see how those work. as mentioned earlier in the thread- for street use i suspect they'll be fine but i wouldn't thrash them over jumps off road. the OEM forks on the bike are junk. more updates to follow.

 

[Barncat]

i installed the new forks and they are a remarkable upgrade, especially for the price. bumps? what bumps . we'll see how they hold up.

i also got the regen working and i'm very pleasantly surprised at the effective braking force. since this setup has no freewheel, regen is both possible and really a must for urban riding. why waste all that braking energy as heat? thing rides like a Tesla now, which i've driven.

will post some pics and details tomorrow.

 

[Barncat]

Yesterday was very cold and rainy in the Tampa area and today was still chilly but sunny and dry, so i put in some test miles after getting in ten miles on the rollerblades.

Still a preliminary thumbs up on the 120mm Chinese air forks that are ubiquitous on ebay. Since they're post mount for disc brakes and the old ones were IS, I also picked up a complete set of two mechanical calipers and 160mm discs on amazon for like 20 bucks. Unbelievable. They work great. And are stylish red.

As for regen: I had max motor brake current at -100A and max battery brake current at -50A for the first test then reduced respective settings to -75/-45. I'll be going back to my first guess as the latter was a bit weak. Bursts of backfeed in to battery pack can be 2.5C without much concern I suppose.

There are two ADC screens in vesc tool where you have to set "control type". I used "current reverse ADC2 brake button".

The kickstand was rendered moot by the taller longer travel fork so I removed it. So a pound less there and close to two pounds less with the fork swap. At somewhere around 46-48lbs, this may be the lightest full suspension ebike in the world that'll break 55mph running downwind on the flat.

The FOC vesc format allows for VERY smooth power delivery and regen that you can tailor easily- once it's all working correctly. If you're like me and not an EE you'll have a steep learning curve. Kudos to BVedder and all you smart guys out there designing this stuff in the public domain.

Speaking of design- I'm finally going to reveal the Barncat Battery. No weld, no solder, no BMS, simple, rugged, efficient, quick to assemble/DISASSEMBLE, two different packs battle tested for 2000 miles, looks way cool. I'll be starting a thread in the battery section...

 

[owhite]

Great pictures and lots of nice touches to your overall build. Congratulations!

So if you did it all over again would you use a HI200? I mentioned you over here in a discussion about controllers.

 

[SlowCo]

Speaking of design- I'm finally going to reveal the Barncat Battery. No weld, no solder, no BMS, simple, rugged, efficient, quick to assemble/DISASSEMBLE, two different packs battle tested for 2000 miles, looks way cool. I'll be starting a thread in the battery section...

Great build and I'm very curious about your battery build. Please put a link in this thread to that battery build thread so I don't miss it.
Enjoy riding you bike :thumb:

 

[Barncat]

Thanks guys 8)

So far I can't think of anything bad to say about the HI200 other than they presumably copied most of the tech. It's not enclosed like the official version- so it must be handled and mounted carefully, but it also costs 2.5x less for similar function. Such is the state of global commerce. Good luck beating China on price... I see there's a big flap on the Flipsky 75100 thread about this sort of thing.

Other than that, I'd have gone with a 20s battery build but last year highV escs didn't really exist save for hobby builds AFAIK. And I still don't know what half of vesc tool parameters mean so I faked my way through it...

It's pretty cool that 12 lbs of battery, 4lbs of motor, and a few ounces of circuitry will propel a bike to these speeds, silently, and with no point source pollution. And unless my calculations are off it only costs about a dime to go 30 miles. We need more vescs and high power compact motors. Faster better cheaper.... it's the golden age of microbrew and ebikes. Cheers!

 

[mxlemming]

The kickstand was rendered moot by the taller longer travel fork so I removed it. So a pound less there and close to two pounds less with the fork swap. At somewhere around 46-48lbs, this may be the lightest full suspension ebike in the world that'll break 55mph running downwind on the flat

O rly... :wink:

But yeah. Seriously cool. I've loved this build since the beginning, and it's now looking fantastic. I'm looking forward to summer. 2022 is going to be a good year for ebike.

I'm a fan of the hi200 and flipsky boards personally. I like the idea of supporting Vedder but 500 bucks for a controller that's really only good for 200A at 16s is pretty steep IMO. I've built my own boards that do that spec and they cost me about 60$each.

I sometimes feel bad for having bought a few flipsky controllers but never a "genuine"VESC because I know just how much effort Benjamin has to put into it (massive massive...) But at the end of the day I can't afford to be risking burning 500$ if i set it up wrong.