Is my project salvageable?

madsci

1 mW
Joined
Jun 25, 2017
Messages
14
Posting in this forum despite my project not being a scooter, but it's also not really a car and the motors are basically scooter-class so I figure folks here might have the answers I need.

I'm building a small vehicle that's going to be the base for a Burning Man art car. The speed limit out there is 5 MPH and the terrain is totally flat and level for miles. There can be rough spots and small 'playa snakes' of piled-up dust, but the roads at least are always a good driving surface and that's enough for my purposes.

The goal is to haul 1-4 people and a ton of LEDs and get a range of at least 10 miles on a charge.

This project was originally going to be a motorized couch but got turned into something more car-like, about 6 feet wide and 7 feet long. Because of the couch origins I stuck with differential steering - it drives like a skid steer loader or a tank.

I'm using a pair of MY1020ZX 36V 800W gear motors, one motor driving both wheels on one side. The wheels are 13" mini ATV wheels, and I'm running an Ion Motion Control MCP266 controller.

Today was my first real drive test of the completed frame, and it started out great. Acceleration needs to be limited in software because it's a bit whiplash-inducing, and according to GPS it topped out just at 7 MPH. I'd planned the reduction ratio to get about 5.5 MPH, but picked up a bit more because the motors came with different sprockets than expected. Cornering is not the smoothest and on asphalt it's murder on the tires, but on playa dust I think that would be OK.

The problem is heat. I drove it about 1/3 of a mile in a lap around a few buildings, stopping four different times to chat with auto shop guys who stopped me to ask what the thing was. When I finished the lap I found the right motor smoking and making a very unpleasant smell.

I realize the MY1020 motors have a reputation for not being terribly rugged and I expected their lifespan would be limited, but this version at least has a fan and seems to be rated for what I need. And I figured if I made it 50 miles over the course of a week that would be a success. Only making it 500 yards is another matter.

I grabbed my thermal camera and found the right motor was at least 10 C hotter than the left. By the time I dragged the car back to my shop the right motor was still almost too hot to hold onto, but the left motor was merely really warm.

The rated current on the motors is 26.7 amps. I had the current limiting on the controller set at 40 amps per channel, thinking that I'd only see them draw that much for a few seconds during acceleration and they'd taper off to something reasonable at speed. I spent some time dragging the car around with a force gauge before I had the motors and the rolling resistance isn't bad.

The right motor was slightly damaged in shipping. The fan cover was dented in, but the fan itself turns freely and as far as I could tell the damage was only cosmetic.

I can't find anything to suggest why one motor would run significantly hotter than the other, unless maybe it was just chance that the one motor started failing first and the failure mode results in higher current and more heat. I don't know how these motors behave when they fail.

All of the wheels seem to turn just as easily and mismatches in speed ought to just make it arc (it was maybe pulling slightly left on straightaways) or at worst skid a bit.

I hooked up a laptop to monitor the motor currents and in the very short drive I made they did seem to be frequently at 40 amps, but I didn't get enough of a run on level ground to really get good numbers. I tried again at 28 amps but my testing got cut short when a software glitch caused the controller to stop responding to my steering inputs. I was going backwards and couldn't reach the shutoff switch fast enough and crashed into a curb.

Am I doing something fundamentally wrong here? Are these motors going to draw vastly more current than I was expecting? Could my low-speed maneuvering with the 40 amp current limit and the motors fighting the tires' traction to turn on asphalt have overheated the motors so quickly?

If the one motor was just a fluke I'll happily replace it and try again, but I don't want to smoke another. And if it really does take a combined 50 amps or more to keep it going, I'm not going to get more than a few miles on my 110 AH lead acid battery pack. Assuming a Peukert number of 1.3 I figure 50 amps would drain the batteries in half an hour.

I was really stoked that my vehicle accelerated so well with 500 lbs of battery, frame, and rider and even handled a steep (but short) driveway with ease. But while I don't know electric motors well, but I've been around electronics long enough to know that they don't work if you let the magic smoke out, and this motor has already lost some of its limited supply before I've really begun.

If anyone's got advice on the subject, I'm all ears. I leave 3 weeks from today and if the project is going to happen, I've got to get it figured out soon.

Thanks!
 
Various possiblities:


Physical damage to the motor could cause heating problems, depending on exactly what happened inside it.

If a magnet was broken, shards could be scattered around, causing either mechanical or magnetic interference.

If a winding was damaged in a way that caused a short within a coil, that coil could experience higher currents (lower resistance) and heat up more.

If a commutator segment or brush was damaged, arc-heating could be greater than normal. Personally I'd guess a brush was damaged or misaligned; they're basically the same as pencil leads, though a different shape, and can be damaged by mechanical shocks.





If one motor is not wound the same as the other (setup for a different RPM) then that would make their currents different at the same load/speed.

This you can test for by spinning the motor at a low voltage, without being mechanically connected to the drivetrain, and measure the RPM at that specific voltage. If it's low enough you can count the RPM by eye, marking one sprocket tooth and counting the number of times it passes the same point within a specific period of time. They should have identical RPM with no load on them, at the identical input voltage, and draw the same current. If htey don't, they're either not wound the same or something is physically wrong with one that adds drag and slows it down (which will also increase current).




The overheating could also be caused by the maneuvers, if there was more use of the righthand motor, or if power usage is not the same between the two sides. If the the terrain you were on wasn't really flat (like a crowned road), but had the overheated side on the downhill you might have had to (conciously or not) keep steering with that motor to push it back uphill, heating it up more than the other.


It could even be that the controller isn't limiting the one as much as it is the other--but that would also result in a different feel to the acceleration on that side.




As to the max speed being higher, it also means the current load at the lower speed might be higher than it would be otherwise (depends on the current limits).



As for the controller's limit being set to nearly double the max current the motors are rated for, it means that every time you accelerate from a stop, or climb enough of an incline, the motor can get more current than it should, and heat up more than it was designed for.

The motors, being made of coils of wire, which have essentially zero resistance for all practical purposes, will draw as much current as the controller can (is programmed to) supply them, whenever they are not spinning fast enough for the back-EMF (BEMF) to counter the voltage across the windings (meaning whenever they are spinning slower than they are being commanded to).


Also, the max current the motor is rated for is commonly while at it's max RPM. If you're using the motors at lower RPMs at still drawing that max current (or more) then the fans aren't spinning as fast as it was made to to draw air thru the motor, and it won't be able to cool it as it was meant to.

If you're not running the motors at max RPM and lower loads (less current) very long vs their over-current at lower RPMs, the fans also don't have much of chance to pull air thru there so heat builds up.

(assuming they're shaft-mounted fans that only spin with the motor).

You can add active-cooling to it with squirrel-cage computer case fans to run off 12v, ducting the fan so it sucks air out of the motor (more effective than blowing it in, typically). Then these fans would run all the time the system is powered on. Some of these fans have a little thermal sensor that sits in the airflow so they run at very slow speeds (taking less power and being quiet) then speed up as the air gets hotter. The thermal sensor can be moved close to or inside the motor if that's insufficent to get them to speed up enough to keep them cool (assuming they can do so at their full RPM).
 
Thanks so much. I went through a lot of that on my own today and it's good to know I'm not totally off track. That's great information about the various types of damage.

I put the car up on jack stands today and did some more testing. I wasn't sure if I trusted the current readouts on the motor controller so I scrounged up a 100V 100A rated current shunt and meter in my shop and hooked it up. At full speed, with the wheels still connected, the good motor was reading 1.8 amps. The bad one, even with its chain removed, was drawing about 15. The 1.8 reading is probably a bit low just because the meter has limited dynamic range.

I've ordered a replacement and I'll keep my fingers crossed that the motor was just damaged when I got it in the first place.

Should the brushes be accessible? I know my brushed motor-driven tools typically have ports to access the brushes for inspection and replacement, but I can't recall having seen ports like that on the MY1020 motors. Now that I've got the bad motor removed from its mount I can at least pull the fan cover off and take a peek at that part.

I was really discouraged yesterday because I saw the current graph showing 40 amps when I was at speed and not accelerating, but I think it was only the one motor showing that. I was having a hard time juggling the steering controls and laptop and hanging on to my perch. (That reminds me, I've got to get my 'safety third' patch sewn on.) The good motor was behaving like I expected it to behave.

These fans are indeed shaft-mounted and I came to the same conclusion, that they'd need to be running at or close to full speed to get any decent cooling. And I was already digging around in my junk pile for PC fans to scavenge. Hmm.. and I just realized I'll need to plan a cover over two more sections of the frame if I'm going to get airflow through there and also keep playa dust out of the chains.

Should the rated current be considered the maximum current, or is the rated current supposed to be a continuous rating and the maximum current would be different? And if the max current is different, since it's not listed in the specs and the MY1020 torque curve I have cuts off at about that point, is there a typical margin between the two?

I was making my torque estimates based on that 27-amp maximum and based on my testing so far I think that's going to be fine. I was worried that I wouldn't have enough power to get moving when loaded if the ground was a little rough, but on my first test drive it drove right over a wheel chock I forgot to remove. Now I'm more concerned about the possibility of it throwing riders off when it starts.

I have spent so much money on the wrong motors, the wrong controllers, the wrong chains, and the wrong sprockets, I'm going to have to build two or three smaller (and lighter and faster) vehicles just to make it worthwhile.

Thanks again!
 
madsci said:
At full speed, with the wheels still connected, the good motor was reading 1.8 amps. The bad one, even with its chain removed, was drawing about 15.
Yeah, a 10x reading on the bad one is pretty far gone. :/


Should the brushes be accessible?
Yes, but probably only when you take it apart...and to put it back together it'll require a bit of DIY, like a wire or a string that pulls the brushes back against their springs so you can slip them over the commutator, then pull the wire or string out of the motor just before you finish reassembling it and tightening it all down. Annoying, but a fair lot of brushed motors are made this way.



I was having a hard time juggling the steering controls and laptop and hanging on to my perch.
Time to make a mount, however temporary, to hold the laptop up where you can see it's screen. ;)

And/or get a wattmeter like the various RC hobby units (or a Cycle Analyst v2.4 standalone model with remote shunt), and use it on the battery-to-controller so you can at least keep track of the whole vehicle's power usage and realtime monitor the amps/watts/etc.



Hmm.. and I just realized I'll need to plan a cover over two more sections of the frame if I'm going to get airflow through there and also keep playa dust out of the chains.

You probably won't keep the dust out of anything, no matter what you do. But you can put a brush over the chain to keep it "dusted off". ;)

If you use any lube on teh chain, use the dry wax stuff so the dust won't have anything to stick to--if you oil it the flinty stuff will actually wear the chain faster than if you clean ALL the lube off of it completely. (especially for a short-lifespan vehicle).



Should the rated current be considered the maximum current, or is the rated current supposed to be a continuous rating and the maximum current would be different? And if the max current is different, since it's not listed in the specs and the MY1020 torque curve I have cuts off at about that point, is there a typical margin between the two?
If it doesn't say, then you could assume that rated is the continuous current. Peak current could be twice or more that amount, but for how long? Dunno--depends on frequency it's applied and how effective cooling is, starting temperature, how much temperature rise there is, etc etc.

Brushed motors don't just heat their windings, they also heat greatly at the commutator and brushes with increased current because the brush arcing gets a lot more intense and thus much hotter.

But remember that the rated current is *at the rated RPM*. If it's not spinning that fast while it's drawing that current, then it's too much current for that speed.

I don't know if it's going to be a linear relationship or if there's a curve (or the shape of the curve).

For brushless hubmotors, you can go to http://ebikes.ca/simulator and see relationships between various factors, including time to overheat, etc. Doesnt' directly show you how your specific motors behave, but might get you some insight into the issues involved.

There's a beta of v2 at http://ebikes.ca/simulator2 if you want to explore it (there's a thread about it by JustinLE here on ES that's recent and ongoing over in the motor technology section).


Now I'm more concerned about the possibility of it throwing riders off when it starts.

You can probably setup a "soft start" in the controller, or use a capacitor/resistor (RC filter) on the throttle output before it goes to the controller, to ramp up it's action (but this also ramps down it's decrease :( ).



I have spent so much money on the wrong motors, the wrong controllers, the wrong chains, and the wrong sprockets, I'm going to have to build two or three smaller (and lighter and faster) vehicles just to make it worthwhile.
EVs are addictive...they're so "easy" to build (vs ICE stuff) that it's hard to not get a bajillion ideas and want to build them all...and see a use for every electric motor you run across. :lol:

I've got everything from my big SB Cruiser long-range cargo trike down to a little "last mile" foot-scooter (which needs work but functions). I'd probably have a regular car/truck/etc EV too, but I don't feel like spending the money on all the insurance/registration/etc for those when I wouldn't use one except once in every few dozen blue moons. ;)
 
I think I'll skip trying to get to the brushes for now. I may see if I can get a replacement motor and salvage the gear head, or see if the gear head will work on the non-geared MY1020 motors I have. That will be AFTER my current deadline crunch is over, though.

Time to make a mount, however temporary, to hold the laptop up where you can see it's screen. ;)

I don't know how many more temporary mounts the frame can handle. =] Everything is held on with C-clamps, zip ties, or Velcro. I'm working on the body/platform today and I'll start firming up placement of some of those parts. Making things like the disconnect switch accessible is my primary concern.

The other problem is that I don't trust the controller with the laptop connected. It's never glitched on me once in standalone operation, but I've had it do weird things when the laptop is connected and the IonStudio program is running. The controller has the current and voltage readings I need, but I'll use something other than IonStudio to get them.

I'm an embedded systems developer and my shop is equipped specifically for that kind of project. I'd have had something rigged up yesterday but the controller's RS-232 port seems to be dead. I've got a custom board on the way that combines a couple of my commercial products and will serve as a data logger, tracking and telemetry transmitter, lighting controller, and will also serve up some web pages with live charts of the controller's readings via wifi. I have more confidence in my ability to do that stuff than to, say, get the chains tensioned properly. =] I'm still actively learning the mechanical stuff.

You probably won't keep the dust out of anything, no matter what you do. But you can put a brush over the chain to keep it "dusted off".

The last time my truck was clean was more than 8 years ago, just before its first trip out there. There's no keeping dust out of everything, it's more a question of magnitude. I was thinking of my car as a box open only on the bottom, but the side panels will hang out over the wheels and without a cover on the sides of the frame it'll throw a ton of dust in there constantly. I'll definitely get some of the wax lube.

I'll keep the current limits set to the rated current of the motors for now. I figure I can throw an override option in if it turns out I really just don't have enough power to get started and want to risk the extra strain on the motors.

You can probably setup a "soft start" in the controller

The MCP266 has an acceleration limit option but it's not in the IonStudio software yet. I sent it the commands myself but it's not responding, so I'm waiting to hear back from IonMC about that.

Finished my coffee, time to get started welding again!
 
madsci said:
I don't know how many more temporary mounts the frame can handle. =] Everything is held on with C-clamps, zip ties, or Velcro.
If there's no baling wire or duct tape yet, then you still have room for more. ;)


I'd have had something rigged up yesterday but the controller's RS-232 port seems to be dead.
If it doesn't show any levels changing on a 'scope readout, it could just be the driver chip (MAX232 is the most common one).


I've got a custom board on the way that combines a couple of my commercial products and will serve as a data logger, tracking and telemetry transmitter, lighting controller, and will also serve up some web pages with live charts of the controller's readings via wifi. I have more confidence in my ability to do that stuff than to, say, get the chains tensioned properly. =] I'm still actively learning the mechanical stuff.

I'm the other way 'round; I can somewhat understand all the software bits, but couldn't create them. Electronics hardware I have a better grasp of. But give me a pile of parts and some tools (even the wrong ones) and I can probably get something to work out, though you might be afraid of it's appearance. ;)
 
Well, I built the upper frame / deck portion today, so now there's much more space for temporarily throwing things on! This part was built with 3/4" square tubing instead of 1" like the lower frame and in retrospect I think 3/4" would have been fine for the whole thing. I always underestimate how strong the stuff is. And I overestimated how strong the plywood would be so I had to add more steel. But now it has a 6.5 x 6 foot plywood deck (plus foam padding and short pile fur) for a much more comfortable surface. I still need to figure out running boards.

I checked the RS-232 port with a mixed-signal logic analyzer, and there's zilch. The transmitter is powered up and I can see when the MCU's I/O lines all initialize and the transmitter's inverted output flips relative to the rest, so the MAX232 or equivalent is fine and it has an active MCU pin connected.

New motor just shipped from Australia. I should be back on track by the weekend, I hope.
 
madsci said:
Well, I built the upper frame / deck portion today, so now there's much more space for temporarily throwing things on! This part was built with 3/4" square tubing instead of 1" like the lower frame and in retrospect I think 3/4" would have been fine for the whole thing. I always underestimate how strong the stuff is. And I overestimated how strong the plywood would be so I had to add more steel. But now it has a 6.5 x 6 foot plywood deck (plus foam padding and short pile fur) for a much more comfortable surface. I still need to figure out running boards.
When I built the SB Cruiser trike I actually used a bunch of smaller tubing (1/2", IIRC) for a number of things, and it ended up being a lot twistier than I wanted. I stiffened it up when I rebuilt it (for other reasons) using more of the 1" version in place of most of the 1/2", and a different construction method.

But I think that if I used the newer design but the narrower tubing, it might still have ended up as stiff as needed while being just a bit lighter. (battery & motors and rider and cargo on it are so heavy that the frame weight isn't really that big a deal).

Since I'm using recycled materials, as available from wherever they can be gotten free, I don't always have the choice of exactly how to build things--sometimes that's at least partly dictated by the materials themselves. It's more challenging...but it is somtimes more fun--and sometimes frustrating. ;)


Regarding the plywood: Wood bolted to a frame (or pressed into the spaces inside it) can make a stiffer structure than either material by itself. This is part of what makes SB Cruiser stiffer now; the big cargo box under the seat is very flexible (no triangulation) as just a frame. But with the planks installed into it, filling up all the space between the tubes on the box faces, it is very stiff. Just a box of wood would be flexible, too.


Plywood is durable, but not particularly stiff, unless it's secured to something else. Boards are often stiffer depending on type and grain and orientation vs stresses, but even those are stiffer connected to something else (or a frame of the same stuff).

There's usually a balance point between stiffness and weight but it can take some iterations to figure out a design that gives that. Can be done on paper (or computer) but I usually end up doing it by experimentation. :oops:



I checked the RS-232 port with a mixed-signal logic analyzer, and there's zilch. The transmitter is powered up and I can see when the MCU's I/O lines all initialize and the transmitter's inverted output flips relative to the rest, so the MAX232 or equivalent is fine and it has an active MCU pin connected.
Then that's weird that it doesn't respond to the port (unless it's just the input pin that's toast).

I suppose it's possible that if it's a regular female D-sub connector, it could have a barrel that's more open than it should be, causing poor or intermittent contact with the male pin of hte mating connnector.

Or there could be a problem with the port on the communicating device (mechanical, electrical or software), or the cable from it to the controller?



New motor just shipped from Australia. I should be back on track by the weekend, I hope.

Maybe next time investigate using the powerchair motors with gearboxes? They're not light but in context with the rest of the vehicle that's irrelevant. ;) You could even parallel their shafts on the same chain to increase torque by multiples of the number of motors, and divide the power dissipation of each motor the same way. (use two or more motors/gearboxes on each side).

See my old Crazybike2's original drivetrain for an example of using a powerchair motor with a chaindrive. (it had enough torque with just one to destroy a number of chains and chainrings due to derailments from frame flex under torque loads).

Or there's lots of other possibilities for inexpensive motors; most of them brushed but the brushless ones would be more efficient.


If you can find them there are brushless powerchair motors also made for high torque / low speed, I have a thread around here about one of the Invacare ones, that I'd still like to get used in a middrive on something.


If there's a medical-equipment repair place near you, make friends with someone there that likes interesting projects, and you might just get free access to otherwise-tossed-out motors, breakers, wiring, etc. ;)
 
I figured out the RS-232 problem this morning. I'm blaming this one on the documentation. The transmit data pin on the controller is marked RX0. I've seen plenty of swapped TX/RX pins and the reason I set up a script on the controller was to check for that and see if I got anything out on either pin.

Well, turns out the port numbers are listed in three different places in the manual. Two of them agree, one doesn't. The one that doesn't agree was right. My test script was sending data to a disabled port. Now I can talk to it on the RS-232 port, but the set acceleration commands still don't work, and in fact they temporarily stop processing of data on the port, which is supposed to have a 1/10 second command timeout. And the controller responds to some commands before the checksum has been sent, so contrary to the docs it's not always checking the checksum.

If nothing else, this project has done a good job of reminding me what it's like to be on the other side of the table. I'm actively working to get my own docs up to date and old bugs squared away.

I rarely have a steady enough stream of recycled stuff available to do much with, but I should probably explore the local scrap yards more. I just picked up about 30 feet of 5/8" tubing this morning and paid $11, new remnants since I didn't have a big enough order to get delivery and had to fit it in my truck.

I'm trying to keep my deck / upper frame basically independent of the powered lower part with the motors and batteries - it just bolts on. The idea is that I ought to be able to reuse the chassis when I have a different idea for the rest of it. I don't know if you've seen the variety of vehicles out there, but it's kind of nuts. [youtube]_A8OTGapssc[/youtube]

Two of us got on the deck and jumped on it this morning and it seems like it's going to be fine. Lots of diagonal bracing and it's plenty rigid for carrying four people. I designed the lower frame in CAD first, but for this part I just started clamping and welding. Easier to work hands on sometimes.

I'll definitely have to learn more about powerchair motors and the like. I decided early on that I'd use 36v since I had chargers already and easy access to batteries and I think then I started looking for the most powerful brushed 36v motors I could find, and I think that took me down the wrong path. I'm also seeing now that I probably don't need as much power as I thought I would, and I almost certainly wouldn't if I had a more efficient steering mechanism. Tank drive makes automation easy, though. I can write a simple GPS autopilot that will navigate me through a series of waypoints. No collision avoidance, but I can work on that later. There are no rules against remotely-operated or autonomous vehicles out there YET...

About the squirrel cage blowers... do you mean to attach the big flat intake side to the back of the motor? Or to hook up a duct like in the thread at https://endless-sphere.com/forums/viewtopic.php?t=119 and reverse the motor? Never tried running a squirrel cage blower that way so I'm not sure if that works right.
 
madsci said:
Well, turns out the port numbers are listed in three different places in the manual. Two of them agree, one doesn't. The one that doesn't agree was right.
Given the stuff I've seen in beta testing music-creation software (mostly by Cakewalk) over the decade or so I used to do that (more than ten years ago now, I think), and "features" that either never got documented, or are in the documentation but dont' exist in the software (or don't work anything like it says they do), I'm not really surprised. :/

(we'll just leave out entirely the discussion about features that simply don't work at all or crash all the time, but were touted as a huge part of the software and were not only never fixed, but completely ignored ever afterwards)



And the controller responds to some commands before the checksum has been sent, so contrary to the docs it's not always checking the checksum.
Now *that* seems odd, cuz it could (theoretically) get corrupt data that accidentally manages to be a command but isn't what was sent, and a wrong operation performed (taht wouldn't be if it checked first..."simon says").


I rarely have a steady enough stream of recycled stuff available to do much with, but I should probably explore the local scrap yards more. I just picked up about 30 feet of 5/8" tubing this morning and paid $11, new remnants since I didn't have a big enough order to get delivery and had to fit it in my truck.
That seems like a pretty good price. I get most of my steel these days from the retail store job I work at, when we throw away old signage and fixtures. Most of it is crappy steel, and a lot heavier than if it were good chromoly, but it's free. :) If I had to buy the stuff I'd never make most of the things I do, or experiment much; it's just not in the budget.

Presently I actually have about enough 1" square tubing and other bits and bike frames to build another complete SB Cruser-style trike, other than the wood for it (I save pallets too, sometimes, when the wood is nice, and light, and they are not the standard 4'x4' size that we send back on the delivery trucks--they'd just be tossed out otherwise). I dont' have any of the 1.5"-2" tubing for the "downtube" and "keel", but I expect I could double up the smaller stuff to make up for it if I had to, or triangulate more.

And I have a bunch of "wire rack" type stuff, originally to be used for baskets and stuff but I went with boxes as they secure better and are actually lighter than baskets made of this heavy stuff, so they're probably going to become shelves for stationary storage instead (and fence sections to keep the St. Bernards out of certain things in the yard).

I've gotten a bunch of galvanized conduit off of Freecycle.org before, and other usable stuff like that; even a couple of treadmills that had a lot of steel and electrical/motor stuff I could use (and bearings, etc).

So if you have time to look around, you might find sources for recycled materials that cost only the trip to get them. :)



I don't know if you've seen the variety of vehicles out there, but it's kind of nuts.
Yeah, I've had a few acquantances that go there off and on. If I could ever swing the money and time to go myself, It'd be interesting (but I'd have to bring my dogs).




I'll definitely have to learn more about powerchair motors and the like. I decided early on that I'd use 36v since I had chargers already and easy access to batteries and I think then I started looking for the most powerful brushed 36v motors I could find, and I think that took me down the wrong path.
You can use PC motors on 36v, too; there will just be more brush/comm heating as the arcing is more intense at higher voltages. But torque will be higher due to higher currents (all else the same), so that's a plus. :)


There are no rules against remotely-operated or autonomous vehicles out there YET...
Not till somebody gets run over by one. ;)

About the squirrel cage blowers... do you mean to attach the big flat intake side to the back of the motor? Or to hook up a duct like in the thread at https://endless-sphere.com/forums/viewtopic.php?t=119 and reverse the motor? Never tried running a squirrel cage blower that way so I'm not sure if that works right.
They don't really work properly in reverse, as they operate by centrifugal force throwing the air out from the center, and have an outer duct shaped to guide that air out.

So to use them on a motor to suck air out, you'd put them flat against the end of the motor, intake on the motor's vents. THen the outlet duct from the fan just blows air away. I wouldn't duct that any further if I didn't have to, as any additional ducting means more eddy currents in the air and so less airflow (and more noise).

If you want to push air thru the motor, you'd have to make a duct from the outlet of the fan to the motor vents, and that will probably be noisier with less airflow, and it'll also take up more space (since the motor wont' just be flat against the end of the motor can).

But you can experiment if you have time and see which works better for your situation.
 
Apparently (undocumented features again) read-only commands don't need a checksum, just the ones that write settings. Still seems unwise to me since the host won't know if a command was misheard and misinterpreted.

I checked and it was actually 36 feet I got, so I think that's a decent price compared to their catalog price. For the 1" stuff I was paying 60 cents/foot. It's really nice to have a forklift and enough yard space to unload 20' sections and keep them intact.

I grew up salvaging anything and everything electronic that I could. I still have a hard time throwing out anything that I know has usable components, even when it's not worth my time to try to recover them. My neighbors down the street were a little older than me and knew all the places to find salvage, AND had a car when the oldest turned 16. I just didn't have the same exposure with mechanical stuff, but I've certainly built a bunch of stuff with reclaimed parts.

Now I just get irritated if I come up with a design I like, but then can't replicate it because it depended on some piece I just happened to have on hand and buying a new one would be too expensive.

I did pick up a few hundred pounds of random scrap steel when I started learning to weld, though. A friend of a friend worked at a scrap yard and drove over with a container full. He weighed it before he left, I took out what I wanted, he weighed the remainder, and charged me something like 10 cents a pound. Not part of their normal service, though.

I didn't think the squirrel cage fans would work right in reverse, but I'd never seen one set up with the intake connected to something. Hmm, I take that back - I did some product testing in a wind tunnel a few months ago and I think the wind tunnel was set up like that, but on a vastly larger scale.

I ordered two blowers today. I'll probably use duct tape to stick them on for now.

I'm glad now I kept the chassis so flat and the deck separate. My transportation plans seem to be falling apart. I'll probably need to disassemble the car to stack it up in a box truck. The chassis it a little more than 6 inches thick with the wheels off and the deck frame just has the four supports sticking up and is otherwise flat. Everything else can fit in the trunk of a car.
 
Thought I'd post a quick update. Got a replacement motor, wound up swapping out the gear head with the dead motor because the old one seemed to run smoother and quieter, and I put squirrel cage blowers on both motors. I used 4" PVC vent caps as mounts and epoxied the fans on, then added some foil tape because the epoxy didn't seem to take well.

The fans are working great and so far the motors have never gotten more than slightly warm. I've kept the current limits a bit under the rated current for the motors, too.

I did a test run with four riders today (well, about 4.6 I guess, not sure how far along Julia is) and it has plenty of power and acceleration in a straight line. Turns are problematic. The differential steering is really giving me trouble with more than one rider on asphalt. On smooth concrete, like in my shop, it spins like crazy.

That's my biggest mechanical concern now. I figure I'm not going to worry about it too much right now. I can arc slowly, and that's good enough for the open playa and following circumferential streets. Worst case, I kick everyone off to make sharp turns. But the playa is dusty and turning should be easier there, and if I have to I'll push the current limits up. I've also got new sprockets coming in that will take my top speed back down to 5 MPH and gain me a bit of torque.

I'm also going to bring the 36t sprockets I ended up not using on a prior iteration of the design. I can put those on and I'll be at a bit under 3 MPH max but I'll gain 40 lbs of force on each side.

If it survives the week, next year I think I'll add a third set of wheels and try to set the middle wheels a tiny bit lower than the others to make it pivot easier. That will also improve my weight distribution, which is a concern because the deck ended up so wide that it'll probably invite 6-8 people to sit on the car. I'm going to have to strategically place decorations to limit seating spots for now.
 
Okay, between this and your other thread https://endless-sphere.com/forums/viewtopic.php?f=34&t=89116&p=1299348#p1299348 I feel like I'm far short of enough information to really start saying this is it. But I know those little motors starting with MY, messed with them quite a bit. If they are they 36v motors you say they are, then they want to run right around 2,600rpm at 36v, if you run them slower their language for expressing their displeasure is thermatic. HEAT!

So, at 2,600rpm turning that 12t pinion the 48t sprocket will be turning at 650rpm, a 1:4 reduction. That means that if you measure the driven wheel that sprocket is attached to around the outside, you get the distance of a single revolution, which you then multiply by 650. You did that twice, 1:16, a little over 160rpm. So a 10 inch wheel, multiplied by Pi, (3.141592) is 31.4 inches around, you'd want to turn it at around 140 rpm to go 5 miles per hour. This is ONLY true if I properly understand what you've put together, but your 13 wheels would be needing to turn at an rpm to go just 440 feet per minute, right? They will go 3.5 feet per revolution you would want 125 revolutions with your gear ratio, while it would need to turn over 160 to keep from overheating.

The smallest pinion I'm familiar with that should fit your motor is 9t. That would take one quarter of the revolutions per minute off the wheel spin when the motor is at 2,600rpm. ONLY IF I'M COVERING ALL THE BASES does this solve your problem. My understanding is when you try to run around at Burning man with the motor under 2,000rpm for whatever extended periods it'll do some serious getting hot. https://www.electricscooterparts.com/bicyclesprockets.html

Dang, building buggies for Burning Man sounds fun.
 
Okay, thinking about this today, it occurs to me that your 13" atv wheel will have a diameter closer to 20", so instead of 3.5 feet per revolution it'll be 5 feet or more. Without the real measurement its too hard to say, but you might want to get your gear ratio up around 30:1. I'm used to thinking in bike wheels where 13" would mean just that.

If you measured the circle of your tire and it was exactly 5 feet you would want it to turn 88 tines a minute to go 440 feet. If you swapped BOTH those 12t pinions for 9t you'd have a 28.444 ratio, which would work fairly well if the wheel wasn't over 5 feet. Another possibility is a 90t sprocket in place of one of the 48t.

This doesn't become reliable without better info.
 
Dauntless said:
Okay, thinking about this today, it occurs to me that your 13" atv wheel will have a diameter closer to 20", so instead of 3.5 feet per revolution it'll be 5 feet or more. Without the real measurement its too hard to say, but you might want to get your gear ratio up around 30:1. I'm used to thinking in bike wheels where 13" would mean just that.

I'll run through what I've got in my spreadsheet. The motors have a 6:1 reduction planetary gearbox built in, so I'm starting with 500 RPM at the motor shaft and 9.2 lbf*ft of torque for each motor. 11t motor sprockets to 18t jackshaft sprockets, to 12t to 24t on the axles. That's a 3.27:1 reduction overall (excluding the gearbox) and 152.78 RPM at the wheels.

To get speed, I'm taking (diameter * pi / (12 * 5280)) * 60, which gives me a top speed of 5.91 MPH at 500 RPM and 55.59 lbf at the wheels on each side.

Reality seems to match up pretty well to my calculations. My GPS receiver showed me briefly hitting 7 MPH with a light load. I tried to switch to 21t sprockets last night in place of the 18t ones, but the spacing wasn't coming out and I don't have time right now to mess with the motor mounts to get the chain tension right. Next time I'll design in more adjustment range.

Dauntless said:
Another possibility is a 90t sprocket in place of one of the 48t.

Not going to work with #40 roller chain. 80t sprockets are 13" in diameter and $100 each at McMaster-Carr. I've already got $160 of 36t sprockets I ended up not using. (And that was from a much cheaper place than McMaster-Carr.)

The blowers seem to be doing a great job of keeping the motors cool. I'm going to try to rig up a control for them so that as soon as the current hits maybe 10 amps they switch on full blast for at least 15 seconds, then drop down to a quieter level at normal cruising load, with another 20 or 30 seconds of cool-down after it stops. That way I'll have full cooling at 0 RPM, where I need it most. Just wish I could get a direct reading of the winding temperatures. I'm afraid a short, hard push like turning with high traction will overheat a winding faster than I can react and it'll burn off insulation and short out like the last motor did.

I keep telling myself it'll all be worth it if this works out. I've been to Burning Man 7 times and I've only spent maybe an hour on art cars in all that time. Having one of my own ought to make my week a very different experience.
 
Thanks to everyone who answered my questions! The project worked out great. Steering was indeed challenging with a heavy load, but the car performed well at Burning Man and carried up to 12 people at a time. Most cruises were 6-8 people. I never found out the maximum range, but it had no trouble taking 8 people on a several mile trip around the playa at 5 MPH.

[youtube]JQDmBcVZkUQ[/youtube]
 
Good to know weren't the title character. This year that's not much of a joke.

So did you have booming music? Did you chase anyone? I meant is that you in the video?
 
i watched the burn (this video was taken just as we were heading out there) but didn't hear about the guy who ran in until hours later.

I'm on the car driving in this video, but I can't say for sure where I was sitting. Maybe on the front left. No booming music on my car - just a cheap car stereo that worked fine when we were far away from everything, but it couldn't begin to compete with the sound systems at the big sound camps or on the larger art cars.

The remote control was great. I'd have the car quietly stalk people, or take off when they sat down on it to rest. I had it backing away from a guy every time he'd take a step toward it and he ran and leaped on, and I rolled him back off again by reversing fast.
 
Back
Top