The SB Cruiser : Amberwolf's 2WD Heavy Cargo Trike & Dog Carrier

Wow, what an ordeal!

I was wondering why we hadn't heard from you.

I have a suggestion for any future experience that's similar to what you went through in solving at least ONE of the great many problems you had to solve. In particular, when you've got the problem you described in this image:

file.php


...I notice that you've got PLENTY of slot below the wheel to work with. So, there's a super easy solution that only requires you have a few bits of junk laying around ( :wink: ) ... a piece of short, thin, and not too narrow flat bar stock and a bolt with nut (and 2 washers would help, one flat, one split) that fits inside the slot... Imagine, if you will, that the flat piece has TWO holes in it, one for the axle, one for the bolt! Then, you slide the flat bar over the axle end (or put it on the inboard side if desired) and bolt the flat bar to the existing drop-out with the bolt below the axle, flat washer against the drop out, lock washer on the flat bar side (with the nut)... You can even take up some gap in the width between dropouts this way, if needed! 8) The split washer will keep it from coming loose and the flat bar positively locates the end of the axle - it can now only move axially! :lol: The flat bar can be damned small, only big enough to get a full face on the washers and enough to reach around the axle end...

HTH, just food for thought for the future...
 
RTIII said:
Wow, what an ordeal!
Yeah, such a simple thing turned into quite the PITA. :/ That sort of thing happens to me regularly, partly because all my stuff and tools are quite disorganized.

I never have the time to organize them *and* do the stuff I need to do, though all the time wasted looking for things would easily have gone to organization if I had the time to do it all at once. Piecemeal organization doesnt' work; I end up having to move stuff I've partly organized but not put in it's final place and then stuff gets mixed up, etc etc. By the time I have more time (and energy) to do organizing, most of the previous stuff isn't any more.

If I could get a month off work (other than for caring for a sick dog like I did with Tiny), during one of the cooler but not cold parts of the year....



Imagine, if you will, that the flat piece has TWO holes in it, one for the axle, one for the bolt! Then, you slide the flat bar over the axle end (or put it on the inboard side if desired) and bolt the flat bar to the existing drop-out with the bolt below the axle, flat washer against the drop out, lock washer on the flat bar side (with the nut)...

If I understand correctly, it's essentially the same type of thing used to secure an old-style derailer into the root of the dropout slot, except it would go *after* the axle and not before it in the slot.


Earlier in this thread, I did actually do something very like this at one time, because I had a problem with the axle nut on the motor being stripped and so unable to securely clamp the axle in place--essentially the problem I have here (though now with no nut at all). In this case I just used a large bolt and nut thru the dropout slot immediately below the axle, with flat washers notched to fit up against the axle/nut, and lockwashers. But it didn't work--it was able to be pushed downward by sideloads on the wheel no matter how tight I got it. IIRC eventually I sort of made it work but ended up redoing something about the dropouts and finding a new nut that wasn't stripped, and that fixed the problem.



Somewhere earlier in this thread, I'd considered something slightly different to this as the original solution instead of pinching dropouts, especially on the outboard sides of each wheel, so I could cut the axle ends down to nearly nothing so it'd fit thru doorways easier and/or give me an extra inch or two of cargo bed frame width. This would have been a plate with two holes in it, one for the bolt and one for the axle. The bolt hole would line up wiht a hole in the dropout plate, above the axle, so I could just bolt it on

The reason I didn't do it then or now is drilling the second hole thru the very thick and hard dropout material just wasn't possible with the stuff I have here; even with oil i destroyed a couple drillbits in the press trying. (and now that they're welded on I'd have to either cut them off like yesterday or figure out how to mount the small press around the frame).


I did also consider welding the bolt head-first to the dropout plate above the axle, but this would negate most or any "height" advantage as it'd stick out significantly farther at that point than if I put the bolt thru the dropout (with nut welded on the inboard side) even if I used a short bolt.

The contraption on there now is actually worse, but at this point the frame itself is too wide to fit thru a doorway (standard house-sized) anyway, so it doesn't matter. I mainly didn't go for the weld-on bolt method yesterday because those weren't the parts that came out when I reached into the bucket. ;)
 
Work commute went fine, although I miss the much stronger braking from the X5304.

I suspect that motor gets much stronger braking than the HSR3548 on the same controller because the X5304 has angled laminations on it's stator, which make it directional--current applied to rotate it in the reverse direction has more drag than in the forward direction (I dunno that "drag" is the right word, but that's what happens mechanically).

The HSR has straight laminations, so it's not directional, and doesn't get this extra "bonus" in braking.

Plus the HSR is narrower both in magnets and stator than the X, which would also reduce the braking power under the same conditions. (and the startup torque).


Startup torque is also less and not as smooth, and much noisier (also due to the difference in lamination style). But it still reaches 20MPH in about the same amount of time (something like 5-6 seconds as a guesstimate), so while the torque curve is different it still does the job.


No problems with the hackjob frame-work, or the makeshift axle support.


Four more days it has to survive. :)



Oh, and I found my splined freewheel removal tool the other day while I was looking for the metal pieces for the axle stub fix, whcih is what enabled me to get the FW off the HSR3548 and use it for this. (it wouldnt' have fit my original plan with it on there, though it would have fit with what I ended up having to do to the frame).

Oddly, it was in a box of stuff that it *couldn't* have been in, in one of the sheds. The box had parts for various dead controllers, none of which I've touched in quite a while (possibly since I moved back in after the fire and sorted stuff out)--but I *have* used the FW tool a number of times since then.

But at least now I know where it is (with the rest of the bike-specific tools, now).

I really ought to paint them all dayglo orange or something so they stand out. :)
 
Glad the patch-up worked out OK so far.

amberwolf said:
I dunno that "drag" is the right word, but that's what happens mechanically.

Yes, there's a thing called magnetic drag - it's a thing. Here in the western world, most of us witness magnetic drag in action, first hand, in the workings of our speedometers (at least for pre-computerized cars): a magnet is spun inside a ferous cup, typically made of steel, in close proximity to the spinning magnet. The cup has a central axis that's mounted on bearings in a housing and the needle is mounted to the cup, but on the opposite side of the bearing. Meanwhile, the magnet is attached via flexible shaft to a wheel or transmission part, sometimes geared. ... The cup is spring-loaded in the opposite rotation to the magnet(s), and the calibration of the spring tension is what ensures the speedometer is accurate; the force of magnetic drag is directly proportional to its rotational speed, so the right spring on the other side (and / or the correct labeling on the dial face) gives an accurate reading of speed...

amberwolf said:
The HSR has straight laminations, so it's not directional, and doesn't get this extra "bonus" in braking.

Interesting speculation, but I have my doubts. However, later today I'll be spending a few hours with a friend who has designed electric motors and I'll ask him.

amberwolf said:
I really ought to paint them all [the bike specific tools] dayglo orange or something so they stand out. :)

...Somehow blood-red seems appropriate - and less gaudy!... AND would match the SBCruiser!
 
Oh, I know more or less how motors work, and magnetics, to a fair degree as a layman, even if I don't remember/use the right terminology. ;) But it'd be interesting to hear what the motor guy has to say about the details.


The lamination difference does create a difference in the drag of the motors--this is why some motors are made that way--they are then more efficient in one direction (but then less so in the reverse). Wind-generation units and other unidirectional motors and generators that need to be efficient are often made this way.

I don't know for sure that it's what makes this motor so much better at braking than the other one, but it is something I can test. If I get it fixed, and have time for the test (would take at least a day), I can swap the left and right wheels, placing this one in the left side so taht it is rotating opposite to it's design, and then test acceleration and braking. I'd still wire it up to the same controller it was using, so all other factors besides direction are the same.

Then I can test using the wheels back where they were, but controllers swapped, since they're similar but not identical and don't do the same braking methods (at least, this is what I assumed based on previous behavior, though it could've been the motor itself all along). (presumably Right is EABS active braking, left is just regen).

Then I can test using the MXUS 3K on both sides, once the second one has it's phase wires replaced and is reassembled and laced into a wheel (in place of the X5304). One is a 3T and doesn't have much motor or braking torque at the low currents these controllers are capable of. The other is a 4T and is as-yet-untested for that, but I expect it to have greater torque for both. (the X5304 is itself a 4T winding, and similar magnet/stator width and diameter, etc).

I could test the latter theory in the http://ebikes.ca/simulator, I suppose, but I still intend to swap to the MXUS from the X5304 anyway, because of the axle and winding damage issues.


I still want to use the X5304 as a front wheel or a middrive or something else at some point, if I can fix the windings.





For the tool color, it has to be something very different than anything else. so I can't lose it amongst any of the stuff. That's how I lose them for weeks or months--I run out of time to work on a project, then I hurriedly put stuff away and something ends up where it shouldn't be because I didn't notice it was one of the special tools I only have one of, etc. The bike tools are the really problematic ones, because I need them very infrequently but when I need them I *really* need them.... :/
 
The wh/mile has gone down with the HSR3548 vs the X5304; it's been about 47wh/mile average for the last two days of work commute.

There's no speedo problem; peak speed was 20.1mph the first day and 20.3mph the second.

So it's a real difference, though there are a few possible sources (assuming any of these are true):

--less drag on the motor when only using the leftside MXUS3k (during right turns mostly).

--lower current average during startup from stops

--lower current during operation

--lower current during active braking

I haven't verified that any of those are the case, but they're all possibilities. Startup and braking are the two most power-intensive things, because I have to do them a lot on the commute, and the trike (plus me) is heavy.

I know that ebraking is not nearly as effective, as I have to use both front brakes in addition to the ebrake, to get even close to the braking that *just* the ebrake used to get with the X5304.

If I have time for it the braking tests described in the previous post would be interesting to see the results of.
 
I've been pondering the braking / angled lamination thing, and realized RTIII might be right in that it shouldn't make a difference; though I think that it would if i were using it sensored...but I'm not, so maybe it shouldn't in this case.

So I made a topic about it here
https://endless-sphere.com/forums/viewtopic.php?f=30&t=90083
for the motor experts around here to help me figure it out (cuz I don't think I can at this point).


So far the hackjob frame work is holding up fine. Wh/mile was a bit higher today, at 52wh/mile, but still notably lower than "normal" with the other motor.


One partly-anticipated side effect of the change in frame (making frame just slightly wider) is a better anti-tip on left turns, and thus being able to maintain a higher speed on left turns (making it a bit safer in traffic or at intersections with impatient cars behind me at the light), since now the righthand tire contact patch is further outboard by more than two inches.

The original trike's contact patch to contact patch width was less than 32", probably 28", maybe 29" (would have to measure). If it's now 2" wider, then that makes it several percent wider, so no surprise it's more stable at a higher left-turning speed.

Of course, with it this much wider, there's zero chance of it fitting thru a house door, so it's now a park-outside-only. I might be able to narrow down the outside dimensions without reducing hte track width much. It may never be important...but I did originally build it specifically so I could take it in the house.


I could move the *inboard* dropout farther inboard, but this will change the width of the cargo area, narrowing it. Not sure by how much, but possibly enough to prevent using the plastic dog crate in there, and that's not acceptable (at least, not until I can build the removable wooden enclosure panels that eliminate the need for the crate).
 
I'm still working out the motor stuff, and how exactly I'm going to permanently modify the frame.

In the meantime, I took Kirin for a ride to my workplace today (since I'm off this week), and while she didn't really like the ride all that much, she did like being at the destination, meeting people, etc.

The ride itself was very different with the HSR3548 than the X5304--it doesn't have anything like the same acceleration. :/ It takes close to 8 seconds to reach 20MPH from a stop with Kirn's 120lbs on the trike, plus the several pounds of plastic kennel. I'd planned on taking both Yogi (in the trailer) and Kirin (in teh trike), but with such poor acceleration (nearly twice the normal time) with just Kirin there, I figured I might never even reach 20MPH with Yogi's 150lbs (plus the trailer's 30-40lbs+) on there, before I'd have to stop again for traffic controls. :roll:

Given how traffic is around here, I didnt' want to risk the dogs even though most people will go around me farther away and sooner when carrying the dogs around than when it's just me. There's always somebody that only cares about the few seconds they might be delayed by being safe and considerate.


The power usage was quite a lot more than usual, too--70.6Wh/mile! Pretty high considering I didn't even try cruising at 20MPH except when in areas cars were passing me (maybe a mile total). Most of the time I was cruising at 17-18MPH.

Full stats for the round trip, which included a couple of small detours to avoid potentially dangerous traffic situations that could be caused by the slow acceleration:

unknown start voltage (forgot to note it, wasn't fully charged, but had cleared stats to get them for just this trip)
53.0v end voltage
48.8V min

5.299 miles
20.2MPH max
13.1MPH avg
24m15s triptime

371.99Wh
70.6Wh/mile

7.110 Ah
0.6% regen
7.156Ah fwd
0.0431Ah regen
-16.5A min
71.41A max

Present odometer is 2871 miles total on the trike.


Since I had such long acceleration times, it gave me a chance to compare the MXUS 3k 3T to the HSR3548, each on it's own. The MXUS takes only a few hundred to several hundred watts for most of the acceleration curve, and it actually starts very low and goes up all the way till i reach cruising speed, at which point I throttle back and it sits around 700-900w depending on wind/road/etc. But the HSR3548 takes 1200-1500w+ the entire time, and gets much hotter than the MXUS, and doesn't have much better acceleration once it passes the first few MPH. It settles down to around 800-1000w or so once I reach 20MPh and let the throttle off.

(The X5304 behaved more like the HSR3548 but didn't take as much power to do the same work)

I'm tempted to retest this with the controllers swapped (as tha'ts way easier than swapping wheels), left to right. If I just phsyically move the controllers around I just have to connect the phse wires (both are sensorless, and both are teh same MPU and very similar inside, both 12FETs).
 
Not sure what you mean by system, but there's a Cycle Analyst v2.3 on the handlebars with a standalone shunt that is mounted between the battery cutoff / charging port on the "downtube" and the battery in the seat/cargo box; the cables from the cutoff also go back to the controllers under the cargo deck. (so I can monitor both charge and discharge currents).

Speed sensor is on the front wheel, reed switch on fork with magnet on spoke. (I don't use motor halls as the speed sensors anymore after having my CAv3 destroyed and a hall in the motor damaged by a wiring fault in the shunt itself that shorted battery power to the speed sensor wire that went to that hall).


Actually noting the data down is either taking pics with the crappy phone camera or just remembering it, depending on how much data I'm preserving (usually just the wh/mile nowadays, as I've already characterized the system with full data previously noted in this thread). Whenever characterizing a new setup I'll post all the data here in the thread to preserve it, then once there's enough to show it's all about the same under various conditions then I usually stop noting down much of anything, just looking thru it on each ride to see if there's anything unusual before resetting the CA's trip stats. If there is, it can clue me into a problem.


I did the same with each build I've had a CA for, or even just plain wattmeters.



Before the housefire four and a half years ago I also kept notebooks with every ride's data, but those were all lost then, and I haven't bothered since. I just do the stuff above.




Today I only used the trike to pull the Mk IV trailer out to the front yard to collect the palm tree cuttings (have to do it about twice a year), and pile it all up on the flatbed to haul it back to the alley so I can cut it up and stick it in the trashcan over the next two or three weeks. The acidity of the palm stuff means I can't just mulch it up and use it to fertilize the yard like I do with everything else. The dogs do like to chew on some pieces of it, and/or play tug-o-war, but I have to use the circular saw or angle grinder to cut the pointy dried palm leaves off so they don't hurt themselves with them.

I used to save the seed "brooms" and use them for decorations along the fences (and yard brooms), but the city started harassing people for that (for no reason I can determine) so I stopped years ago.
 
The MXUS 4T stator is now on the trike for testing, in the 3T's rotor/wheel, on the left side.

It pulls significantly harder than the 3T, as expected, though it's still not enough (need a bigger controller for that). I'd call it comparable to the X5304. If I had that one on the right and this one on the left I'd see some definite improvements in startup-from-stop acceleration times to 20MPH. Dunno how much, would have to test that.

It also draws lots of watts right from a stop, unlike the 3T that gradually goes up until I let throttle off (doesn't level off before I reach 20MPH). (but like the X5304)


It wasn't quite straightforward, though. The axles are different between the two versions of MXUS 3k stator. On the 4T, the wire-side axle is quite a bit thicker than that of the 3T. So I had to use the bearing and side cover from teh 4T on the 3T's rotor/wheel. The drive-side axle is the same, so was able to use the drive side cover/bearing as already installed on the 3T's rotor/wheel.

I also had to modify the wire-side axle to extend the flats farther inward, just as I did on the 3T's axle, so it would fit in the leftside dropouts (otherwise would've been nearly 1/4" too wide between the shoulders). However, I didn't get to cutting the axle on the drive side yet, so it doesn't fully fit into the clamping dropout on the inboard side--the axle is long enough to overlap the pinch bolt, so just like the 3T's axle I'll hve to cut the 4T's as well.

It still fits well enough to do the test rides the start of this post reports on, but it's not all the way into the dropouts yet so I have to take it out and cut the axle to fix that.

Then I'll run the 4T for a while to see what changes there are on the ride.

In the meantime I'll setup the drill press to drill the 4T's rotor spoke flange for smaller holes between it's large ones like I did the 3T. Then I can use it on the right side in place of the HSR3548, X5304, or on CrazyBike2 etc. just by lacing it up into a rim off one of those other two motors.


If I can fix the X5304's winding scrapes, I'm also looking at putting it back on the right side, as long as teh axle repair holds up. I put some Hard As Nails polish on them, after having spread them apart a tad since i don't know which ones are part of teh same winding and which are just passing by. Would suck to fry the X5304 after everythign else it's been thru, from a tiny mistake. :(
 
I forgot to post the pics from yesterday, so:

First up was fixing the 4T V3 MXUS from Korpin, so I could use the stator in my 3T V2; this is what it looks like now, after the work below.
dsc06621.jpg



Since I'm using sensorless I just replaced the phase wires, and tied both hall wire sets inside to simplify getting the wires thru the axle (I wanted quick, not hassle--eventually I'll have to redo that with bigger phase wires plus a set of halls, but not now; there's no point).
dsc06599.jpg
dsc06600.jpg
dsc06601.jpg

The V3 4T has two sets of halls, vs the V2 3T's single set.
dsc06620.jpg


The V3 is also one lamination wider than the V2, which might be a manufacturing error in one or the other.
dsc06610.jpg
View attachment 9
dsc06612.jpg

The V3's axle is thicker on the wire exit side
dsc06613.jpg
dsc06614.jpg
dsc06618.jpg


but the same on the other side
dsc06615.jpg
dsc06616.jpg
dsc06619.jpg


There's not much other difference but there are a few things, like the way the stator support is "clamped" to the stator lams
dsc06608.jpg
dsc06609.jpg
 
Today I finished fixing the X5304; it's not tested on the trike yet but it's ready to once i can redo teh frame to remount it.

Was planning on doing it today, but instead had to do a bunch of yard work today, while it was still cool enough to do it, as it's going to be several degrees hotter the next few days, and it was already 105F today; earlier this week it was 110F+ on a day they predicted 103F, and its' supposed to be 105-106F...so I'd guess it'll be 110-111F instead. :/ Can't do the work in that heat (and it'll get hot fast in the morning too)...so at least I got the front yard and the street part of the back done...but still got the whole backyard inside the fence to do. :/

So...anyway, the X5304:

This is what the axle repair looks like now, with the new phase / hall wires, shown with bearing in place for fit test
dsc06602.jpg
dsc06603.jpg
dsc06604.jpg


The primary damage to the windings, before covering with the Hard As Nails (nailpolish clearcoat borrowed from my brother Raine, best I could come up with out of what we had here).
View attachment 5
I thought I took pics of *all* the damage, including post-repair, but apparently the camera didnt' actually take them when I pressed the shutter. Have had that problem a few times with my replacement button (since the original broke off).

Here's the reassembled wheel.





I also got pics of the trike with the swapped-stator/cover MXUS on the left, with teh dog kennel in place, plus a couple of styrofoam coolers above it (for cargo and keeping direct sun from shining on the top of the kennel so it didnt' heat it up while taking Kirin up to work earlier this week).
dsc06622.jpg
dsc06623.jpg
View attachment 2
dsc06625.jpg
dsc06626.jpg
 
Wow AW you certainly have the Large bike/trike category covered with your twin rear 5304 Clytes to haul her big ass, well done and the dogs look happy
 
Hi Amberwolf,

thanks for the write-up on your data collection strategy - the way you rattled it all off, I figured you had a handy-dandy computer going on, and in a sense, you do; you're own head! :) ... It's a bit less automated than I'd figured, but well done. And sorry to hear you had a fire and lost your logs. :( I worry about that kind of thing and have a lot of my key items near an exit door so if I have to, they're easy to get out - or at least easier than they might otherwise be!

Thanks also for the update on your various repairs. What you could use is a few hundred to buy a pair of new motors for the back so they're matched and can live long service lives instead of all this constant repair work. However, you do seem to be pretty well adept at doing these repairs, so more power to you!

While you've been doing all this, I've got my new stepthrough nearly complete. I'm waiting on a "braze-on" connector and a replacement shipment for a brake package (v-brakes) that was missing a few critical pieces (like one of the four cantilever arms)... Should have been here yeasterday but wasn't! :( Maybe today! ... OH, and I still have to fabricate a new trailer hitch, so I don't have to move it back and forth between bikes. ... And I want to figure out how to do a trailer light hookup, too! 8)
 
dingoEsride said:
Wow AW you certainly have the Large bike/trike category covered with your twin rear 5304 Clytes to haul her big ass, well done and the dogs look happy
Thanks--they are even happier today--see the post a couple down from here about the almsot 600lbs of dog food haul. That is probably the biggest single load of anything so far, I think.

I actually dont' have any X5304's on the trike right now, the one I did have (on the right side) is still taken off after the axle broke around a couple weeks ago, and I repaired it (and the wiring) in the posts above.

At present it is an MXUS 3k 45H v3 4T (M4504) on the left, and a Crystalyte HSR3548 on the right (taken off CrazyBike2's rear till I get one of the others worked out for the trike).
 
RTIII said:
thanks for the write-up on your data collection strategy - the way you rattled it all off, I figured you had a handy-dandy computer going on, and in a sense, you do; you're own head! :) ... It's a bit less automated than I'd figured, but well done.
The only problem with the mental computer is that while it's memory is perfect, the lookup tables are corrupt. ;)



And sorry to hear you had a fire and lost your logs. :( I worry about that kind of thing and have a lot of my key items near an exit door so if I have to, they're easy to get out - or at least easier than they might otherwise be!
Other than a way for the dogs to get out, I'd never really planned for a fire as far as belongings go, the best I had done was have everything I really needed next to the bed so I could just grab it and go whenever I went somewhere that I needed it or in an emergency. Unfortunately that's the room the fire was in, and I wasn't home at the time, just out for lunch with a friend (so everything was all still sitting there).

And I'd been on vacation that week, with one of the many things planned to do being to install the doggie door I'd picked up for cheap. If I had already done that, all but one of them (in a kennel in the bedroom cuz she'd start fights with one of the others when I wasn't there) could've escaped. They might still not have--but they *could* have. :cry:



Thanks also for the update on your various repairs. What you could use is a few hundred to buy a pair of new motors for the back so they're matched and can live long service lives instead of all this constant repair work. However, you do seem to be pretty well adept at doing these repairs, so more power to you!
Actually the motors aren't such a big deal--and one of the repairs (the MXUS v3 4t (M4504) now on the left rear) came as such a deal because it needed rewiring:
https://endless-sphere.com/forums/viewtopic.php?f=31&t=53943&p=1281779#p1281779
but was otherwise fully functional.

The X5304 didn't surprise me really when the axle broke--it's a really common failure with them--at one time there were multiple members selling new redesigned axles for them!

The other MXUS v2 3T (M4503) works fine, it's just not the wind I need for this trike. Still works, just that the powerband is not in the right spot given the low-power controller I have on it.

I'm pretty sure I can break *any* motor if I try hard enough; especially since these things aren't meant to take the weight and shock loads I'm putting on them, since there's no rear suspension at all other than the spokes and tires.



What I really need is to finish saving up for the rest of the parts (and get them ordered) for my Lebowski controllers, then build those and get them setup on there. I have the PCBs and some of the passive parts, but have to get the MCUs, drivers, power chips, etc, and the rest of the passive parts that I can't recycle from other things I already have.

More powerful controllers would take care of the problem of startup from a stop, especially with heavy loads. Even the 30A ones I have now are reasonably adequate, just not what I'd like to have for a better safety margin and quicker startups. (don't need more speed, just more torque).


I also have another trike in the design stages, that would be a middrive, but that needs a differential that can take significant torque thru it and efficiently transmit it to the wheels, which would be larger 26" bicycle wheels (for a better ride over the potholes). I'd use even larger wheels but I already have 26" ones aplenty, and some tires for them.

See the Raine Trike thread for the basic idea (though that one for my brother will use geared hubmotors in the 26" wheels).
https://endless-sphere.com/forums/viewtopic.php?f=2&t=80951

OH, and I still have to fabricate a new trailer hitch, so I don't have to move it back and forth between bikes. ... And I want to figure out how to do a trailer light hookup, too! 8)
I use a 5-pin DIN, like MIDI cables or the old AT-style keyboards (because at one time it was really easy to get cheap keyboard extension cables, which you can just cut in half and use male on trailer and female on bike, and plug together, or use a coily extension between, etc.).

it's not waterproof at all (dielectric grease packed in there helps, but it's very dry here for the most part), but it works; it's what I've used on every trailer so far (so any trailer connects to any bike/trike). 5 wires supports:
+12v / taillight / downlighting
Ground / taillight / downlighting
Brake active low (or high, depending on the brake/taillight module)
Left turn active low
Right turn active low

Since I usually use LED lights on there then it doesn't take much current, and the connector easily handles that, even if I am using keyboard extension cables with thin wires.

See the next post for how my trailer looks today, hooked up to the trike with a load on it.
 
(below is crossposted between the SB Cruiser thread and the Mk IV trailer thread)

Heaviest load so far, I think, at 591lbs of dog food.
dsc06638.jpg
dsc06648.jpg

The trailer carried most of it, 468.5lbs, while I only put 122.5lbs of it in the trike's bed.

The trike carried 3 bags of food, plus 50lbs of cans (whcih I use a spoonful of plus some water along with a hair under two cups of dry kibble for each of their meals, twice a day for each of two dogs; makes their meals more enjoyable. They'll eat anything, but they go for it more enthusiastically with the wet food in there).

FWIW, each bag of food averages about 1.5 to 2 weeks of food, so this trip gives me around half a year of food. At the extreme clearance prices I got it all at, it was only $250 for all of it (I think the receipt said I saved over a thousand dollars, and that doesn't include all the savings!). I couldn't readily feed these monsters at regular prices (at least, not anything even halfway decent, and almost all this stuff is pretty good).

The trailer wiggled a little under this load in turns and on the washboard sections of road at intersections and whatnot, but nothing broke or shifted around. It also balanced just fine, and I had no problems with it pushing the trike around or causing any handling problems.

The trailer tires, inflated to 50psi just before I left the house to go get it, were still squished quite far by the load (which they aren't designed to take, of course), but not enough to be damaged visibly by the end of the ride home, and when checked after unlaoding didn't appear to have leaked.


It did temporarily deform the bed into a slight curve, and canted the tops of the wheels toward each other--not unexpected. Didn't interfere with the load or rub on anything except for the top edge of one bag on the right side that looks like it touched the tire at some point--probably in a right turn.

Once unloaded it appears to be flat again, as normal. If it was built out of "real" steel tubing, rather than old retail signage, it'd probably be stiff enough to not have this problem (and might even weigh less). Alternately if I had a rail around the whole thing, that would also stiffen it up. That was part of the original design plan (to be removable for flatbed use), but like brakes has not been implemented yet.

Speaking of weight, I'm not sure I ever actually measured the trailer's dry weight, which turns out to be about 75lbs, holding it vertically standing on it's nose on a scale. Heavier than I thought, but lighter than it could be. This includes the deck and frame and wheels and hitch and lightbar on the back.

I bolted the SMV sign to the vertical of the lightbar. Usually this sign is bolted to teh tailgate of the trike, or if I have the dog kennel on the trike, it's bolted to the door of the kennel. Whatever the rearmost largest vertical surface is, that's where I try to put it for visibility.


I actually had only a couple of drivers get a bit testy as they went around me, gunning their engines to accelerate as fast as they could--but both of them did so in areas where there was no other traffic but them and me, and they had the complete left lane clear to pass me in. One used that, but the other came close enough to almost hit the trailer's left edge with their tires, and their door mirror overlapped the space the trailer took up.

Nobody honked or yelled, and everybody else just went around me normally, most of them switching lanes well before they got to me, which surprised me because most don't seem to think of that until they get right up behind me.

Was also surprising because I was going around 15-16MPH most of the way home, as braking from 20MPH with this load takes quite a long distance (still haven't built brakes into the trailer--I really need to do that :oops: ). Braking from 15MPh is less than half the distance.


As I was approaching home, about half a mile away, a police car followed me, pacing me at a distance of several car lengths behind for around 1/4 mile, but then they turned off into a different neighborhood. Probably they were just curious as to what the heck this contraption was. ;)

Trip stats below. Note that stats are comparable to teh trip with just Kirin (only about 140lbs load on the trike, no trailer), though load was same on trip there and trip home with Kirin, and here it was just the trailer's 75lbs on the way there, plus 591lbs more on the way back.

Speeds were same boht ways with Kirin's trip, speed there today was normal, but on way home was about 4-5MPH less than usual. I saw about the same power consumptions at the lower speeds though, due to the load on the way back, and possibly because of the different leftside motor (4t vs 3t)


unknown start voltage (forgot to note it, wasn't fully charged, but had cleared stats to get them for just this trip)
53.6v end voltage
49.2V min

4.821 miles
20.1MPH max
12.3MPH avg
(forgot to note triptime)

365.65Wh
76.0Wh/mile

6.971 Ah
0.5% regen
7.0101Ah fwd
0.0376Ah regen
-19.0A min
70.01A max

Present odometer is 2881 miles total on the trike.
 

Attachments

  • dsc06640.jpg
    dsc06640.jpg
    85.9 KB · Views: 2,747
  • dsc06641.jpg
    dsc06641.jpg
    76.8 KB · Views: 2,747
  • dsc06642.jpg
    dsc06642.jpg
    72.9 KB · Views: 2,747
  • dsc06643.jpg
    dsc06643.jpg
    75 KB · Views: 2,747
  • dsc06644.jpg
    dsc06644.jpg
    73.7 KB · Views: 2,747
  • dsc06645.jpg
    dsc06645.jpg
    84.9 KB · Views: 2,747
  • dsc06646.jpg
    dsc06646.jpg
    86.5 KB · Views: 2,747
  • dsc06647.jpg
    dsc06647.jpg
    72.4 KB · Views: 2,747
  • dsc06649.jpg
    dsc06649.jpg
    79.4 KB · Views: 2,747
amberwolf said:
I actually dont' have any X5304's on the trike right now, the one I did have (on the right side) is still taken off after the axle broke around a couple weeks ago, and I repaired it (and the wiring) in the posts above.

At present it is an MXUS 3k 45H v3 4T (M4504) on the left, and a Crystalyte HSR3548 on the right (taken off CrazyBike2's rear till I get one of the others worked out for the trike).

Sorry mate, I just did a quick skim through, well it's good that you are still mobile,
my trike is at a standstill till I get some time to finish it, (I think I posted it in here) I've stretched it and got it running ok but I want to lace the motor into a 24" rim and build the tray for the dogs still, as you can see it is also quite long, about 8', standing it upright to save space
 

Attachments

  • trike20170826_131222.jpg
    trike20170826_131222.jpg
    56.2 KB · Views: 2,739
The cat (Mouse) can practically use food in small quantities like that...but that's like a couple days for the giant monsters like Yogi and Kirin (they eat ~8 cups a day between them).

Pics of Mouse (who is a cat that doesn't know she's a cat, and thinks she's really more like a rabid ferret-squirrel) below are all from Raine (Mouse is really his cat), starting with the newest and going back to the oldest:
 

Attachments

  • mouse plotting again 20161210_054926.jpg
    mouse plotting again 20161210_054926.jpg
    88.7 KB · Views: 2,701
  • mouse 20160915_223930.jpg
    mouse 20160915_223930.jpg
    49.6 KB · Views: 2,701
  • mouse 20160915_223804.jpg
    mouse 20160915_223804.jpg
    37.2 KB · Views: 2,701
  • mouse 20160616_011600.jpg
    mouse 20160616_011600.jpg
    24.9 KB · Views: 2,701
  • mouse 20160315_003209-1.jpg
    mouse 20160315_003209-1.jpg
    68.6 KB · Views: 2,701
  • mouse 20160225_164131-1.jpg
    mouse 20160225_164131-1.jpg
    73.5 KB · Views: 2,701
  • apparently killing the cat tree is hard work20161215_231137.jpg
    apparently killing the cat tree is hard work20161215_231137.jpg
    37.9 KB · Views: 2,701
  • mouse 20151124_012708-1.jpg
    mouse 20151124_012708-1.jpg
    56.9 KB · Views: 2,701
  • mouse 20150521_200412.jpg
    mouse 20150521_200412.jpg
    19.6 KB · Views: 2,701
  • mouse 20150415_080101.jpg
    mouse 20150415_080101.jpg
    35.5 KB · Views: 2,701
  • mouse 20150415_071959.jpg
    mouse 20150415_071959.jpg
    34.4 KB · Views: 2,701
  • mouse 20150407_215203.jpg
    mouse 20150407_215203.jpg
    21.9 KB · Views: 2,701
  • mouse 20150407_214830.jpg
    mouse 20150407_214830.jpg
    31.7 KB · Views: 2,701
  • mouse 20150322_202603-1.jpg
    mouse 20150322_202603-1.jpg
    26.3 KB · Views: 2,701
They're always doing something crazy (and cute). :)


Today while on a grocery run, I did some specific testing of the individual motors vs acceleration times vs cruising speeds, etc. All of the tests were after picking up the groceries, which weighed about 75lbs and filled two small and two large styrofoam coolers filling the cargo area under the rack.


The trike as a whole, both M4504 left and HSR3548 right, takes about 6 seconds to reach 20MPh from a stop. It takes around 600-800w to maintain that speed. Fluctuates a lot as I'm using independent throttles and it's pretty well impossible to get exactly the same output from each, etc.


The M4504 takes around 10 seconds to reach 20MPH from a stop. It takes about 500-600W to maintain that speed. That's much lower than any other motor so far.


The HSR3548 takes around 12 seconds to reach 20MPH from a stop. It takes about 700-800w to maintain that speed.


I don't have them on there now, so the data below is purely from memory, and could be wrong. I tried to look thru this thread for the data I already posted on them, but the only one I ran across was this one:
https://endless-sphere.com/forums/viewtopic.php?f=2&t=67833&p=1261626#p1261626



I think with the M4503 and X5304 it took about 5-6 seconds to reach 20MPh from a stop, and it took 800-900w to maintain that.

Using just the X5304, I think it took 8-10 seconds to reach 20MPH from a stop, and it was still around 800-900w to maintain.

The M4503 by itself took a long while, seems like 13+ seconds, to reach 20MPH, and it also took around 800-900w to maintain.




The above are all with the same pair fo 30-35A 12fet generic sensorless controllers, which usually draw about 60A peak, but sometimes I'll see a 70A peak, for unknown reasons. (it isn't because the system isn't trying to draw it, it's something to do with how the controllers react to teh same situation in different ways at different times for whatever reason).


When I did the M4503 + X5304 test using a 40A 12FET sensored on the M4503 (but generic one still on the X5304), it was under 5 seconds to reach 20MPH.





On a completely separate note:

Back when I was testing with that 40A controller, I was using it sensored with the M4503, and then something happened that seemed to be a hall issue, though I didn't know if it was motor or controller. As I just needed it to work at that time, I pulled the 40A sensored controlller off and just put the phase wires on the sensorless 30A controller.

When i pulled the M4503 stator out to put the M4504 stator in, I found one of the hall sensors (probably the one I'd replaced?) loose and no longer seated in the notch in the laminations, so it was probably not being reliably triggered by the rotor magnets passing it. That would definitely be able to cause the problem I saw.

So there's probably nothing wrong with the 40A controller (which I still haven't gotten to test), and so I can put it back on CrazyBike2's rear wheel, and put the HSR3548 back on there, too.




However, I've been considering taking both the 40A controllers, the generic (ecrazyman?) discussed above, and a Grinfineon that runs the front wheel of CB2, and putting them on the trike, because I use the trike much more (CB2 has become just a backup bike and actually hasn't been used in many months :( ), and because I need the extra power much more on the trike than the bike.

The disadvantages are:

--those 40A controllers are sensored-only, whcih means if I have any hall problems they won't work, whereas with the sensorless 30A controllers they don't require anything but good phase connections.

--I don't have any hall wires run out thru the axle for the M4504 or X5304, though I do have them on the M4503 and the HSR3548. However, the HSR3548 will go back on CrazyBike2, while the M4503 will take it's place on the right side of SBC once I unlace the X5304 and put it's spokes and wheels on the M4503. I'll then have to take the M4504 apart and add in the hall wires.


The advantages are:

-- I'll get much more reliable startup from a stop, and probably a better startup torque and time from the difference in the sensored startup method vs sensorless.

--the higher power limit will also improve startup from a stop.


I don't remember if either of the 40A sensored controllers has a reverse function, though, and I use that function regularly on the one of the 30A sensorless controllers that has it.
 
More testing in the same vein as the above:

On the ride home from work, I tested at 10MPH in some parking lot stretches and back streets, and got about the same for both M4504 and HSR3548 motors, each used separately: 250-280W.

Together its' about 270-300W to do the same thing.

I don't remmeber if I ever did that on the other motors, so no data there.


I tried at 5MPH but apparently there's too much variation in surfaces (tilting/sloping) to get steady enough readings to be useful. It was hard enough at 10MPH.

I saw anything from 50-60W to 280-300W, so while it's probably an average of those, with the former being downslope and the latter being upslope, I wouldn't count it as good data.
 
Gotta say the 600 lbs of dog food is an incredible load, most impressive. Should be in Guinness Book of Records for most animal feed transported on a tricycle :D
 
Back
Top