Safe's Electric Bike Project #002

[RLT]

OK, Never mind.

I was just relaying my personal experiences, hoping to help you make a great machine; Just my way of of giving thanks for the information and ideas you and the others here have given me on my projects.

I have no desire to make and sell spot welders nor to start building battery packs commercially, although I would have been willing to help people here out with a few welds occasionally and to provide the information on building their own welders if they want, assuming that either of my devices do a good enough job, which is far from certain at this point.

Here is just one quick illustration though: of the 18 NiMH C & D cell batteries I have sitting on my counter now, here are four that have dents of significance These are all merely from the pressure of the spring pushing the + battery tips into the negative base, in MagLights under fairly normal use. The ones that actually sarted leaking were in lights that saw rougher use... especially if they were dropped either base first or head first and had the full force of the impact mostly perpendicular to the base plane of the battery. (Sorry, but I didn't keep the ones that were badly damaged around to prove it.)

Now admittedly, these ARE Powerizer batteries which don't have the greatest reputation for quality in the world. Maybe other brands are more stoutly built.

And of course flat top batteries don't have the projecting + points, so pressure from springs and/or impact would be more widely distributed, and less likely to cause problems.... But then, making proper contact using flat tops without tabs is even more of an 'engineering' problem.

EDIT: Oh, one thing I just now thought of , and it is kind of embarrassing that I just NOW thought of it instead of a couple of years ago. The ones that actually started leaking may not have been due only to a mechanical stress failure of the battery case, but it is quite possible that the dents made the batteries short out internally and overheat. There was no significant bulging of the cases, or melted shrink wrap to make a short seem self evident, but t could definitey be a contributing factor.

If you think I'm a lying troll just trying to cause you trouble or make money off of you, there is no point in continuing this discussion and I'll go elsewhere.

And with that I'll wish you good luck on your project and hope that you discover that you were totally right and I was totally wrong, and thank you for sharing your knowledge and your discoveries with everyone.

 

[TylerDurden]

Heck, I used to use a carbide lamp too. When the guy in front of you gets stuck in a tight spot, you come up on his rear and give him a "carbide assist". Beware of barking spiders when doing this.

eeeww.

 

[fechter]

Listen to RLT.

He has *actual* experience. (see my signature line).

 

[safe]

He has *actual* experience.

You know when I first got started with these electric bikes most everywhere that I read it said that the hub motor was the only way to go. My gut feeling was that gears could give me better performance than a fixed gear, but I lacked the knowledge to be able to perform the mathematical proof that my hunch was going to be valid. After building the bike and now learning all the formulas and doing the hard work of figuring everything out I've vindicated what my gut feeling was at the start. It's a fact... gears do work and you don't need 25 lbs of heat sink iron to make it work.

The idea of pressing "D" cell batteries together is an idea worth trying. There are many benefits of having things done this way like ease of servicing, ease of pack building (no effort at all) and better performance. There are "pioneers" out there in the RC World already doing this with success, so I know that it can be done. I feel that I owe it to myself and the advancement of knowledge in general to try this out and see what possible problems might pop up.

My intentions go like this:

First, start with a very low amount of spring pressure. See how things go and if needed increase the pressure based on how well the batteries perform. If there is a desire for better conductivity then try the thin layer of silver paste. If there are still problems then try some more pressure. If after trying, adapting, and trying some more I find that the idea simply doesn't work, then I revert back to a rigid soldered connection.

My guess is that the few people who have tried this probably used a screw cap approach and did not think about the thermal expansion and contraction issues and then quickly gave up.

Come to think of it, the SMARTEST thing to do is to have a spring on BOTH ends. (this is in reaction to the overpressure controversy) This way you can control the cell to cell pressure and ALSO suspend the batteries against hard pounding caused by the road. (reducing the chance of denting)

So it might be the double spring that is the key to success.

Only time and testing will let me know...

 

[vanilla ice]

With all that weight stacked up, I think you would need some pretty heavy springs to keep them from bottoming out. Like cylinder head valve spring heavy.

 

[Beagle123]

I think this thread illustrates our biggest hurdle: making a modular, balanced battery pack. I'm constantly amazed that a bulletproof solution isn't available. And when I say bulletproof, I mean bulletproof against inefficiency too.

I think we have to look at all possibilities to find a solution. I don't think the current ones are acceptable at all (but I'm definately not saying good work hasn't been done).

 

[safe]

With all that weight stacked up, I think you would need some pretty heavy springs to keep them from bottoming out. Like cylinder head valve spring heavy.

Maybe, maybe not.

As Beagle123 just said we need to keep trying new things. Until they start mass producing automotive sized NiMh or LiFePo4 batteries at a low price we will need to attempt to find better ways of doing things.

I suspect that little actual pressure would be required because the RC solderless batteries have discovered a phenomenon called "zapping". Basically what happens is the electrical current tends to effectively file down the high spots due to heat, the more narrow the protrusion rising up in the metal the more heat it takes and the faster it gets "zapped". The RC guys claim that the batteries get better over time because of this.

So my strategy would be to take it slow and start off with the "zapping" process for several rides. Then when that seems to have reached a plateau I would experiment with the silver paste if I'm still unsatisfied. If after all that proves unsatisfactory then I begin the process of forceful pressure rather than gentle pressure. I'm hoping never to get to the point of forceful pressure.

The failures in the past seem to be with people who used forceful pressure combined with a screw type fitting. I know in advance that the screw type cap is an error, so I'm starting off with an advance over the most naive (natural) first approach.

 

[Malcolm]

Small pads of copper braid or something similar between batteries might also help with cushioning as well as preventing momentary open circuits. You would obviously have to make sure that the pads could not migrate or fragment.

 

[safe]

Small pads of copper braid or something similar between batteries might also help with cushioning as well as preventing momentary open circuits. You would obviously have to make sure that the pads could not migrate or fragment.

Direct physical connection is pretty darn good. Solder has a 10 times higher resistance than steel and silver is the lowest of them all. I just don't see conductivity as a problem, in fact the previous tests have all returned with evidence that conductivity will be IMPROVED with a direct connection.

The significant issues are:

Over pressure on the metal part of the cells causing a dent.

Side pressure on the surface of the cell wrapping digging into the cell because of movement.

The first problem is all about spring pressure. The second problem has to do with the packaging of the tubes, how smooth it is. Both of these things seem manageable.

Sometimes you just have to "be brave" and "be a pioneer".

My "guess" is that in the coming years the batteries will be sold in automotive sizes because the electric cars would never bother with little cells like we do now. So the future will probably look more like my existing bike, three or four big batteries rather than many small cells.

 

[v_tach]

With 20 D sized cells arranged end to end, wouldn't the mass of the cells alone and any potential internal forces created from bumps and jarring in the road surface create significant compressive stress? Especially to those cells near the ends.

 

[CGameProgrammer]

Solder has a 10 times higher resistance than steel and silver is the lowest of them all.

That's absurd. According to this data, expressing electrical conductivity as siemens per one-millionth of a metre:

Steel with 1% carbon, 99% iron is 0.0983.
Lead solder (40% tin, 60% lead) is 0.0655. About 2/3 as conductive as steel.
Silver solder (96% tin 4% silver) is 0.113. 15% more conductive than steel.
For reference, copper is 0.596 and silver is 0.630.

So lead solder is roughly 1/10 as conductive as pure copper or silver, but certainly not steel; it's 2/3 as conductive as that.

And unless you're welding those springs, the steel is always like a cold solder joint, i.e. not very good. The melted solder, on the other hand, forms a great connection.

 

[safe]

With 20 D sized cells arranged end to end, wouldn't the mass of the cells alone and any potential internal forces created from bumps and jarring in the road surface create significant compressive stress? Especially to those cells near the ends.

Well they would have the friction of the insides of the tubes to inhibit movement. I doubt that bouncing is going to cause a denting problem, but it would more likely cause an abrasion problem on the sides of the cells. After a year or more of riding the abrasion could wear into the cells and possibly cause a problem. But I can guarantee that for that first summer of riding I'll be taking things apart and checking everything often. That's one of the benefits of the free cells is that you can take it apart and test each cell and see if any of them are not doing well. The rigid pack concept means that if there is a problem your ability to find where it is can be difficult.

 

[safe]

So lead solder is roughly 1/10 as conductive as pure copper or silver, but certainly not steel; it's 2/3 as conductive as that.

The springs need not have contact with the battery, they just need to provide pressure. The actual contact would more likely be copper surrounded by some type of electrical insulation material. I was thinking of possibly even using about an inch "cubed" of rubber used as a plug that would fit the 1.5" x 1.5" square tube. The rubber would act as the spring and you could just drill a hole in the middle and snake a wire through to the copper contact point. People have done this as springs for a front fork, they use these rubber things and you can get a more dampened suspension feel that way. The "ideal" would be like a shock absorber, with full dampening and all, but that's kind of going overboard.

Rubber would do two tasks at once in that it would provide a spring and also provide electrical insulation.

Obviously I haven't been to the hardware store to see what I can cook up for this yet... :wink:

 

[safe]

More Sanding

I filled in the inside area so that I can get a more precise result. By making this perfectly flat in the middle I know where everything is relative to the flat baseline.

The obvious thing to say is:

"Gee, that's not that hard of a shape, why didn't you just do it like that in the beginning?"

Well, I really didn't know exactly what I wanted in the beginning and now I do and I'm on my way to getting there. It's sort of artistic... you have to build stuff that appeals to your own personal sense of style. I want this thing to look awesome and you just can't rush that...

 

[CGameProgrammer]

While we're on the subject, what's the resistance of common connectors like quick-disconnect, bullet, or ring&screw? It's occurred to me that a good connection method for allowing quick insertion/removal of individual cells, without risking open circuits, is to solder wires to each cell but to use a connector on the other end, so you just plug the cells into each other. The problem is that for small cells, the power would have to traverse many of these before it's done. For example, in a 10s6p pack, it can travel through up to 14 of these before leaving the pack.

 

[safe]

It's occurred to me that a good connection method for allowing quick insertion/removal of individual cells, without risking open circuits, is to solder wires to each cell but to use a connector on the other end, so you just plug the cells into each other.

Well you also have to add the length of the wires too. The Deans Connectors that I use claim that they are as conductive as the wires that connect to them and I'd believe it, they seem okay. This would actually be MORE work than less however. The central idea of the direct cell to cell connection is that there's work involved to make it happen. All you do is throw the cell into a tube and then make sure your end caps are set with the proper spring pressure. (which could be a chunk of rubber and not an actual spring)

The "real" answer is automotive sized batteries or even something that is about the size of the 12V 10Ah SLA's they use for scooters and some electric bikes. My bike is designed to accept four of these inside the front fairing. (2.56" x 5.94" x 4.56")

But that's the future.

My bike is designed so that in the future there is room for a battery box underneath the frame and right behind the front wheel inside the front fairing. I fully expect that eventually they will sell large batteries and so I've designed for that. (though it's nearly impossible to predict the exact dimensions that will be the future standard)

 

[CGameProgrammer]

I don't think there's anything really bad with some "work" involved in replacing cells if it's easy and not error-prone. Desoldering is not easy because you risk shorts or heat-related damage as you're trying to remove a cell surrounded by many others. But simply unplugging something or unscrewing something is very simple and easy. And sure you have to prep the new cell before insertion, but that's perfectly safe and predictable and not time-consuming for just one or two.

 

[safe]

I don't think there's anything really bad with some "work" involved in replacing cells if it's easy and not error-prone.

But if the unmodified cell approach (solderless) works (which I hope to prove) why do ANY work if you don't have to? Of course this presumes that the unmodified cell concept does in fact work which is the whole point of trying it out. Using the old "pioneer spirit" allows me to try new things... (it's the American Way :wink: )

 

[safe]

Pretty Much Done

At this point improving what I have is getting harder and harder and I end up having to stare at it longer before I can even notice problems. The weather is getting cold and when it starts getting below about 60 degrees it's harder to do fiberglass work. So I'm at the point where I'm going to have to decide whether to prepare an indoor room for this type of work or not. The possibility of awful smells leaking into other parts of my house during the middle of winter is not too appealing to me.

Gosh, I hate to think that I'll have to wait until spring, but my options are pretty limited...

Even with a fully ventilated garage and a very powerful overhead fan that sucks the air out it's still nasty, nasty, work dealing with all this dust. Fiberglass is even worse and the little particles get into everything. You practically need a hazardous materials room with a ventilator to use this stuff. On a good day were you can sit out in the open and the dust or fiberglass can dissipate into the wind it's not so bad, but in a closed room this stuff is terrible.

It's looking nice though.

 

[dirty_d]

have you put your 8 speed hub in a wheel and make sure everything lines up and works ok. mount the shifter for it and get that working, plenty of stuff to do

 

[safe]

have you put your 8 speed hub in a wheel and make sure everything lines up and works ok. mount the shifter for it and get that working, plenty of stuff to do

I was saying that I was "pretty much done" with the original molds. I still need to make the actual fairings and that means first female molds (done over the originals) and then the final product. (done inside the female molds) The limiting factor is the temperature... if it get's below 60 degrees the fiberglass has a hard time hardening. Using more catalyst works up to a point and then it begins to have problems.

At this rate the bike won't be in action until next spring.

I've bullt the rear wheel already with the 8 speed hub in it, but I don't have it mounted in the picture.

 

[safe]

Fork Replacement

I've decided to replace the fork with a rigid one. It will be like the one on the Project #003, but with larger diameter tubing. Pictures will come in time... this thread will be brought back to life... (the 60 volt concept would be applied here)

 

[safe]

Fork Parts

I've started to assemble the parts I need for a rigid front fork. The tubes are 1.25" verses the 1.125" on the project #003 so they should be a little stiffer even though they are actually thinner walled tubing. We will see...

This is as far as 6 hours of work will get you.

 

[safe]

Improved Triple Clamps

Given the close tolerances that exist between the frame and the fork on this bike I figured it was worth it to go that extra step and put a small amount of offset into the triple clamps themselves. On the Project #003 bike the triple clamps have the steerer tube centered and the fork legs do all the offset.

 

[safe]

Ten Days Later...

Probably eight can be considered actual "work days", but overall this stuff just takes waaaaay too much time. Custom building something like a triple clamp where you need it to be both very strong and also very light weight (and precise too) means that you have to be very careful at each step. That takes up a lot of time.

It's looking great... probably my best work yet... but I'd never be able to make a living doing this sort of thing. :?

In a factory process you could probably cut the fabrication time down to under a minute... (ideally this would be cut from aluminum or something like a motorcycle triple clamp)

Custom, custom, custom... time, time, time...