Safe's Electric Bike Project #001

[recumpence]

Looks good. Even though you didn't invent this helper cell technology, I think it is cool you sort of figured it out for yourself.

I must say, though, this thing is looking more and more like a motorcycle by the day.

Matt

 

[safe]

Just need to figure out where the best deal is on cells is at then maybe I can imitate your pvc setup.

I think Dirty_D bought a batch too... it seems to be a really good deal and All Battery is a reputable place to buy from. Their internet price is cheaper than their website price by a large margin... I have no idea why...

$133.94 delivered for 96 cells... or $1.40 per cell...

That's enough to make two 40 cell strings (80) of 48 volts making a total of 4.8Ah. (exactly what I have on the bike) Plus you would have 16 cells left over for rotation of the runts or to keep around for something else. If you break the sets of 40 down to sets of 20 then you have 24 volts and can buy their 24 volt chargers, again, that's what I did.

http://cgi.ebay.com/96-NiCd-Sub-C-2400mAh-Batteries-for-PowerTools-Flat-Top_W0QQitemZ330198582341QQihZ014QQcategoryZ40975QQtcZphotoQQcmdZViewItemQQ_trksidZp1742.m153.l1262

 

[safe]

Todays Ride

Trip Length - 16.93 miles
Trip Time - 35.47 minutes (2147 seconds)

(16.93 / 2147) * 3600 = 28.39 mph average.

This was a nice ride through the backroads in no big hurry, but holding a pretty decent time. It rained recently so everything is really green and it was just plain beautiful to be out riding the bike. At the end of the ride their was still very little sag.

The Solderless Tubes are working without problems.

 

[safe]

Even though you didn't invent this helper cell technology, I think it is cool you sort of figured it out for yourself.

It was just supposed to be a test of the Solderless Tubes. But now it's been a nice thing to know because it means that you can buy 80% SLA and 20% NiCad and really expand the usefulness of the SLA. It ended up an SLA enhancement technique.

The best things to learn in science and otherwise are when you stumble across them.. Forrest Gump would be proud... :lol: (though I had heard of the idea previously... I just never bothered to test it before)

 

[D-Man]

You should try the nicads by themselves with the voltmeter on board, note the voltage drop. Then try sla's by themselves, note the voltage drop. Then both together and note the voltage drop. That way we have a better idea of whats really going on.

 

[safe]

You should try the nicads by themselves with the voltmeter on board, note the voltage drop. Then try sla's by themselves, note the voltage drop. Then both together and note the voltage drop. That way we have a better idea of whats really going on.

The NiCads, being only 4.8Ah would struggle to deliver even at their 10C rate. The current requirement is 40 amps by the controller. So you would be simply stressing the NiCads by trying to pull the whole bike with them alone.

The SLA cells I've ridden with before and while they can deliver the max amps at first (while at full voltage) they start to "sag" in voltage while trying to deliver the same amps.

It's only together that you get the benefits of both.

So I just don't see what good that test might deliver other than possibly stressing out my NiCads. That kind of test is one of those things where afterwards (after your NiCads are destroyed) you are saying to yourself:

"How could I have done something so dumb?"

(no offense, but I don't want to risk it)

I can simply tell that the bike works better... it's not something that is like me to make up stuff that isn't true, if something works then I say that it does, if it doesn't work so great I'll say that too. The logic for mixing chemistries is simple enough in that you get the best of both worlds. SLA is cheap, but tends to sag and has a Peukerts problem. NiCads are a little more expensive and you have to use the SubC's to get the high "C" rate performance. Mix the two at a ratio of about 80/20 and it seems to work well.

The real trick would be to play with different ratios.

What might be the minimum ratio of NiCads to be able to see a noticeable improvement?

The NiCads are in effect acting like an ultracapacitor that buffers (and assists) the SLA cells. It's the cheaper and easier way to build "pseudo" ultracapacitors into your bike.

 

[TPA]

Do you have to fully discharge your nicads before recharging?

 

[safe]

Do you have to fully discharge your nicads before recharging?

I went to Battery University to check on that because I know that if you don't fully discharge the NiCads every once and a while they get a memory effect that reduces capacity.

From Battery University:

NiCad Service needs:

"Discharge to 1V/cell every 1 to 2 months to prevent memory. Do not discharge before each charge."

...so at some point I will need to do this. It might be a good idea after all to do as D-Man suggested and drain the Nicads (without the SLA's connected) while at the same time being careful not overdo it with the "C" rate. The 4.8Ah cells run at the full 10C would be 48 amps at peak which is within the 40 amp limit. Hmmmmm... I'm going to have to think about this... :?

 

[recumpence]

I ran nicads for alot of years. If you do not discharge them after every run, they get imballanced very easily. In fact, I have seen as much as 50% imballance within only a few cycles.

However, running them with the SLAs may change that affect.

Matt

 

[safe]

17 - 17 - 15

17 miles, then 17 more miles, then another 15 more... that's a lot of riding in one day. The extra batteries make it so I'm getting tired out more than the bike.

That's 49... almost 50 miles of full "on" road racing action all in one day... and without suspension... with a 47 year old man... yikes!

 

[safe]

In fact, I have seen as much as 50% imbalance within only a few cycles.

I did notice that my ending voltage on this last run was something like:

24.3
24.3
25.2
23.4

...or something to that effect. It's as though one string of 24 volts went low while the other went high. This will all be interesting to sort out over time. In order to break the memory effect I'm supposed to go all the way down to 1V per cell, so that means down to 20 volts... way lower than what the SLA's are allowing. I've tended to want to shorten the rides for the sake of the SLA's.

Hmmmm.....

Maybe on the last part of the ride I should disconnect the SLA's (that are sagging anyway) and run with the NiCads until they completely drain themselves down to 20 volts. I'll have to try a bunch of stuff.

 

[safe]

Drop the Sagging Cells and Empty The NiCads?

First, I just took the bike out under NiCad only power and it's just amazing that these little NiCads (only 4.8Ah at 48 volts) are able to power my bike completely. This DOES increase the heat generated in the NiCads by a large margin compared to using them slowly with the SLA's.

The idea here is that the sequence that would seem to make the most sense is:

Begin the ride with both SLA and NiCads so that they can assist each other in making the most efficient power.

Then disconnect the SLA's when sag starts to take place (about 15 miles) and you are ready to end the ride.

This will drain the NiCads and if you get the hang of it they will mostly be emptied (20 volts) by the time you get home. In the process your SLA's will be more protected than if they participated and the NiCad's will give decent power for the short time they have left.

 

[johnrobholmes]

How many cycles do you have on your nicads? What charge rates are you using?

 

[dirty_d]

you drained a 48V nicad pack down to 20V? my pack with the same cells at 36V 12Ah is pretty much dead at 30V under load. although that doesnt really make sense, if they can hold up 30V stalled at full throttle then i should have a lot more torque, i mean like it wont even barely move in grass, maybe its the lvc or something kicking on and off.

 

[recumpence]

Nicads have most of their output power available untill the very end, then they die "All at once".

Matt

 

[Toorbough ULL-Zeveigh]

You should try the nicads by themselves with the voltmeter on board, note the voltage drop. Then try sla's by themselves, note the voltage drop. Then both together and note the voltage drop. That way we have a better idea of whats really going on.

if 20% 4.8Ah NiCd is good, then 100% 24 Ah NiCd would be 5 times better.
'safe' has moved 20% down the road of admitting to himself what everyone has been telling him all along.
The less lead-acid the better.

 

[safe]

How many cycles do you have on your nicads? What charge rates are you using?

I just got these so I'm just testing them out for the first time.

The chargers are 24 volt 1.5 amp and the cells are 2.4Ah, so that makes them being charged at 0.625C. They get recharged really fast compared to my SLA charger, but they don't heat up as fast while charging as they do when you try to ride them at 10C... so 0.625C charging does not appear to be a big problem for them.

The main goal of this was testing of the Solderless Tubes... that was the central advancement and the reason for testing. At this point I'd have to say the Solderless Tubes do appear to be a success. But people like to hear stories of thousands of miles of use without flaws and so that's going to take a while. Plus there are all the issues about eventual runt identification and replacement... so this will be a looooong story to complete.

Be patient...

 

[safe]

you drained a 48V nicad pack down to 20V?

No, I'm looking to drain the 24 volt subpacks down to 20 volts.

I haven't even done it yet and after checking my spreadsheet it will take a while because these NiCads calculate out to about 7 miles of range all by themselves... this alone explains why the Peukert's Effect disappears for the first 15 miles, the first 7 miles could have been done by the NiCads alone. Working together the "C" rate of the SLA's are probably cut in half, so the calculated full throttle range "should" be 16 miles just with the batteries as they are, but I went 21 miles as a first test. It's possible that more mileage is possible if I'm willing to accept some sag. As it is I'm getting at least 15 miles of sag free performance while at full power all the time. That's very good for this bike.

 

[johnrobholmes]

Nicads take 5-10 cycles to break in. Wait until after that to weed out runts unless you know that there are dead cells already. You should be able to charge up to 1.5C depending on the resistance of them, but your charge rate will make them last longer and have more MAH in a charge.

 

[safe]

Nicads take 5-10 cycles to break in. Wait until after that to weed out runts unless you know that there are dead cells already.

Yeah I know... I'm just getting these things tested.

The Solderless Tubes appear to be a success and that was the area the was a real question mark before. So the structural issue of how to contain and conduct the cells is working (Solderless Tubes) and now I'm looking to play around with the quirks of the chemistry. I'll dial it in... it just takes some time.

The main reason for the attraction to the SubC NiCads was the price... the fact that All Battery has this ebay sale going on is making it so attractive that the price actually beats SLA within about a year or two of ownership. Add in the fact that they are a lot lighter and it's a bargain.

Weight is less critical on my bike because it's for the street. It's able to carry 86 lbs of lead and now I'm running less than it's designed for so it's light by comparision. (I'm lighter by about 15 lbs)

SLA is still a good deal on price... but NiCad SubC's (on sale) combined with Solderless Tubes seems to be the formula to a low priced pack. It may be the low budget answer of the moment.

 

[safe]

New 10 Mile Average Speed Record (for this bike)

Trip Length - 10.00 miles
Trip Time - 16.45 minutes (1005 seconds)

(10.00 / 1005) * 3600 = 35.82 mph average.

I started the day by riding around on NiCad only power in order to completely drain everything down to the 1 volt per cell level. One tube was holding so much more capacity than the others I had to recharge the others a little just so that I could drain it. (tubes need to be combined in order to make the motor run) Finally (after about an hour) I got all the tube sets to where they are supposed to be (20 volts) and then recharged normally. After almost two hours all four chargers finished within a few minutes of each other, so that's going to mean they are pretty close.

For this ride I started off going about 3.5 miles just to get to the starting point and then went and did exactly ten miles and checked the time. That's why it's exactly 10.00 miles I made sure of it. The total ride came close to 20 miles. When I got home the NiCads and the SLA's were pretty close to being equal within their voltages.

That's a new average speed record... I suspect there's room for another one or two mph average because even on this route there was a u-turn and a few stops. Pretty good though.

 

[safe]

I passed 4,400 miles today...

From now until, say, 5,000 miles I'm effectively testing the Solderless Tubes. Always good to be testing something while you ride.

Can you believe 100 miles in less than three days...

 

[recumpence]

I have seen people pedal without any assist over 100 miles in one day.

Yes, yes, yes, I do agree 100 miles in 3 days is pretty good. It took me over a week to hit that. Of course, I ride 6 miles at a time and spend time with the family while I repeak my pack.

Matt

 

[D-Man]

That's it! With all your tubes and sla's you have to conduct a 24 hour marathon and see how many miles you can go. Its a charge and go marathon. The winner gets bragging rights and an online trophy!

 

[John in CR]

Safe,

Should have waited a month or so. 4500mah subC NiMH's with a 40A drain rate with or without tabs are available now on Ebay for $.58/wh http://cgi.ebay.com/100-Pcs-of-Matched-Sub-C-4500mAh-NiMH-Battery-with-Tabs_W0QQitemZ330224562651QQihZ014QQcategoryZ34061QQtcZphotoQQcmdZViewItemQQ_trksidZp1742.m153.l1262
Your packs would have cost about the same, but take up a lot less space and would weigh only 11 pounds plus the tube cases.

John