What a fantastic sight Neal, it was great to meet you and I hope to hear a lot more of your Swizzbee experiences

Had a fiddle with Swb #2 today , nothing major .

 

The lights weren't working but about an hours patient fiddling and all is well again, continuity wire checked all four wire runs and no issues. Must have been a bad brass eyelet contact on the rear where the guards have a filament element running through them front to back.

 

Removed the boost button and wiring run from the bars, the system has enough power without and over boost button.

 

Added a ting ping bell and a refitted the original supplied multi function cycle computer. Added a rack bag with drop down panniers so I can put my 100a HE2 pack in there.

 

Partly dismantled the in frame battery case 20s 1p 9ah nimh cells only registered 3.5/3.6v, boy compared to lion they are heavy for so few cells.

In time I expect I will build a 120/160A 24ah 30Q 8P 7S soft cell pack with each parallel group as 4 x 2 x 7s as the internal case it self isn't overly wide but is long enough for 14 cells, first need to search for a high output 7s bms.

Edited October 18, 20187 yr by Nealh

60a continuous bms (180a over current) on it's way , < £16 .

Nice ride out with #2 today, Garmin say's near as a damn 63 miles & cycle computer with correct wheel size input say's 65ish.

My minor battery repairs mentioned in post #95 to the HE2 pack worked a treat this time a full discharge to cell cut off 3.1v.

Might have a tinker with the front Magura Julie as the braking performance is nowhere as good as a Shimano m446 or similar of the range, may need bleeding ?

 

On a woodland trail.

Waiting for a sand/ballast dredger 'City of Cardiff ' to negotiate the Shoreham Harbour lock.

Two rare pieces of history, lovely old 1944 Tiger Moth as the back drop.

I just love visiting this airfield, reminds me when I was a young boy and used to regularly cycle to South Nutfield when the 'Tiger Club' was in residence.

Edited October 23, 20187 yr by Nealh

Managed about 42 miles of powered riding from the 12.5a HE2 & 15.9a Boston cells which meant 21/22 miles unassisted into the wind.

On their own the Bostons are not much good and the 39a of the pack sags terribly, with the 100a HE2 pack things improve dramatically but range isn't quite enough for the high amp draw Swizzbee.

Planning ahead the HE2 pack I will most likely deconstruct it and use the 20a rated cells in my 48v drinks bottle battery to replace the very mediocre 29E's.

5ah 40a is plenty for my local commute/errand bike.

 

A 56 cell 7s 30Q battery will replace the HE2's giving almost double the ah/wh with an inc from 300wh to 576wh + the Boston pack.

Had a play with the old NIMH pack last Thursday and managed to get it to charge up, I repaired the series link John had cut and after a bit of de-soldering the charge board from the battery rigged up a temporary connection between the pair but nothing happen until a gnd wire on another part of the charge system showed up so I re-soldered this and a charge was accepted.

The pack stopped at 26.4v a tad over 1.3v per cell and all cells appear to be pretty close, I might put all back together and test ride it to see if the voltage holds up and see how many miles it will do out of interest.

 

The weight of the 20 cell pack is quite heavy at 3.5kg for 9ah compared to a 12.5ah 18659 with bms at 1 3/4kg. Even a 56 cell 24ah 18650 will be lighter.

Edited October 23, 20187 yr by Nealh

Loving the looks of the Swizzbee, retro cool and if it was around today I reckon it would have pretty strong sales with a monster battery and a good motor. The maintenance would need to be sorted, so respect for keeping this on the road.

Swizzbee #2 battery build in the pipe line for this winter.

This evening I met a very nice man in a dark and wet car parking lot near Gatwick where an unused MJ1 50 cell pack was handed over for Kirstin's use.

 

At the same time I received 3 x PF 50 cell packs for a £1 a cell, two of these will carefully be de-constructed and the ends carefully dremmel cleaned.

Then I will spot weld up 98 cells for a 8p7s & 6p7s batteries to run paralleled for a 140a 14p 7s pack giving 40.6ah /974wh of power/range.

Once I have stripped out the in frame removable case I have a space in the lower tier to fit a 8p 7s in a (2x4p config x 7s for a 14 cell long battery), the upper tier/shelf houses part of the old NIMH charging apparatus. Once this is stripped out I will see if it will take the rest of the cells if not they will go in a top rack bag at the rear where I currently use HE2's on Swizzbee #1.

Edited November 28, 20187 yr by Nealh

The last lot of cells I took apart after pulling off the nickel strip I found it easier and quicker to cut the little bits that are left on with a Stanley blade rather than using a dremal.

 

Sent from my K107 using Tapatalk

Will give it a try Craiggor.

When I pulled my DAS KIT 26F cell pack apart I did find some bits of nickel were stead fast spot welded to the cell ends and only being careful with a Dremmel fine pink abrasive attachment soon shifted the sharp and prickly bits. I welded the 26F's in to 6p 3s packs and they have been damn good in series with 36v PF's to from a 48v battery.

Swizzbee #2 battery build in the pipe line for this winter.

This evening I met a very nice man in a dark and wet car parking lot near Gatwick where an unused MJ1 50 cell pack was handed over for Kirstin's use.

 

At the same time I received 3 x PF 50 cell packs for a £1 a cell, two of these will carefully be de-constructed and the ends carefully dremmel cleaned.

Then I will spot weld up 98 cells for a 8p7s & 6p7s batteries to run paralleled for a 140a 14p 7s pack giving 40.6ah /974wh of power/range.

Once I have stripped out the in frame removable case I have a space in the lower tier to fit a 8p 7s in a (2x4p config x 7s for a 14 cell long battery), the upper tier/shelf houses part of the old NIMH charging apparatus. Once this is stripped out I will see if it will take the rest of the cells if not they will go in a top rack bag at the rear where I currently use HE2's on Swizzbee #1.

 

What does the controller draw on that bike? I can only imagine how happy I would be with 140 A discharge on the GSM, a small controller tweak down to 14 Amps = 0.1C draw on the battery!

What does the controller draw on that bike? I can only imagine how happy I would be with 140 A discharge on the GSM, a small controller tweak down to 14 Amps = 0.1C draw on the battery!

 

I haven't wired up a watt meter to view amp peak flow, though Easy Rider a gut call Chris on Pedelec forum .de, he has many spares and rebuilt his battery box and states that up to 80a peak is possible from the controller. He has details of his lithium build with 6p 8s PF soft pack for his Swizzbee for 17ah/510wh and says about 60km/37.5miles is feasible.

 

My 15.9ah 39a peak draw Boston cells that Jimmy built for me sags terribly,

the 5 leds light display goes down to 2 on the move with about 9 -12 miles range. I then paralleled in a HE2 100a 12.5ah battery and the leds occasionally losses one led briefly under load but soon reappears once steady riding has presumed, assisted range was better at about 48 miles miles for about 680wh.

 

The bikes draw is greedy and not a frugal battery guzzler.

Edited November 28, 20187 yr by Nealh

That is a handful of Watts!

Yes quite a handful of watts for a 24v 270w Heinzman brushed mid mount motor, one reason why I couldn't resist 150 unused PF cells at a £1 a cell.

£98 to make a 98 cell of 14p 7s near to 1000wh battery pack. I was planning on something a bit less using 30Q or similar.

I have got a watt meter with remote 100a shunt , the watt meter I plan on mounting in a small project box so I can view the parameters as I'm riding to actually see real term data of voltage, draw and wh be used. Just need to find a suitable locale to position it and a method of disconnection for when removing the in frame battery compartment.

100 PF cells with weld spots cleaned off from de-constructed long cell packs, interconnectors were 0.15mm nickel and quite easy to remove leaving only nickel.weld spikes to clean off.

 

 

Cell connectors and pure nickel H strip arrived from China today, ordered from my favourite AliEx seller AnnBattery. Placed order on Nov 19 and here 6 days later they provide a very good service.

The nickel from Ann is reliable and pure though I always test a small piece first for 24 - 48 hrs.

 

As the Arduino boss spot welder isn't really designed to do copper, I was going to layer the nickel 3 times which means lot of welding.

The H nickel is 7.5mm x 0.15mm so 1.125 mm2 and good for about 5a optimal or 7.4a acceptable.

Nickel ampacity is only 4.46a per 1 mm2 for optimal temperature with out becoming warm, though on ES they say 6.6a is an acceptable draw with warming above this it gets v.hot and glows red at 10-12a loading.

 

Having browsed and searched ES for ideas I have found a couple of posts where a copper sheet solution may be off use without the need to spot weld it.

If this solution works then it would act as a massive heat sink with great conductivity converting more battery energy in to watts output instead of loosing battery energy to heat by the warming and poor conductivity of nickel strips.

 

Copper is four times more conductive then nickel, silver is even better then copper but expensive and hard to solder and weld due to damaging heat increase to cells.

Edited December 6, 20187 yr by Nealh

I have finished and installed the Panasonic PF pack and it is now all but finished and installed in to the bespoke battery case.

I hmmmed and arghed about construction and interconnectors and then after browsing ES battery building some more settled for one to try as an experiment, the battery case isn't very wide so only a 4p is really possible so I connected the cells up as 2 x 4p for 8p final discharge capacity/amps, the pack only connects as a full 8p battery with soldering the two banks of cells at the discharge wires.

I utilised cell spacers for mechanical holding as the battery is a regular shape.

 

Top side even cell numbers. I started spot welding pure nickel P strips with the 0v group and then changed my mind for the rest of the battery. As I didn't wan to tear the nickel off and Dremel the cell ends again I let them be. The X parallel groups allow for the each cell to flow it's current equally to the center.

 

Bottom side odd cell numbers.

 

 

Series connecting the even cells top side.

For series interconnectors I used 9.5mm x 0.4mm copper, then another parallel nickel X on top. The pair of parallel X's have are capable of 13.2a each to the center. Cells are 10a rated the copper buss is 60a rated.

I used a a pair 2mm x 0.15mm braided flat copper for the cell balance, these spot welded very easily to the nickel.

 

Bottom side odd cell numbers.

Edited February 26, 20197 yr by Nealh

Bottom side odd number cells.

I forgot pics for the sense and discharge wiring so drew them on.

Top side odd cell numbers.

 

 

For sense & final discharge I soldered and culminated a pair of 12awg wiring in to one soldering to a 10awg

 

The battery case is aluminium so on the bottom of the battery I used 1mm epoxy glass fibre sheet for protection/insulation and 0.4mm sheet on the top then used three bands of shrink wrap. All in part visible in the below pic.

Almost finished battery pack cable tied in to case.

Blue line is the panel mount 5.5mm x 2.1mm charge jack with dust cap, this is fitted to a bit of 1mm epoxy sheet. Waiting for some epoxy resin to glue the piece in place. I am utilising the old charge plug wire hole as the old Nimh charger was of no use so no onboard charger will be carried.

Heat shrinking the whole of the battery wasn't necessaty as it is within a case.

 

 

 

Added also a watt meter with 100a remote shunt. The shunt can be seen on the upper shelf to the rhs, this sits on a pair of 4mm poly carb doughnuts for insulation.

The watt meter screen is constantly on so the feed wire I have soldered to the cases switch (circled yellow) otherwise the lcd will drain the battery, the cases leds have been rendered useless now the old nimh charger board has gone so the watt meter will be my fuel gauge.

BMS is waterproof compact one with 60a constant discharge and capable of over 100a for a short time , the battery is 80a 23.2ah and will be paralleled with a 7s7p 140a 17.5ah HE2 battery. Currently the HE2 is 7s 5p, I have a 7s 3p I use with a 36v battery so the 3p will be deconstructed to increase the battery ah/wh.

The watt meter lcd I am mounting the top of the head set via an aluminium bracket which is in progress, connection is via a 4 pin jst lead so can be easily disconnected to remove the battery case.

 

Case with side on and almost finished.

Each battery will only be asked for 30 - 40 amps max each, although at times 60- 80 amps may be asked for typically this will be less.

 

The build is experimental and if I need to I will return to it for further work if required.

Edited February 26, 20197 yr by Nealh

With those currents I can see your point about using copper. For those of us using 15-17 Amp controllers I think pure nickel will work just fine.

With those currents I can see your point about using copper. For those of us using 15-17 Amp controllers I think pure nickel will work just fine.

 

Yes agree for up to 20a pure nickel is all that is needed, just a case of using correct size for cells in P to series to carry the amperage across.

 

In the experimental X pattern for parallel I used, parallel busses only need to be able to deliver the individual cell current to the center, it is the series buss that needs to handle the combined amperage.

 

Generic China battery with branded cells tend to use 8mm - 12mm x 0.15 nickel sometimes layered.

Really cheap unbranded cells I don't know but on ES some use plated steel and pics show the rusting.

Bosch builds are quality and use bespoke laser cut buss and can be 0.2 or 0.3mm nickel x about 12mm.

Edited February 26, 20197 yr by Nealh

Epoxied the charge port in to day and the product sticks like Sh** to a blanket in about 3-5 mins. In the past have used 2-part Araldite but wasn't overly impressed, this time I used the Gorilla 2 - part epoxy and it is superb.

Going to make up the steerer tube bracket tomorrow for the amp/watt meter and a go at repairing the Boss spot welder. Spent today using the rest of the PF's in a 10s4P which is all finished, used the Melectric which is a great bit of kit.

I replaced the Watt meter with ext shunt as I simply couldn't get it to work (likely me not installing it correctly) but no matter what I tried zilch, nothing, zero in any info read out.

So I now have a nice smaller watt/ammeter installed in a small project box on the head set, current measurement is via a directional hall sensor and the V+ wire to the controller.

https://www.aliexpress.com/item/32904528079.html?spm=a2g0s.9042311.0.0.dfba4c4dUM0Q22

 

Todays ride showed a max of 34a in high torque mode and some 800+w.

Typical usage on flat terrain is 10 - 16a continuous.

Edited October 30, 20196 yr by Nealh

I replaced the Watt meter with ext shunt as I simply couldn't get it to work (likely me not installing it correctly) but no matter what I tried zilch, nothing, zero in any info read out.

So I now have a nice smaller watt/ammeter installed in a small project box on the head set, current measurement is via a directional hall sensor and the V+ wire to the controller.

https://www.aliexpress.com/item/32904528079.html?spm=a2g0s.9042311.0.0.dfba4c4dUM0Q22

 

Todays ride showed a max of 34a in high torque mode and some 800+w.

Typical usage on flat terrain is 10 - 16a continuous.

 

Hungry but beautiful!

I replaced the Watt meter with ext shunt as I simply couldn't get it to work (likely me not installing it correctly) but no matter what I tried zilch, nothing, zero in any info read out.

So I now have a nice smaller watt/ammeter installed in a small project box on the head set, current measurement is via a directional hall sensor and the V+ wire to the controller.

https://www.aliexpress.com/item/32904528079.html?spm=a2g0s.9042311.0.0.dfba4c4dUM0Q22

 

Todays ride showed a max of 34a in high torque mode and some 800+w.

Typical usage on flat terrain is 10 - 16a continuous.

 

OOOOO! I like that one. Mine has developed a tiny bug and shows 907 peak Amps... I think I would have noticed a burning smell if that was the correct number...

 

The battery being about 2 metres from where I would like to mount a multimeter is a bit of a challenge. Multimeter with Bluetooth transmitter?