Weber and Coulomb's MX-5

Post up a thread for your EV. Progress pics, description and assorted alliteration
Squiggles
Senior Member
Posts: 742
Joined: Wed, 22 Apr 2009, 03:19
Real Name: Neil
Location: Newcastle NSW

Weber and Coulomb's MX-5

Post by Squiggles » Fri, 30 Apr 2010, 14:05

Maybe he was just a little sloshed at the time....

User avatar
EV2Go
Senior Member
Posts: 2059
Joined: Wed, 16 Jul 2008, 00:21
Real Name: Paul
Location: Brisbane 1963

Weber and Coulomb's MX-5

Post by EV2Go » Fri, 30 Apr 2010, 14:54

yep or maybe it was a Fraudian slip and he was thinking of getting sloshed.

User avatar
coulomb
Site Admin
Posts: 3766
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by coulomb » Fri, 30 Apr 2010, 16:42

EV2Go wrote: Is that a back slosh or a forward slosh Image

Sorry, where I used to work, the "slosh" was the obvious name for a back slash. A sort of "self defining term".

Like "bang" for the windier "exclamation mark", and "hack" or "hash" for the "sharp or pound" sign. As in "hack-define" in C/C++ programs. Also "tick" for "reverse apostrophe" or "back quote". "Splat" for "asterisk" (even though there is a perfectly short "star").
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.

User avatar
EV2Go
Senior Member
Posts: 2059
Joined: Wed, 16 Jul 2008, 00:21
Real Name: Paul
Location: Brisbane 1963

Weber and Coulomb's MX-5

Post by EV2Go » Fri, 30 Apr 2010, 16:47

splat * love it Image have to remember that one.

Nevilleh
Senior Member
Posts: 773
Joined: Thu, 15 Jan 2009, 18:09
Real Name: Neville Harlick
Location: Tauranga NZ

Weber and Coulomb's MX-5

Post by Nevilleh » Fri, 30 Apr 2010, 16:52

All this horrible abuse of the English language was obviously brought about by the invention of software writers who couldn't actually type, so they try to minimise their keystrokes! And it soon spreads into their speech.

User avatar
Electrocycle
Senior Member
Posts: 985
Joined: Sun, 19 Oct 2008, 20:23
Real Name: Andrew
Location: Sydney
MSN: dumhed@dumhed.com
Contact:

Weber and Coulomb's MX-5

Post by Electrocycle » Fri, 30 Apr 2010, 23:24

Where I used to work the @ symbol was called a "snail"

A snail was outputted whenever there was a watchdog reset, and a major failure could be recognised by "snails of death" :)
(this is embedded comms gear)
The Engine Whisperer - fixer of things

Squiggles
Senior Member
Posts: 742
Joined: Wed, 22 Apr 2009, 03:19
Real Name: Neil
Location: Newcastle NSW

Weber and Coulomb's MX-5

Post by Squiggles » Fri, 30 Apr 2010, 23:57

coulomb wrote:
Like "bang" for the windier "exclamation mark",


I thought that was a question mark ?

User avatar
woody
Senior Member
Posts: 1715
Joined: Sat, 21 Jun 2008, 02:03
Real Name: Anthony Wood
Location: Mt Colah

Weber and Coulomb's MX-5

Post by woody » Sat, 01 May 2010, 00:23

Long-winded! not winding as in "The long and winding road" :-)
Planned EV: '63 Cortina using AC and LiFePO4 Battery Pack

User avatar
coulomb
Site Admin
Posts: 3766
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by coulomb » Fri, 07 May 2010, 06:54

Inspired by this now obsolete version of the Tritium Battery Management System, ...

Image

and with many thanks to Tritium_James for the idea, we realised that we could save some wiring and at the same time make it practical to measure link voltage drops as well:

Image

{Please excuse the low quality of the image; it was saved with compression quality 10% to squeak it under the 100k limit, and I wanted to get a hint at the grandeur of the panellised artwork, hence the large size.)

We measure the voltage of the links connected to the negative terminals. This has the advantage that one side of the link is connected to the negative reference of the micro, so there is no need to subtract two similar voltages, or even to make two measurements (and the current through the links could change between measurements, leading to bad readings).

The disadvantage of this scheme is that we can't measure link voltage drops during charging or regeneration; voltages less than zero read as zero.

I suspect that with our short, single squiggle joins, and with three of them "in parallel", we'll end up with too much stiffness. But possibly the best way to find out is to make several boards and try them, so we intend to get a panel (34 boards) made like this. Doing one panel separate from the rest will cost us a repeat job charge, $90. We're using B.E.C. Manufacturing because they are in Brisbane and we know that they do excellent quality work. In fact, the $90 buys us the option to spend another $390 for another setup charge, if we find we have a mistake. We should have worked out all these details at the much cheaper prototyping stage (using BatchPCB), but now time is running out.

As usual, if you guys spot any glaring errors, or have any suggestions about the overall concept, you could save us a bundle. So here is a sharper image of one board in one "rotation":

Image

I might add that getting these two extra tracks onto this packed PCB and have it work with the two rotations has taken ridiculous amounts of time; we're the kind of people that don't know when to give up. The current solution is a considerable improvement over my design, which had one of the tracks running around 3/4 of the board. Remember that we are keeping 5 mm (200 mil) clearance between parts of the PCB that could be 900 V different in potential to the main part of the board. A spray-on conformal coating will be used as well, and wires running off the board will be siliconed.

Edit: the thick purple "tracks" represent router paths.

Edit: We may be able to have as many as 9 boards joined (more if we use an extra-large panel). However, this makes them hard to populate; we can only fit 5 in a row in our high tech oven (an electric frypan). The panel is made with rows of 8,9,8,9, for a total of 34 boards. This will give us maximum flexibility for where to make the breaks. We can still measure link voltages over board breaks, by using 3 wires instead of the two we were planning on. But most boards will have zero wires, assuming this all works out.

Edit2: Full name for Tritium BMS
Last edited by coulomb on Thu, 20 Mar 2014, 18:25, edited 1 time in total.
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.

User avatar
Electrocycle
Senior Member
Posts: 985
Joined: Sun, 19 Oct 2008, 20:23
Real Name: Andrew
Location: Sydney
MSN: dumhed@dumhed.com
Contact:

Weber and Coulomb's MX-5

Post by Electrocycle » Fri, 07 May 2010, 13:32

try saving PCB artwork as .gif files. It works a lot better than jpeg for drawings.

Heh, just noticed the second one is a gif :P
The Engine Whisperer - fixer of things

User avatar
EV2Go
Senior Member
Posts: 2059
Joined: Wed, 16 Jul 2008, 00:21
Real Name: Paul
Location: Brisbane 1963

Weber and Coulomb's MX-5

Post by EV2Go » Fri, 07 May 2010, 13:53

Still have no idea of what I am looking at (electrically, know what it is) but it sure is pretty Image

Squiggles
Senior Member
Posts: 742
Joined: Wed, 22 Apr 2009, 03:19
Real Name: Neil
Location: Newcastle NSW

Weber and Coulomb's MX-5

Post by Squiggles » Fri, 07 May 2010, 13:58

I assume the fancy routing between boards is to allow flex.
Have you considered making the tracks in the flex area larger and on upper and lower as a precaution against fracture?

User avatar
coulomb
Site Admin
Posts: 3766
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by coulomb » Fri, 07 May 2010, 14:06

Electrocycle wrote: try saving PCB artwork as .gif files. It works a lot better than jpeg for drawings.

Yes, if don't care about the size. GIF is lossless, so I can't compress a big image like that into less than the 100k limit on post images. I suppose I could have copied it to a bit bucket somewhere and pointed to it, rather than hosting the image on the AEVA wiki.
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.

User avatar
weber
Site Admin
Posts: 2623
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by weber » Fri, 07 May 2010, 14:30

Great post Coulomb. Here's the schematic to help make sense of the PCB layout. All criticism and suggestions gratefully received. Thanks squiggles.

Image
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

User avatar
Electrocycle
Senior Member
Posts: 985
Joined: Sun, 19 Oct 2008, 20:23
Real Name: Andrew
Location: Sydney
MSN: dumhed@dumhed.com
Contact:

Weber and Coulomb's MX-5

Post by Electrocycle » Fri, 07 May 2010, 17:15

coulomb wrote:
Electrocycle wrote: try saving PCB artwork as .gif files. It works a lot better than jpeg for drawings.

Yes, if don't care about the size. GIF is lossless, so I can't compress a big image like that into less than the 100k limit on post images. I suppose I could have copied it to a bit bucket somewhere and pointed to it, rather than hosting the image on the AEVA wiki.



hmm I would have thought it'd get low enough in size.

maybe too many colours in the pallette?

I generally find for schematics you can get a gif pretty small
The Engine Whisperer - fixer of things

User avatar
weber
Site Admin
Posts: 2623
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by weber » Sat, 08 May 2010, 05:17

By the way, we took your advice about 2 weeks ago and ordered a Tritium Wavesculptor200 (the 450 Vdc version) for the MX-5, and will wire the battery as two strings of 114 cells, so the two strings can be parallelled. When the 900 Vdc version becomes available we can put them in series as originally planned.

Anyone know anyone who wants a top-of-the-line new industrial 75/90 kW Variable Frequency Drive for $9,900. Control Techniques/Emerson Unidrive SP4502. I obviously don't expect anyone to buy it for EV use now that Tritium's AC controller is available.

As well as Weber, Coulomb and Newton, we now have another SI unit who has volunteered to help us. Pascal (Warrick Beatie) is designing and making the motor/gearbox adaptor plate and the adapter plate for the belt drive of the aircon and power steering. He is also doing the drawings for a new flywheel with integral taperlock hub that was the recommendation of Ultramotive engineer Ben Guymer who kindly visited us and the MX-5 and gave us the benefit of his knowledge and experience.

We have set a deadline for the MX-5 to move under its own power by EV Tuesday 22nd of June (yes, 2010 Image ). It will by no means be finished then, but we need some kind of milestone like this to work towards if we are to have any hope of getting it to the AGM in Adelaide this year.

[Edit: "Waveshaper" -> "Wavesculptor" and added link]
Last edited by weber on Wed, 19 Mar 2014, 08:07, edited 1 time in total.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

7circle
Senior Member
Posts: 540
Joined: Thu, 04 Feb 2010, 02:18
Real Name: Ken
Location: Melbourne

Weber and Coulomb's MX-5

Post by 7circle » Sat, 08 May 2010, 06:33

You guys are unreleanting.Image

I'm still not to sure how the PCB's connect. LINK OUT to above battery LINK IN I assume. Cant tell from Track design.

Just a thought that as the link is a diagnostic fault function it would be sad to loose it during charging. If you added a pull up res of 47K, say using the spare I/O pin so it could be disabled. This would give a positve voltage offset, thus bringing negative voltages on the link signal into range of the AD input. Using the P1.3 set high to Vcc, would supply 2.5V from IC2 regulator, and resistor 47K to Link net. This would pass current through the link path (47K(Rpu) + 47K(R11) + BAT_LINK) to Vss(GND). So when Batt zero current Vlink is 1.25V. If there was 1V across BAT_LINK when charging Vlink would be 2.5V - 1/2 x (2.5 + Vlink) = 0.75V. So upto 2.5V accross BAT_LINK when charging.
When discharging the Vlink will just range from 1.25V to 2.5V.

{Edit - If the BAT_LINK is Open circuit Vlink will be 2.5V not 1.25 when Ibat is zero. Good for diagnostics}

Have you considered using the squiggle links as only reaching tenticles with vias on the ends that overlap onto aligned vias on the next PCB.
So each PCB is the same (or two types for aligned the LEDs). This still requires three solder joins with wire down the vias for strength.

But Top stuff from W&C.
Thanks for sharing.

{Edit- Looking closely at track layout "Lo" net on Neg-Left PCB is hanging out as a stub which is a bit of an antenna. Might be worth tiding this net if noise input is concerning. }
Last edited by 7circle on Fri, 07 May 2010, 21:04, edited 1 time in total.

User avatar
coulomb
Site Admin
Posts: 3766
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by coulomb » Sat, 08 May 2010, 14:52

7circle wrote: I'm still not to sure how the PCB's connect. LINK OUT to above battery LINK IN I assume.
Yes, that's right.
Just a thought that as the link is a diagnostic fault function it would be sad to loose it during charging. ...
Yes, actually, at the start of charging, when the current will usually be maximum and steady, would be an ideal time to check the links. The internal pullups of the MSP430 are about 36K from memory, and they don't guarantee much accuracy. However, I imagine that the stability would be quite good, and we're not looking for more than 10% accuracy here. Also, we already have to calibrate the cell voltage and temperature; we were expecting to have to calibrate the cell input as well.

The fact that charging current is small compared to maximum discharge current (5.5 A vs 120 or 240 A in paralleled pack configuration) may be more than compensated for by the convenience of being able to check link voltages when the car isn't moving. We might even be able to use a non-contact infra-red thermometer while charging to verify what the BMS is telling is. And it only costs one tiny piece of copper track to have that facility. The only possible downside would be if the extra input, in digital input mode without the pullup, might cause problems with extra capacitance or drift current. But I think the chances of that are small.

We essentially only need a "still good" or "check me pronto" signal, though a "check me when you get around to it, no big deal yet" indication would also be useful. And there does happen to be a spare pin right next to the one we're using to read the link voltage... brilliant! Thanks heaps for the idea!

Edit: I meant to add that Weber might be very pleased with this idea. He was devastated when we had to add the R15 pullup when the micro provides programmable pullups. Of course, the value of pullup it provides is 24 times too high, but it happens to be about right for this application. We could also adjust the values of R10 and R11 (the link resistors) to make the internal pullup value even more suitable. So this would be a case where the microcontroller reduces parts count, as it was meant to!
Last edited by coulomb on Sat, 08 May 2010, 04:58, edited 1 time in total.
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.

Squiggles
Senior Member
Posts: 742
Joined: Wed, 22 Apr 2009, 03:19
Real Name: Neil
Location: Newcastle NSW

Weber and Coulomb's MX-5

Post by Squiggles » Sat, 08 May 2010, 14:58

7circle wrote: Have you considered using the squiggle links


Thanks for naming them after me, but honestly I can't take any credit for their invention Image

User avatar
coulomb
Site Admin
Posts: 3766
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by coulomb » Sat, 08 May 2010, 15:15

7circle wrote: Edit- Looking closely at track layout "Lo" net on Neg-Left PCB is hanging out as a stub which is a bit of an antenna. Might be worth tiding this net if noise input is concerning.

Actually, that's also a good point. We've been thinking in terms of keeping the printed circuit design completely unspecialised, since we don't want to have to make a change more than once. But we already have a 17-step procedure for panellisation (it's on the schematic; we haven't shown it before to save space).

Step 17 is:
"Clean up the top and bottom ends of the four boards and ensure their boundaries are closed in Mechanical layer 1." We could make the cleaning up process explicitly include removing unneeded stub tracks at the top and bottom end.

There are a few stubs that result from our rotation procedure, but I think it's a bit much to fix 34 copies of those (more if we use the larger panel size).

Panellisation is already a gruelling procedure, so adding a few more steps won't be much extra effort in proportion. We'll still have to grind off PCB material, and leave track stubs, on boards that we have to cut; we can't do much about that.
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.

User avatar
weber
Site Admin
Posts: 2623
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by weber » Sat, 08 May 2010, 15:20

Squiggles wrote:
7circle wrote: Have you considered using the squiggle links


Thanks for naming them after me, but honestly I can't take any credit for their invention Image


Hee hee. I had noticed that. But then I realised that "squiggle" is what Coulomb calls the tilde character "~" in his colourful hackers' jargon of bangs ! splats * and sloshes \.

We have implemented your suggestion of wider tracks and using both sides of the squiggles. Thanks.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

User avatar
coulomb
Site Admin
Posts: 3766
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by coulomb » Sat, 08 May 2010, 17:35

woody wrote: Long-winded! not winding as in "The long and winding road" :-)

Well, I wanted a word that said "long-windeder". Maybe it should be winddier?

Well, windier is correct, per
http://www.thefreedictionary.com/windier.

I don't know how to spell the word that means the most bendy (windy, in the other sense).

Obligatory pedant-free content: Weber provided me with this page on names for ascii characters:

http://www.codejacked.com/know-your-key ... lat-whack/

No sn@il, but I like strudel.

Finally, an emailed comment from Weber: "Something I haven't seen: If this / is slash and this \ is slosh then if this ' is tick then this ` should be tock." Certainly!   Image
Last edited by coulomb on Sat, 08 May 2010, 07:37, edited 1 time in total.
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.

User avatar
coulomb
Site Admin
Posts: 3766
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by coulomb » Wed, 12 May 2010, 15:43

Well, yesterday was a whirl of activity. I started populating a battery rack, mainly to try out our idea of giant cable ties for clamping the cells:

Image

You can see a pair of old, analogue style BMU boards installed; at least they are the right size. Also, I installed a few Sky Energy links upside down, noting that they do fit. It's probably not a great idea, as there is almost no room at all for the links to give, at least not without distorting, which would put a lot of load on the cell terminals. We were considering upside down links at one stage when we thought we might connect boards along the side with a sort of loop of PCB material. However, we're well past that now.

Image

All of the links from one particular plastic sealable bag were really shiny and bright; you could use them as shaving mirrors. These links would be a year old now, and all the others are quite tarnished. So if you know that you won't be needing your links for some time, consider storing them in plastic sealable bags, and possibly make an effort to exclude most of the air.

Image

Perhaps we should have done the same with the cells themselves; some of the terminals were tarnished rather badly; a few of them had copper terminals that were completely black. I touched them up with fine sandpaper as best I could.

The cable ties (which were quite expensive, about $2 each I think for a bag of 25) didn't work out well. They fit, and you can put a bit of tension on them, but nowhere near enough. Nylon is 6 times stretchier than steel, as technical guitar aficionados (like Weber) will know.

At the moment, we're back to "what was so bad about the threaded rod again?". Weber has shown that they can be welded (they're supposed to be stainless) to make the appropriate lengths, so that's something. It's just going to be quite tricky getting cells in and out of battery boxes.

Meanwhile, Weber and Newton were attempting to mount one of the recently made battery boxes, so we can paint it before it rusts. (Even our motor's rotor is rusting... definitely need to get to that.) Alas, it failed the "fitness test", as in, "it fiddn' dit!"   Image It was decided to "move" the corners of the protusion with 6 cells in it, 16 mm at one end, and 30 mm at the other, losing 46 mm overall (1 cell width). Unfortunately, "moving" corners like that is almost as much work as building a new box, but it was done by late evening. So now we're back to where we thought we were at the start of EV day. I guess these things are to be expected.

Weber was 95% finished with panellising the BMU boards for production; we've decided that $400 is too much to pay for an artwork modification, and we don't have time for another Batch PCB prototype, and quick prototypes are also expensive. So we're going to make 250 boards, order parts for 250 boards, and make them work. The "squiggle joins" (or should I capitalise the "s" now? Image ) are completely untested, but we have the fallback of guillotining them off and just using wires. After lunch, he decided to finish the panellisation, so the order could be sent off. The BMU artwork has been taking up most of our time for a while, so it will be good to have the order sent. I showed Weber some pointers for quicker deleting of tracks (Tools Unroute Connection on Protel), and I learned some tips from him as well.

A half hour after I left him to finish that job, I was called upstairs. "There's all this green!". Green on Protel indicates a design rule violation, such as a short circuit, track too close to a via, etc. It's good to fix these, as some manufacturers will refuse to make the board if there are such violations, and also of course, it might mean an actual error in the design. It took a bit of detective work to establish that the problem was some text "uTXD+" on the bottom layer. Sadly, we don't have such text on the board. Although we did, a hundred revisions ago, back in the analogue days. Back then, we used the "comment" field of a component as actual text in the copper, back before we were paying extra for a second silk screen overlay. We had changed this text to appear on the bottom layer and mirrored. When we didn't want that any more, we clicked the "hidden" box, so it doesn't appear on final artwork. It seems that the Protel design rule checking didn't respect the hidden checkbox, and the text, now floating off to one side of the board, caused no problem until panellisation, when there is now another board there, and the clearance check failed.

The solution was to change that text to be on the bottom overlay, but that meant 34 changes. Protel does have a global change facility, but when I tried it, it asked "you are about to make 1620 changes; ok or cancel?". With a bit of fiddling, I worked out how to only change the texts that were on the same layer, i.e. the bottom layer, to the bottom overlay layer. This time, 32 objects were changed (I know I had done one by hand, and must have done another). Anyway, the violations were gone, so by about 4pm, we had the order off to BEC Manufacturing.

Next on the list was ordering the charger; we've decided to get a 2 kW Elcon charger that should be able to charge helf the pack. It might end up being our primary charger, depending on how things go with controller charging. As others have mentioned, Elcon also sell DC/DC converters, and some of those are high voltage and isolated. Alas, only the "800 W" models (600 W continuous, 800 W peak) will take 420 V, which isn't quite enough. Still, we could turn off the DC/DC while charging, and the Elcons may be quite economical. Indeed, all their DC/DC models are under US$100 (before shipping), except the 800 W models, which are over US$350. We'd eventually need two of these for 730 V working, and two of them would be huge. So we just ordered the charger. It will "go into production" as soon as Weber wires over the money. It came to US$649, including US$165 ( Image ) for shipping. We've ordered the CAN version, even though others have pointed out that the RS232 option may be available in the non-CAN version, and is more convenient for us to use. But we can't be sure about that, and getting technical information like that from a Chinese company... well, I guess most of us know how likely that is. The CAN option is US$44, so it's not a deal breaker.

Edit: added two more images.
Last edited by coulomb on Wed, 12 May 2010, 06:00, edited 1 time in total.
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.

Nevilleh
Senior Member
Posts: 773
Joined: Thu, 15 Jan 2009, 18:09
Real Name: Neville Harlick
Location: Tauranga NZ

Weber and Coulomb's MX-5

Post by Nevilleh » Wed, 12 May 2010, 16:12

Bit of "deja vu" there while reading that post! I tried big cable ties on my scooter batteries to make packs of 4 but that didn't work too well either. Gave up and bought some threaded rod in 1m lengths and made up clamps that way. Which worked very well.
My SE cells are contained in a steel box and are just a tight fit. This is working very well so far as although the pressure is not high it is applied over the whole cell surface which results in quite a bit of total force. But I can still get them in and out OK after some six months.

My jaw drops when I look at the complexity of your BMS!

I know how much work it has been getting mine going and you have heaps more than I have. I hope it works out for you. But at least you have twice my workforce available!

Mine is working well now at both ends and I have to get some more slave boards made so I can connect up more cells than just eight. But it works OK with the laptop pretending to be more boards, so I am optimistic.

bga
Senior Member
Posts: 492
Joined: Mon, 01 Sep 2008, 19:27
Real Name: Bruce Armstrong
Location: Perth WA

Weber and Coulomb's MX-5

Post by bga » Thu, 13 May 2010, 16:55

coulomb wrote:
EV2Go wrote: Is that a back slosh or a forward slosh Image

Sorry, where I used to work, the "slosh" was the obvious name for a back slash. A sort of "self defining term".

Like "bang" for the windier "exclamation mark", and "hack" or "hash" for the "sharp or pound" sign. As in "hack-define" in C/C++ programs. Also "tick" for "reverse apostrophe" or "back quote". "Splat" for "asterisk" (even though there is a perfectly short "star").
Some immortal poetry:

< > ! * ' ' #
^ " ` $ $ -
! * = @ $ _
% * < > ~ # 4
& [ ] . . /
| { , , SYSTEM HALTED


The poem can only be appreciated by reading it aloud, to wit:

      <     >    !    *     '    '   #
     Waka waka bang splat tick tick hash,

       ^    "        `        $      $    -
     Caret quote back-tick dollar dollar dash,

       !    *     =    @    $        _
     Bang splat equal at dollar under-score,

        %      *    <    >     ~     #     4
     Percent splat waka waka tilde number four,

          &       [       ]     .    .    /
     Ampersand bracket bracket dot dot slash,

          |            {          ,     ,   SYSTEM HALTED
     Vertical-bar curly-bracket comma comma CRASH.

from Here
It's not the end of the world, but I can see it from here.

Post Reply