Weber and Coulomb's MX-5

Post up a thread for your EV. Progress pics, description and assorted alliteration
User avatar
jonescg
Senior Member
Posts: 2368
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

Re: Weber and Coulomb's MX-5

Post by jonescg » Wed, 10 Jan 2018, 16:25

weber wrote:
Mon, 29 Feb 2016, 03:54
And with the auxiliary battery fixed, it became worthwhile to fix the power steering, again with help from Bladecar.
For some reason I found myself looking at radio controlled hobby motors and found some in-runner motors that spin up to 5000 rpm and are good for about 2 kW. I thought to myself maybe these could be good for driving power steering pumps in conversions where replacing the power steering system with a complete electric system would be too difficult.

Do you think it's possible to use one of these motors:
RC_Motor_inrunner.jpg
RC_Motor_inrunner.jpg (49.51 KiB) Viewed 1194 times
And one of these speed controllers:
RC_Motor_ESC.jpg
RC_Motor_ESC.jpg (31.78 KiB) Viewed 1194 times
And control the speed of the motor (and therefore power consumption) by making it inversely proportional to the speed of the vehicle?

I have no idea how much power a power steering unit needs, but surely a 1 kW brushless motor is up to the task?
AEVA National Secretary, WA branch vice-chair

User avatar
weber
Site Admin
Posts: 2485
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Re: Weber and Coulomb's MX-5

Post by weber » Wed, 10 Jan 2018, 16:54

Not really sure Chris. I suspect it might be a bit too high speed and low torque. But mine is current limited to 30 A at 12 V (360 W) so maybe.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

User avatar
coulomb
Site Admin
Posts: 3399
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: Weber and Coulomb's MX-5

Post by coulomb » Mon, 02 Jul 2018, 16:47

It's project day at Weber's house, and he's finding lower range than usual (80 km compared with 130 km when new). One of the cells has been playing up, and was replaced a few weeks ago. He replaced it with one that seemed to have the lowest internal resistance of his spares. But that one seems to have low capacity. So today's project was to find the best of the spare 8 cells, 4 of which are being used as his auxiliary battery.

We spent half the day getting the 8 cells top balanced. Well, 7 of them; one of them ruled itself out of contention by its behaviour when charged at a paltry 6 A. While that was happening, Weber constructed the load resistor: about 2.5 metres of 0.7 mm dia steel wire, aiming for about 1 Ω. This was wound around a piece of wood (ceramic would have been better, but wasn't to hand). The ends were soldered to some copper wire and battery clips from some other project. The resistor and board went into a 10 litre bucket of water. The wood tended to float (who'd have thought), so a dead drill battery was used as a weight to hold it down.

This is near the end of the test:

Making bubbles sm4.jpg
Making bubbles sm4.jpg (129.81 KiB) Viewed 857 times
You can see hydrogen and oxygen bubbles streaming from it, due to electrolysis of the water. The blue thing is an ice-brick.

After about 70 minutes, the first cell went low voltage, with its Cell Monitoring Unit beeping. The others had a good enough spread of voltage that we could pick the winner, and in fact rank the rest of the cells in case more are needed later.
Learning how to patch and repair PIP-4048 inverter-chargers and Elcon chargers.

User avatar
jonescg
Senior Member
Posts: 2368
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

Re: Weber and Coulomb's MX-5

Post by jonescg » Tue, 03 Jul 2018, 16:31

For only the strong survive.
AEVA National Secretary, WA branch vice-chair

User avatar
weber
Site Admin
Posts: 2485
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Re: Weber and Coulomb's MX-5

Post by weber » Sun, 04 Nov 2018, 12:14

Since the last post in this thread four months ago, Mexy has mostly been working fine. There has been the occasional difficulty in getting started, that seems to occur when the auxiliary battery is low. We really should have staggered the turn-on of the eleven EV200 contactors. We have since designed and built suitable turn-on-delay boards that we use in our Black Monolith solar power systems. We could publish the schematic and veroboard design if anyone is interested. But I haven't got around to installing any in Mexy.

There is a bigger problem with Mexy now. The battery is nearly 10 years old and a few cells have high internal resistance (which limits power) and only about 50% of their original capacity (which limits range). Our BMS tells us which cells these are, and it is blatantly obvious that they are the cells that have spent the most time at higher temperatures—those in the rear window and those directly above the motor and power steering reservoir. Replacing these should give Mexy a new lease of life.

Unfortunately I have not been able to find any place to buy LFP cells of the same dimensions. They are the Sky Energy 40 Ah cells (blue). Sky Energy became CALB and the cells became grey, but CALB no longer make the grey cells of that size either. However Greg MacDonald recently advertised a bulk buy of UBetter brand 50 Ah cells with almost the same dimensions. They are 10 mm smaller in their smallest dimension, 7 mm taller at their terminal contact areas, and exactly the same width. The terminals are a similar spacing, except they have M8 studs instead of M6 holes. But I can drill out the holes in the CMUs. So I bought 16 of the UBetters to replace the worst of the old Sky Energys.

Because our BMS' cell monitoring units (CMUs) are designed as a single board that goes across 8 cells, I need to pack (or "shim") the new cells out by 5 mm on each broad face (or 10 mm between new cells), and I need to pack the old cells up by 7 mm. Fortunately there is enough spare height for this.

Yesterday I decided I had to go and buy some material to do this packing. So I tried a crazy search strategy. I went onto the Bunnings website and searched on simply "5mm". Then I started slogging through the 2314 items that came up. At 432 I struck gold! Well yellow. No, not Yello (the Swiss electronica duo from the mid '80s that Coulomb introduced me to). But having typed that, I had to go to YouTube and find The Race. :)

This kind of yellow:
https://www.bunnings.com.au/macsim-fast ... k_p0044303

Since the broad face of the old cells is 116 x 180 mm and these shims are 100 x 150 mm, they don't require any cutting or joining. Brilliant! Bunnings also have them in 10 mm (black) for between new cells, and 2 mm (blue) to go with the 5 mm (yellow) under the old cells. Like this:

IMG_1978b.jpg
IMG_1978b.jpg (163.79 KiB) Viewed 313 times

What's that you say? The thing in the background? That's the hot buttered 15 mm fruit shim (brown). ;)

One thing that concerned me was that the new cells are in aluminium cases, which are known to carry a voltage with respect to the terminals, and have only a thin blue plastic film to insulate them. In Mexy, the cells sit in steel angle. What if the thin insulation wore through with vibration over time and two distant parts of our 720 V battery were shorted together via the chassis? :o

So I decided to measure the resistance between the case and the terminals. I first noted that a fully-charged rested cell with a terminal voltage of 3.35 V had 0.61 V between case and positive, and 2.74 V between case and negative. I then connected various resistances between the case and one or the other terminal and discovered that for both terminals it behaves like approximately 180 kilohms in parallel with 1 millifarad. So there is no significant danger, even if the insulation does wear through. And in any case, we have an insulation monitor that checks the battery every time we turn the key to start, that will raise an alarm.

IMG_1977b.jpg
IMG_1977b.jpg (165.38 KiB) Viewed 313 times
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

User avatar
jonescg
Senior Member
Posts: 2368
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

Re: Weber and Coulomb's MX-5

Post by jonescg » Sun, 04 Nov 2018, 20:47

Yes stray capacitance can be a hazard. Not because it will shock you (although that can be unpleasant) but because you might flinch and drop a tool somewhere you shouldn't...
How long until you pull all of Mexy's cells out and put some NMC cells in? You can probably get some impressive range out of it for the same mass of cells.
AEVA National Secretary, WA branch vice-chair

User avatar
weber
Site Admin
Posts: 2485
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Re: Weber and Coulomb's MX-5

Post by weber » Mon, 05 Nov 2018, 06:56

Hi Chris. That's true about the surprise factor being a hazard. But of course it shouldn't matter if you drop a tool. There is no excuse for using uninsulated tools on batteries. Even if you're unwilling to spend the money on a proper insulated socket and handle (click the image for an eBay search on: insulated socket 3/8"),

Image

it doesn't take long to wrap a tool in two layers of electrical tape.

I also recommend one of these digital torque adapters.

Image

I have successfully modified one of these for use with insulated sockets, by carefully cutting away part of the case on the male side with a 20 mm holesaw and stabilising the shaft inside with neutral cure silicone, and insulating the female side with a short piece of 32 mm corrugated conduit siliconed in place.

Yes, NMC is more energy-dense, but it also burns much more fiercely than LFP in the event of BMS failure (or human error). :o And all electronics fails (and humans make mistakes) eventually.

Mexy contains such a huge investment of hours in battery box design (thanks Mike and Jeff), not to mention fabrication—10 boxes squeezing cells into every available space, earning her the "Tardis" epithet—and the 218 CMUs ($5000 replacement cost), that any replacement cells need to reuse the existing boxes and CMUs if possible. So I'll be looking at how these 16 UBetter's perform over the next few months to a year, and may replace all my cells with those if they perform adequately (assuming they're still available).

If anyone knows of any other prismatic cells in the 40 Ah to 70 Ah range that are less than 117 mm wide and less than 190 mm high, of any chemistry, please let me know.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

User avatar
jonescg
Senior Member
Posts: 2368
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

Re: Weber and Coulomb's MX-5

Post by jonescg » Mon, 05 Nov 2018, 08:25

I use my insulated tools regularly :)
96s2p nearing completion.jpg
96s2p nearing completion.jpg (200.04 KiB) Viewed 262 times
AEVA National Secretary, WA branch vice-chair

User avatar
weber
Site Admin
Posts: 2485
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Re: Weber and Coulomb's MX-5

Post by weber » Mon, 05 Nov 2018, 08:33

Of course I knew you would use insulated tools (nice looking pack). My use of "you" was aimed at the general reader. :)
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

User avatar
weber
Site Admin
Posts: 2485
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Re: Weber and Coulomb's MX-5

Post by weber » Tue, 06 Nov 2018, 18:51

Here are the 16 new cells installed, with shims, in the battery box Coulomb calls "the neck breakers" (it mounts through the parcel shelf, on the back of the rollbar). The drilling out of the cell high-current links, and the CMUs, from 6 mm holes to 8 mm holes was done with a step drill. This produces a much cleaner result than a twist drill when enlarging existing holes in thin material.

IMG_1983b.jpg
IMG_1983b.jpg (189.48 KiB) Viewed 223 times
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

T1 Terry
Senior Member
Posts: 725
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Weber and Coulomb's MX-5

Post by T1 Terry » Wed, 07 Nov 2018, 07:48

Is there enough clearance each side to add a piece of the barbeque plate Teflon sheet to act as an insulator so you avoid any nasty experiences or alarm trips?

T1 Terry
Green but want to learn

User avatar
Johny
Senior Member
Posts: 3705
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Re: Weber and Coulomb's MX-5

Post by Johny » Wed, 07 Nov 2018, 12:50

Good to hear that you have a solution guys. My headways have lost about 20% in the last year. July 2017 the degradation was hard to measure but mine are just over 10 years old too. No particular cells down but all 384 cells are at pretty much the same temp. daily.
Not a bother as range is still fine for me.

User avatar
weber
Site Admin
Posts: 2485
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Re: Weber and Coulomb's MX-5

Post by weber » Thu, 08 Nov 2018, 07:30

Many thanks to Jeff Owen for his help yesterday, in using the engine crane to remove, modify and replace 3 more battery boxes. This leaves only 2 of the worst 16 cells not replaced. As they were only the 11th worst and 16th worst cells, and each is the only one in its respective battery box, I will leave them for now, and see how I go.

T1 Terry wrote:
Wed, 07 Nov 2018, 07:48
Is there enough clearance each side to add a piece of the barbeque plate Teflon sheet to act as an insulator so you avoid any nasty experiences or alarm trips?
That's a brilliant idea. Yes, there is enough clearance. I didn't know that stuff existed, but I see now that it's available from Bunnings and Woolworths at $7 for roughly an A3 sized sheet.

Unfortunately your suggestion was too late for this round of cell replacement, but I will keep it in mind for the future. Thanks Terry.

Hi Johny. It's good that we both got the 10 years we were promised, with the understandable exception of my hot spots, and with the capacity-loss and resistance-increase being somewhat higher in Queensland, as expected from Arrhenius' law.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

T1 Terry
Senior Member
Posts: 725
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Weber and Coulomb's MX-5

Post by T1 Terry » Thu, 08 Nov 2018, 10:53

Can't take the credit for that one, a member of the South Australian branch used it to insulate his (Winston I think) aluminium cells from each other in his PHEV battery pack for his Prius. He made a 3 sides envelope so each pairr were insulated from the bottom, short sides and the one long side, the next pairr provided the insulation between itself and the previous cell pair.
I'm watching with interest to see how your cells and those Evric is using in his battery upgrade hold up compared to the LYP Winston cells. I want to either upgrade the capacity in my PHEV Prius I bought recently, or building a pack for the taxi Prius we use as our business run around car. Eric says the smaller physical size compared to the 90Ah Thundersky cells he is replacing allows for double the capacity at a price close to the same as the Thundersky replacements, that got me very interested :lol:

T1 Terry
Green but want to learn

T1 Terry
Senior Member
Posts: 725
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Weber and Coulomb's MX-5

Post by T1 Terry » Thu, 08 Nov 2018, 11:00

Still trying to get my head around what is actually causing the loss of capacity. The house battery packs I've tested recently still have over 100% of the advertised capacity, one 200Ah pack is over 7 yrs continuous service and the other over 5 yrs continuous service. The 5 yr old pack has spent quite a bit of its life in far nth Qld and over some of the roughest roads you can imagine, so vibration isn't causing a problem and neither is the heat apparently because the couple that own the van think anything under 36*C is coat and long pants weather

T1 Terry
Green but want to learn

User avatar
weber
Site Admin
Posts: 2485
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Re: Weber and Coulomb's MX-5

Post by weber » Fri, 09 Nov 2018, 06:11

T1 Terry wrote:
Thu, 08 Nov 2018, 11:00
Still trying to get my head around what is actually causing the loss of capacity. The house battery packs I've tested recently still have over 100% of the advertised capacity, one 200Ah pack is over 7 yrs continuous service and the other over 5 yrs continuous service. The 5 yr old pack has spent quite a bit of its life in far nth Qld and over some of the roughest roads you can imagine, so vibration isn't causing a problem and neither is the heat apparently because the couple that own the van think anything under 36*C is coat and long pants weather
Hi Terry. 100% of advertised capacity doesn't tell us what percentage of original capacity is left, which is what I am referring to (based on range in km). How many weeks per year does the 5 year old pack spend in far north Qld? Where does it spend the rest of its time?

The temperature in Brisbane is on average about 5 degrees higher than in Adelaide, so based on Arrhenius law alone, we should expect it would take 14 years in Adelaide to produce the same degradation as 10 years in Brisbane. But of course there are also differences in usage between traction packs and house packs, and higher probability of low-capacity outliers (that determine the capacity of the whole pack) with 218 cells in series versus 16 cells in series.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

User avatar
weber
Site Admin
Posts: 2485
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Re: Weber and Coulomb's MX-5

Post by weber » Mon, 12 Nov 2018, 19:33

Many thanks to Mike Van Emmerik (Coulomb) for his help on Friday. With only 14 of the 16 worst cells replaced, we did a range test and found that it had gone up from 65 km to 78 km (to turtle mode in both cases). Then we examined the data logged by the CMUs and found that the cells that had gone to the lowest voltage on the trip, were now in the battery box under the bonnet (over the motor) in a row that Jeff and I had not touched on Wednesday (cells B32 and B33). Fortunately it was a row where it was possible to loosen the clamp rods without removing the battery box from the vehicle. So we swapped those cells out and hopefully gained another couple of kilometres. So overall, by replacing the 16 worst cells, the range increased from 65 km to probably 80 km. Of course that's a lot less than the original 130 km, but it makes it possible for me to drive to a new work location at Ipswich (and back).
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

Post Reply