Page 4 of 68

Weber and Coulomb's MX-5

Posted: Mon, 06 Apr 2009, 23:29
by weber
Here we were, thinking the owner of that white MX-5 above was a bit naughty, cutting into a cross-member. Andrew Schwarz was doing the right thing, even leaving the spot-welding flanges beside the rails and cross members. Then he invited his mate "Dave" over to help, and Dave just cut them all away and whacked out the whole cross-member!

            Before Dave.      (Click to enlarge)     After Dave.

Image    Image

<Gulp> ... I've seen grown MX-5's gnaw their own wheels off rather than talk to "Dave". We sure won't be letting "Dave" anywhere near our MX-5. Image

Weber and Coulomb's MX-5

Posted: Mon, 06 Apr 2009, 23:44
by weber
EV2Go wrote:I'm lost... you wouldn't ever use a drop saw to cut a piece out of the boot it is designed to be used on solid surface like the floor or a bench.

Quite right. Coulomb read the safety brochure I posted on angle-grinders and saw that the overriding consideration is not to use them with cutting disks (as opposed to grinding) if there's any other option (because they are so bloody dangerous).

There is in fact an alternative for car body work, even in quite tight spaces. The sabre saw. It produces a far cleaner cut too. Here's one being used by Andrew Schwartz on his Miata (MX-5), before "Dave" got hold of it.

Image

(Click to enlarge)

Weber and Coulomb's MX-5

Posted: Mon, 06 Apr 2009, 23:47
by juk
Might as well bolt an outboard to the back of Andrew's; it might as well look like a boat if it's going to handle like one.

Weber and Coulomb's MX-5

Posted: Tue, 07 Apr 2009, 00:07
by coulomb
EV2Go wrote:I'm lost... you wouldn't ever use a drop saw to cut a piece out of the boot

Well, exactly. It's just that the West Australian Government article linked by Weber above said that angle grinders should never be used for cutting, always use a special purpose cutting tool like a drop saw.

I guess they weren't thinking EV conversions when they were writing that page...

Weber and Coulomb's MX-5

Posted: Tue, 07 Apr 2009, 05:22
by woody
Nibbler? Gas Axe? Hacksaw? Acid? Thermite? :-)

Weber and Coulomb's MX-5

Posted: Tue, 07 Apr 2009, 14:41
by EV2Go
coulomb wrote:
EV2Go wrote:I'm lost... you wouldn't ever use a drop saw to cut a piece out of the boot

Well, exactly. It's just that the West Australian Government article linked by Weber above said that angle grinders should never be used for cutting, always use a special purpose cutting tool like a drop saw.

I guess they weren't thinking EV conversions when they were writing that page...


Nah that’s crap I could cut delicate ice sculptures with an 4” angle grinder, it’s not the tool, it’s the tool behind the tool where it all goes wrong...

Some people have absolutely no forethought when it comes to cutting things. Things like brake lines behind the panel they are trying to cut, or the effects of the heat the put into the metal that is in close proximity of plastic and other easily melted items, little matter of fires you might start etc.

Angle grinders are as safe as houses if treated with the proper respect they deserve.

Weber and Coulomb's MX-5

Posted: Tue, 07 Apr 2009, 16:25
by weber
It turns out that "Dave" wasn't such a bad guy after all. He didn't cut off the spot-welded flages as I had thought, but bent them upward. And he at least partly replaced the cut-out cross-member with a piece of angle welded in further back. You can see these things in this photo.

Image
(Click to enlarge)

We'll run this by our engineer.

But yes, what a boat. All that lead above and behind the back axle. I can hear the poor MX-5 groaning in pain. We should only have about half that weight at the back, with the LFPs.

Weber and Coulomb's MX-5

Posted: Tue, 07 Apr 2009, 16:35
by acmotor
I do second that EV2Go. Image
Common sense, experience and operator skill prevail.

You can see the safety people have a big task though.

For instance a cutting disc (thin, parallel sided) should never be used for side load grinding, it is not built for it and side wear removes the reinforcing webbing and wheel thickness resulting in failure.
A grinding disk is more tollerant, although should only be loaded on the edge or face, never the back. Any wheel damaged in a drop or a jamb resulting in twisting should be replaced and the old wheel broken to prevent accidental re-use.
You can cut with either wheel but only clean up edges with a grinding wheel.
A grinding wheel produces a wider, hotter cut than a cutting wheel.
(and is using more mechanical/electrical power)
(Ummm, most angle grinders are AC/DC universal motors and would run off the EV battery pack !)

Then there are the different grades and compounds of disc abrasives suitable for Ali, steel, ceramics, masonary, plastics and so on in both cutting and grinding format for angle grinders. My fav. is the diamond disk I use on bathroom tiles. Cool, clean cuts like butter.

How did all that get started ? Oh,I remember, safety boots ! Image

Nice red weber ! Image

Weber and Coulomb's MX-5

Posted: Tue, 07 Apr 2009, 17:01
by weber
acmotor wrote:Nice red weber ! Image

If you mean the painted battery racks in the above images, those are not ours. We're not that far along yet. They belong to Andrew Schwarz's MX-5 in the US.

Weber and Coulomb's MX-5

Posted: Tue, 07 Apr 2009, 17:10
by acmotor
... and there I thought you had been onto photo shop. Image

Ok, that's good, cause I would not pass the welded angle iron that runs between the "chassis" rails. It will suffer fatigue failure as it pretends to control the chassis dynamics. (particularly with a hundred kg of cells on it).

Weber and Coulomb's MX-5

Posted: Tue, 07 Apr 2009, 18:12
by weber
ACmotor, there is a very substantial subframe bolted under those rails, hidden by the newspaper. You can see it in this picture.

Image     (Click to enlarge)

So it may not be as bad as you thought in regard to suspension dynamics. But I still tend to agree with you.

In regard to tools. It's nonsense to say it's only the user that makes it dangerous. Even the best of us sometimes make mistakes, have a bad day, get distracted momentarily, make a wrong judgment. Angle grinders are more dangerous than many other tools because they are fairly unforgiving of such mistakes. Google "angle grinder accidents" if you have a strong stomach.

Weber and Coulomb's MX-5

Posted: Tue, 07 Apr 2009, 18:50
by coulomb
acmotor wrote:Ok, that's good, cause I would not pass the welded angle iron that runs between the "chassis" rails. It will suffer fatigue failure as it pretends to control the chassis dynamics. (particularly with a hundred kg of cells on it).

Is that basically because it will deflect so far, having relatively low cross section compared to its length? Plus, boxes are (presumably) stiffer and stronger than L shapes of the same cross sectional area?

I'm guessing that an actual battery box with a steel floor could be as strong as the original... would you agree? I must have done exercises like this at Uni, but that was a few years ago now.

Weber and Coulomb's MX-5

Posted: Tue, 07 Apr 2009, 19:01
by acmotor
Are you suggesting that a newspaper hides the real story ! Image

Re tools..
No no, I thought I was saying that it is the user that makes it safe.
That includes the user chosing the right tool or at least having the skill / judgement to manage the tool.
e.g. ladder under the tilted vehicle. Knowledge (calculation and measurement) and care means it is up to the task. (not suggesting a new standard here !) In this case I am not certain there is an MX5 45° side tilt stand available from Auto Cheap !

But beware the kitchen knife in the wrong hands !

BTW, is the suspension rubber mounted to the body ? Sorry to show my ignorance of the MX5.

Weber and Coulomb's MX-5

Posted: Tue, 07 Apr 2009, 19:37
by weber
acmotor wrote: Are you suggesting that a newspaper hides the real story ! Image
Good one.
Re tools..
No no, I thought I was saying that it is the user that makes it safe.
That includes the user chosing the right tool or at least having the skill / judgement to manage the tool.
Sorry acmotor. I was referring to EV2go's comment "...it’s not the tool, it’s the tool behind the tool where it all goes wrong..."
BTW, is the suspension rubber mounted to the body ? Sorry to show my ignorance of the MX5.

No need to apologise. I would not expect anyone to know such things. I didn't until a few weeks ago.

If you mean the tops of the coil-over-shock-absorber units, then yes there's a rubber mount there. If you mean the rear subframe that carries the diff and wishbones and ultimately the wheels and the lower ends of the coilovers, then no. It is solidly bolted to the monocoque chassis rails by three 19 mm high tensile bolts per side, two of which go all the way through from the top. There is about a 3 mm sheet of rubber between the monocoque and the subframe, but I don't suppose that's what you had in mind.

Weber and Coulomb's MX-5

Posted: Tue, 07 Apr 2009, 19:52
by weber
A clever suggestion for weighing a car, from my friend Mike Stasse:

Paint the tyre with something so it leaves an imprint of its contact-area on something. Estimate that area, measure the tyre pressure, and use the following formula.

mass(kg) = pressure(Pa) * area(m^2) / gravity(9.8 N/kg)

Not sure how accurate it would be. You'd have to let the jack down very gently.

Weber and Coulomb's MX-5

Posted: Tue, 07 Apr 2009, 20:07
by Electrocycle
if anyone in sydney wants to weigh cars I know a workshop with proper scales, and I may be able to borrow a set if necessary :)

Weber and Coulomb's MX-5

Posted: Tue, 07 Apr 2009, 20:52
by acmotor
coulomb wrote: Is that basically because it will deflect so far, having relatively low cross section compared to its length? Plus, boxes are (presumably) stiffer and stronger than L shapes of the same cross sectional area?

I'm guessing that an actual battery box with a steel floor could be as strong as the original... would you agree? I must have done exercises like this at Uni, but that was a few years ago now.


The chassis will flex as will the subframe. The angle iron cross bar will fight this and lose. It is too rigid axially. Failure will be at the welds. IMHO. Don't underestimate chassis flex.
The interaction between the cross bar and the battery box is another matter. Deflecting the cross bar will induce significant axial loads so the battery box would itself need to be rigid so why then fit the bar ?

Weight is all important. I would hope you plan aluminium for as many add ons to the vehicle as possible including the battery box.
If it is OK structurally by your mechanical signatory to remove the boot floor, then replace it with as light as possible battery box supported as much as possible by a hanging action off the 'chassis' rails and don't try to make the box structural. (unless you have be requested to do so ?) I am interested here.



Weber and Coulomb's MX-5

Posted: Fri, 10 Apr 2009, 02:52
by coulomb
We bought one of these today:
Image
Note near the top right: "with electric brake": it even has regen!!! Image

It's fitted with hacksaw blades, of course, for cutting sheet metal. It made rather quick work of the remaining cuts that we dare to do before we see our engineer (after Easter).
Image

It was a bit awkward to use in tight spots, and when the clearance was low, but with a bit of practice even those weren't bad at all. It could also even out some of the worst angle grinder cuts. We did need the angle grinder to get started in several places.

We ended up with this:
Image    Image

Weber was happy with the results:
Image

The shiny metalwork that protected the fuel tank area can be reused to protect the mid battery racks:
Image

Weber and Coulomb's MX-5

Posted: Fri, 10 Apr 2009, 05:40
by coulomb
A possible battery layout, assuming 208 Thunder Sky 40AH:

Image

The big concern is the front end, particularly the lower row of 14 cells on each side. It mostly depends on how much we can move around things like air conditioning pipes to make room. We've really set ourselves a challenge wanting air conditioning and power steering on the first conversion.

Edit: the background graphic is from one of several MX-5 models available for free from [URL=http://sketchup.google.com/download/ Google Sketchup 7]. It's an amazing program (from just playing with it for half an hour), and I think it might be useful for answering "what if" questions about battery layout, in three dimensions.

To get the graphic, I just rotated a model into approximately plan view, and did a "print screen" to the clipboard. I then used Gimp (a free paint program for Linux and Windows) and increased the brightness to wash it out.

Edit: the 56 batteries in the front is just not going to happen, sadly. We're considering other options.

Weber and Coulomb's MX-5

Posted: Fri, 10 Apr 2009, 09:15
by Mesuge
Hm, just realized that these UK made "Parker SSD/Eurotherm" VFD drives, when fitted with comm., cloose loop, and breaking boards are not that much competetive anymore to the other brands we discussed in the AC thread. I'm talking about the power range of ~45kW nominal. Perhaps for your higher output ~100kW range it can stay bellow Danfoss and other top brands to make a difference.

Weber and Coulomb's MX-5

Posted: Sat, 11 Apr 2009, 03:44
by coulomb
I made a minor discovery today, which may have application to vehicles other than MX-5s:

Image   Image   Image

At left is the right hand rear tiedown, in its normal position. Note the "R" stamped on the metal near the middle. At middle is the view from the opposite side of the rear post that the tie-down is mounted to. Note how symmetrical the bolts are, and the two holes that line up with the bolts that hold the tie-down to the post.

It turns out that things are symmetrical enough that the rear tie-downs can be swapped, so that the "left" tie down can be bolted on the outside of the post (and the same on the other side of course). The last picture shows this. I used an air conditioning bolt, which is too long, to prove that a 65mm bolt would fit and thread into the captive nut on the inside of the post. The bottom hole needed a little filing for the bolt to fit; the top one fits cleanly. I wonder if this swap capability was designed or accidental.

The advantage of this is that space for batteries or the controller are reduced by some 25mm in the crucial under-the-boot space (the largest space in the MX-5 for such items). Sometimes, you really need that space, and if approved, this modification claims back that space without losing the rear tie-downs. These tie downs are useful for towing, but also save the back of the vehicle from scraping expensive parts. (The tie-downs have marks suggesting many scrapes over the years. Better they scrape than other things!)

We'll run this past our engineer soon and see if he salutes it.

Edit:
Postscript. This turns out to be a bad idea, because of the captive nut on the inside of the post. It's welded to the post for convenience, not strength. So when the bolt is put on the other side of the post and tightened, it tries to pull the nut off the post. Certainly, this is not strong enough for towing (even though the rear tie downs have 2 other bolts). We're just going to leave off these tie downs now, as they get in the way of the latest rear battery box design.

Weber and Coulomb's MX-5

Posted: Sat, 11 Apr 2009, 05:21
by woody
coulomb wrote: A possible battery layout, assuming 208 Thunder Sky 40AH:
I second the vote for Sketchup: Image (old old image of my cramming efforts)

Do you need / want the range that 40Ah TS will provide? Or are you just going for them because of the current required?

If you just want 100A+ continuous, then smaller capacity, high output could be the go, e.g. EVPST 7AH, 8AH, 10AH which will all do 15C continuous + double that for 10 seconds.

Range isn't that big a deal for me (Carry an extension lead and a smile :-), so 224 EVPST 7AH will give me ~50km range, fit in the hole where the fuel tank was, and deliver 210Amps for a quick blast.

For me the advantages are lighter weight, lower total cost, smaller size.
Disadvantages: unproven brand, high price per A/H (~US$2.55/Ah FOB Ghanzhou).

cheers,
Woody

Weber and Coulomb's MX-5

Posted: Sat, 11 Apr 2009, 06:21
by weber
woody wrote:Do you need / want the range that 40Ah TS will provide? Or are you just going for them because of the current required?
Hi woody,

We want 240 A for 15 seconds every now and then. That's 6C for the 40 Ah cells. TS specify that their cells can do 10C for 8 seconds (but it shortens their life). This 240 A is for that "quick blast" (of around 100 kW at motor shaft) that you speak of. While it would be nice to also have a range of 120 to 160 km driving normally, we could maybe live with half that. But being a sports car, you kinda want it to make it from Brisbane to say Caloundra or Tweed heads or Toowoomba on one charge if possible. Hence the 120 km minimum.
If you just want 100A+ continuous, then smaller capacity, high output could be the go, e.g. EVPST 7AH, 8AH, 10AH which will all do 15C continuous + double that for 10 seconds.
Yikes! I find it hard to belive there are cells that can do 30C without suffering damage from internal hotspots. I'm not familiar with EVPST cells. Can you point me to some test data or manufacturer's specs?

We'd want at least 30 Ah with 8C for 15 seconds. Hence we plan to test a China HiPower cell against a ThunderSky.

I'm interested in how you arrived at the figure of 224 cells, and acmotor's 220. It seems to me that with a 750 V max DC bus voltage (or is it 780 V?) you wouldn't have much headroom for regen with that many cells. We've been working on a figure of 208 cells.

Weber and Coulomb's MX-5

Posted: Sat, 11 Apr 2009, 06:37
by antiscab
with 208 cells, at 6C you will have a bus voltage of around 500v (when the cells are around 30 deg C)
meaning your max motor side voltage is 353vac.
the higher the voltage, the higher the rpm at which you can get maximum torque.

dont regen to a cell average of any higher than 3.6v. otherwise some cells will experience overshoot.
the 40AH cells on my bike, while accepting 40A of regen, only rise to 3.37vpc average, so a high head room definately isnt needed.

Matt

Weber and Coulomb's MX-5

Posted: Sat, 11 Apr 2009, 07:56
by Mesuge
Yes, according to "Parker SSD/Eurotherm" VFD manual, there is defined limit of 750VDC for the DC bus, so 208x cells @ 3.6V, allowing for regen safety overhead looks reasonable, also there might be hidden few % extra as safety margin beyond that in the drive. Btw. I think only the Danfoss brand can go up to ~800VDC, that's at least what I recall from the pre-reviewed selection of VFDs we have discussed here so far. In addition, some excess juice can be burned via breaking resistors on the DC bus, but this applies only to specific cases like driving downhill (brutal long slopes) with full batt. pack just after charging..