Weber and Coulomb's MX-5

Post up a thread for your EV. Progress pics, description and assorted alliteration
User avatar
acmotor
Senior Member
Posts: 3614
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Weber and Coulomb's MX-5

Post by acmotor »

Agreed on your final thinking on regen.
Man if the last application of brakes at the end of a long trip caused a bypass I'd be happy !!

Not certain I like the idea of expecting the cells to all go OV at the same time/amount at the end of a recharge. All 220 of them.
converted RedSuzi, the first industrial AC induction motor conversion
on to iMiEV MY12 did 114,463km
now Tesla Model 3, 4/2021 MIC pearl white
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by coulomb »

acmotor wrote: Not certain I like the idea of expecting the cells to all go OV at the same time/amount at the end of a recharge. All 220 of them.

We're not certain either!   Image   We may have to apply extra current only to the group where the cell(s) of interest reside(s).

The problem is, without an extra daisy chain line, the only "signal" we can send to all cells that they aren't likely to receive in actual driving is overvoltage. That's assuming a non-microprocessor based BMS.

Maybe people can convince us that they are just as reliable, and we can switch to such a design.
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by coulomb »

me wrote: The second problem is more serious. When the bypass current comes on, the change in comparator voltage (despite bypass caps everywhere) seems to trigger the latch to reset itself, at least on our breadboard.

In fact, it was only on our breadboard. Putting a 2200 uF (it was to hand) across the power supply output made the problem go away. Perhaps the power supply (an Electronics Australia kit) had overshoot when the bypass resistors come on, and the capacitor absorbs the overshoot.

We're hoping that on a PCB right on top of the cell, this won't happen. But of course, there are all sorts of transients in an electric vehicle, so time will tell. At least it's a "convenience function", not essential to the cells' welfare.
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
User avatar
acmotor
Senior Member
Posts: 3614
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Weber and Coulomb's MX-5

Post by acmotor »

There are pros and cons for analogue/discrete vs uP. We can all list them and many have been discussed on this thread.
I'd be happier with KISS unless going Nevilleh's 8 cell uP direction.
It comes down to cost (and build time) and what you are most comfortable working with. (and what you really need out of your BMS)
I have made numerous projects using Pic,atmel and silicon labs uP embedded systems using C code and can run a web page at each cell if you like, but cost and overkill are a concern. So I would not discourage your direction.

Remote readout of voltage is the biggest win for uP BMS IMHO.
That is not to encourage a uP vs analogue debate, just my comment.

Re the reset. Just thinking. If the 'visual' latch is for service / diagnostics when you have the battery pack open, then a tiny membrane switch or a magnet / hall effect could be used as reset.... Or be committed to the Rx opto in a uP system. Bidirectional comms would be advisable with reprogram option via serial with a uP system anyway. More wires remember. Everything is x 220 in my case !

edit: spelling... at least the mistakes I noticed Image
Last edited by acmotor on Fri, 28 Aug 2009, 17:42, edited 1 time in total.
converted RedSuzi, the first industrial AC induction motor conversion
on to iMiEV MY12 did 114,463km
now Tesla Model 3, 4/2021 MIC pearl white
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by coulomb »

acmotor wrote: Re the reset. Just thinking. If the 'visual' latch is for service / diagnostics when you have the battery pack open, then a tiny membrane switch...

Good idea. We were going to use the "short R6 to reset" technique, but most of the components are inaccessible under the bypass resistors, and besides most of the resistors are in arrays now, where it's much easier to short to the neighbouring resistor than across one resistor.
It's going onto the PCB design right now (between the pigtails of two bypass resistors).

I was saying that the overvoltage only has to be a few seconds; well really it only has to be milliseconds, say one mains half cycle (10 ms). Not all of them need resetting, and the ones on cells that are weak (and hence undervoltaged) will probably reset first, since they will have higher internal resistance. Well, internal resistance or wiring/lug resistance; either way they are signalling attention.

The fact that some cells may end up with higher resistance, and hence a higher voltage drop while charging and regenerating, indicates to us that perhaps it may be important to keep the reset voltage higher than the bypass voltage.
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
User avatar
acmotor
Senior Member
Posts: 3614
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Weber and Coulomb's MX-5

Post by acmotor »

It is likeky the uP versions of BMS will require some form of reset option (other than WDT) or reset via (isolated) programming port. but even that does not always work from experience. A hardware reset or power off is sometimes required. Every uP device I can think of has either a reset button or ability to remove power. This may be hard with a bolt on BMS.
Sorry, of track.
converted RedSuzi, the first industrial AC induction motor conversion
on to iMiEV MY12 did 114,463km
now Tesla Model 3, 4/2021 MIC pearl white
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by coulomb »

The PCB has been sent to BatchPCB. US$37.41.

Image

[Edit: updated the image to emphasise the cutouts for high voltage isolation.]

It gets panellised with many other designs, so we will have a long wait to get our prototype.

The bypass resistors fit over the top of most of the other components; their pigtails fit horizontally into the four large pairs of donuts, with the yellow circles around them. (The yellow circles are the size of ceramic standoffs, so this is a sort of keep-out zone for components).

The heavy tracks on the top (red) at the left art for the thermistor and its transistors; they attempt to detect over temperature of the negative terminal.

It has over and under voltage, and a latch for undervoltage or over temperature. There are monitor leads at the bottom, via 1W resistors for safety and voltage capability. That way we can measure individual cell voltages either at a terminal block with a multimeter, or by some as yet undesigned monitoring system.

But now we concentrate on battery racks.
Last edited by coulomb on Tue, 01 Sep 2009, 08:04, edited 1 time in total.
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
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 »

acmotor wrote: It is likeky the uP versions of BMS will require some form of reset option (other than WDT) or reset via (isolated) programming port. but even that does not always work from experience. A hardware reset or power off is sometimes required. Every uP device I can think of has either a reset button or ability to remove power. This may be hard with a bolt on BMS.
Sorry, of track.


How do you live with that paranoia   Image
You could have a simple RC (or is it RL I forget) circuit that is pulsed by the uP to keep it charged. If the charge drops the reset is activated.
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 »

coulomb wrote:

The bypass resistors fit over the top of most of the other components; their pigtails fit horizontally into the four large pairs of donuts, with the yellow circles around them. (The yellow circles are the size of ceramic standoffs, so this is a sort of keep-out zone for components).


How are you going to hold these to prevent vibration related failure of leads & joints?
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by coulomb »

Oh, and the shape of the board (designed for 40 Ah Sky Energy) will tile, hopefully allowing more PCBs per panel and reducing the final PCB cost.
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by coulomb »

Squiggles wrote: How are you going to hold these to prevent vibration related failure of leads & joints?

Good question. We were hoping that two ceramic standoffs would be enough to couple them mechanically to the board and hence to the cells.

There is a gap between the resistors for air flow, I think we'd want to retain that. If needed, we could possibly design a clip that held them, did a little heat sinking, yet allowed most of the air flow.
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
User avatar
Johny
Senior Member
Posts: 3749
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Weber and Coulomb's MX-5

Post by Johny »

Squiggles wrote:How are you going to hold these to prevent vibration related failure of leads & joints?
Epoxy 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 »

coulomb wrote:
Squiggles wrote: How are you going to hold these to prevent vibration related failure of leads & joints?

Good question. We were hoping that two ceramic standoffs would be enough to couple them mechanically to the board and hence to the cells.

There is a gap between the resistors for air flow, I think we'd want to retain that. If needed, we could possibly design a clip that held them, did a little heat sinking, yet allowed most of the air flow.


We commonly used a little bead of neutral cure silastic to add rigidity, not sure how that goes with heat though.

Edit: apparently it is good for 200+C, how hot will the resistors get?
Last edited by Squiggles on Tue, 01 Sep 2009, 06:24, edited 1 time in total.
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by coulomb »

Squiggles wrote: We commonly used a little bead of neutral cure silastic to add rigidity, not sure how that goes with heat though.

Edit: apparently it is good for 200+C, how hot will the resistors get?

They'll get to less than 110°C, so wire insulation won't melt. So that should be good. It turns out that Weber had silicone in mind all along; good to hear that it's been used before.
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
User avatar
acmotor
Senior Member
Posts: 3614
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Weber and Coulomb's MX-5

Post by acmotor »

10W ww ceramic in free air at 10W input, ambient 21°C, I just measured 271°C with IR therm.
That's why I derate considerably for the BMS.   2W keeps it under 100°C
Image

Squiggles... paranoia ? nope... reality. Image
converted RedSuzi, the first industrial AC induction motor conversion
on to iMiEV MY12 did 114,463km
now Tesla Model 3, 4/2021 MIC pearl white
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by coulomb »

For completeness, here is the latest schematic:

Image

Some of the craziness of the circuit is to keep Protel happy, and will check that the circuit and layout agree. The many labels are net names, which appear on PCB pads and vias while routing.

We're down to 27 components, from the shameful 42 that we thank TJ for noting   Image .

We saved 8 components by subsuming 11 resistors into 3 quad arrays. Another 6 were saved by leaving off 3 LEDs and their resistors. A further 2 were saved by using single 1 W thru-hole resistors instead of two SMD resistors in series, to get the 500 V working, 1000 V withstand rating for the voltage monitoring resistors. We added an optional push-button switch as suggested, for resetting the latch.

We thank everyone for their helpful suggestions, particularly ACmotor for providing the TLV431-based circuit that was our starting point, and for suggesting the addition of a latch.

We'll post Protel files after the board is fully debugged; this is an open design project. The reason for not posting now is that there is a slight chance that there may be changes before the first production run   Image
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by coulomb »

acmotor wrote:   2W keeps it under 100°C

We'll have a max of 1.9 W in 7W resistors; that's about 27% of rating, or equivalent to 2.7W in a 10 W resistor. So we're not quite as conservative as that, but close.

Edit: speelunk.
Last edited by coulomb on Tue, 01 Sep 2009, 08:02, edited 1 time in total.
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
User avatar
acmotor
Senior Member
Posts: 3614
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Weber and Coulomb's MX-5

Post by acmotor »

So is it 6R8 or 3R9 as in the pic ?
converted RedSuzi, the first industrial AC induction motor conversion
on to iMiEV MY12 did 114,463km
now Tesla Model 3, 4/2021 MIC pearl white
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by coulomb »

acmotor wrote: So is it 6R8 or 3R9 as in the pic ?

We have been testing with a single 3R9; the original design was for 4A of bypassing using 4 x 3R9. However the temperature issue forced us to switch to 4 x 6R8 in production, for ~2.1A of bypass.
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
User avatar
acmotor
Senior Member
Posts: 3614
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Weber and Coulomb's MX-5

Post by acmotor »

I'm happy now. Image

Looking good guys !
converted RedSuzi, the first industrial AC induction motor conversion
on to iMiEV MY12 did 114,463km
now Tesla Model 3, 4/2021 MIC pearl white
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 »

Johny wrote:
Squiggles wrote:How are you going to hold these to prevent vibration related failure of leads & joints?
Epoxy Image


Make sure it is the right type then! Araldite for example goes soft at about 80C.
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 »

coulomb wrote:
Squiggles wrote: We commonly used a little bead of neutral cure silastic to add rigidity, not sure how that goes with heat though.

Edit: apparently it is good for 200+C, how hot will the resistors get?

They'll get to less than 110°C, so wire insulation won't melt. So that should be good. It turns out that Weber had silicone in mind all along; good to hear that it's been used before.


Just make sure you use Neutral cure variety.....learn from the mistakes of others!
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 »

coulomb wrote:
Some of the craziness of the circuit is to keep Protel happy, and will check that the circuit and layout agree. The many labels are net names, which appear on PCB pads and vias while routing.


You can break a net in Protel as long as you get the naming right. Sometimes (often) it helps with the visual layout of the schematic.
User avatar
acmotor
Senior Member
Posts: 3614
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Weber and Coulomb's MX-5

Post by acmotor »

This may have already been covered, but are your inter cell straps/leads above or below the PCB ?
I'm guessing below. All BMS connections are from top layer of PCB ?
Are the cell pads double sided and plated through ? or will the BMS only connect to the cell via the terminal bolts ?
Do you envisage all load current flowing where you intend ?
How close are the through hole components to the large terminal nuts on the cell ?
Sorry if you've covered this already.
converted RedSuzi, the first industrial AC induction motor conversion
on to iMiEV MY12 did 114,463km
now Tesla Model 3, 4/2021 MIC pearl white
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by coulomb »

acmotor wrote: ...are your inter cell straps/leads above or below the PCB ? I'm guessing below.
Yes, below.
All BMS connections are from top layer of PCB ?
They're all through hole plated, but yes, all components and connections are on top of the board.
Are the cell pads double sided and plated through ? or will the BMS only connect to the cell via the terminal bolts ?
Yes, the cell connections are plated through, but we explicitly don't intend for traction current to go through the board and its holes. At one stage, we had a row of holes in case someone misused our board design, but we've removed them. Incorrect assembly would surely become obvious straight away, although the big hole for the terminal bolts might hold the traction current for a few minutes. So only bypass current (2.1 A max) will flow through the terminal bolts and/or pads.
Do you envisage all load current flowing where you intend ?
I'm not sure how to answer that. The intention is for the current to flow where intended, yes Image .
How close are the through hole components to the large terminal nuts on the cell ?
The edges of the bypass resistors are 190 mil (0.19" - sorry, had to do the board in imperial units) from the line between the board proper edge and the "wings". This line is the maximum extent of the cell terminal nut. But there will be straps and/or lugs under there, so the extent of the terminal nuts won't matter. The resistors are centred on the board for that reason. (Plus Weber wanted symmetry.   Image )
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
Post Reply