Weber and Coulomb's MX-5

Post up a thread for your EV. Progress pics, description and assorted alliteration
User avatar
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by weber »

coulomb wrote:I think that carbon motor would reverse spin our "little" motor till the rotor shattered!
I think the CARBON motor's rotor would be damaged first, as it is a much larger diameter and the force is proportional to the radius (and the square of the RPM). I just learned from James that he probably won't be able to test our motor much past 3500 rpm due to the speed limitation of the CARBON motor, whereas our motor is warranted to 4500 rpm and capable of maybe twice that for brief periods when properly balanced.

The CARBON motor has more peak power than ours, but it is a multipole motor designed for use without a gearbox and so has higher torque but lower RPM. The Ultramotive CARBON has peak 150 kW (522 Nm at 2744 rpm), continuous 58 kW (122 Nm at 4540 rpm). The ABB 3GAA 131 008-••E should have approximately (by simple extrapolation when run in delta at the nominal star voltage) peak 135 kW (270 Nm at 4800 rpm), continuous 38 kW (72 Nm at 5050 rpm).
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
User avatar
Geerant
Groupie
Posts: 67
Joined: Mon, 25 May 2009, 20:45
Real Name: Ben
Location: Sydney - West
MSN: geerant101@hotmail.com

Weber and Coulomb's MX-5

Post by Geerant »

My motor is currently with james and will be for the next few weeks. He is currently making an adaptor plate up for it. If you needed to get your motor tested against mine James might be able to sort something out...

Ben


www.evalbum.com/2715
Last edited by Geerant on Sat, 16 Oct 2010, 05:25, edited 1 time in total.
Working on the "Magnatron".

[url=mailto:geerant101@hotmail.com]Email[/url] me for more details.
User avatar
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by weber »

Thanks Ben. Here's a link to your motor specs, for others.
http://www.evo-electric.com/inc/files/A ... t-V1.1.pdf
If James wants to do that, as part of testing and configuring his VF drive, I certainly have no objection. But it sounds like a lot of work.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
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 »

Geerant wrote: My motor is currently with James and will be for the next few weeks. He is currently making an adaptor plate up for it. If you needed to get your motor tested against mine James might be able to sort something out...
That could be interesting. The Carbon motors don't like to go too fast, but yours (being "only" 4-pole) is a better match for our overvoltaged 2-pole.

Wow, interesting conversion. Huge pack, high tech motor and controller... it will be great to see it when completed. Maybe we can even compare performance. Your member's profile says you are in Sydney, but the EV Album page says the conversion is in Brisbane.

If you don't mind me asking, how much was the EVO motor?
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
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 »

Nice looking project Ben can't wait to see it finished, same for the MX5.
Last edited by EV2Go on Sat, 16 Oct 2010, 10:40, edited 1 time in total.
User avatar
Geerant
Groupie
Posts: 67
Joined: Mon, 25 May 2009, 20:45
Real Name: Ben
Location: Sydney - West
MSN: geerant101@hotmail.com

Weber and Coulomb's MX-5

Post by Geerant »

Yeah i like it. Its on the expensive side but in my mind its a quality build, and a dream of mine so the cost didnt matter soo much...

It has been a very long time in the making, spending more than 7 years in the "On paper / theoretical" stage, and going through numerous changes to the motor / battery configuration / and virtually everything haha.

I gave my motor to James last week for testing against the Carbon and another motor he has there aswell. I guess to test yours you would need an adaptor plate to match it to the testing jig.

I live in Brisbane now (Was in Sydney). Been a while since i have been on aeva with the move to Brisbane and getting a new job. I have updated my profile now :)

The motor is priced in GBP (Being it comes from London) and will vary alot with the AUD (I payed for it a few months back now) and it cost $11'000 at the time, with about $1600 worth of GST an the likes to get through customs. Worth every cent in my mind as its the best motor I have ever seen...

I have bought all the parts now and am just in the process of determining where the batteries will all fit and getting the motor coupling / adaptor plate / battery boxed fabricated.
Working on the "Magnatron".

[url=mailto:geerant101@hotmail.com]Email[/url] me for more details.
User avatar
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by weber »

Geerant wrote:I gave my motor to James last week for testing against the Carbon and another motor he has there aswell. I guess to test yours you would need an adaptor plate to match it to the testing jig.
I suspect the other motor you're referring to is ours. It has a Euro standard 132-frame mounting flange, which Tritium already had a mounting plate for. See this post with photo.

forum_posts.asp?TID=980&PID=28337#28337
Last edited by weber on Wed, 19 Mar 2014, 08:12, edited 1 time in total.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
User avatar
Geerant
Groupie
Posts: 67
Joined: Mon, 25 May 2009, 20:45
Real Name: Ben
Location: Sydney - West
MSN: geerant101@hotmail.com

Weber and Coulomb's MX-5

Post by Geerant »

Yeah thats it! i seen it when i was over there the other week :) Nice looking motor
Working on the "Magnatron".

[url=mailto:geerant101@hotmail.com]Email[/url] me for more details.
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 »

Things have been a bit hectic here at the Coulomb Castle, but I thought I should make a brief post about our motor. Finally it went on the dynamometer last Wednesday, and it made 350 Nm. We have been working on 273 Nm (72 * 3.8; I hope those figures are right). When I expressed surprise at this, TJ pointed out that the breakdown torque figure (273 Nm) pertains to direct-on-line (DOL) connections; it's mostly a limitation caused by the field current not being independently controlled to the rotor current. With a controller like the WaveSculptor, there is separate control of these two currents, so this breakdown torque is not a real limitation for EVs.

This possibly means some other conversions might get a pleasant surprise.

Hopefully we'll get some data to massage and publish in the next few days.

TJ calculated that we might get 75-80 kW mechanical from the motor with ~ 390 VDC, though the power (edit: was torque) will peak at around 1800 RPM, giving us roughly constant power above that (see post after the next). Maybe this would not be too bad for direct drive.

With the higher voltage controller, we'd get a bit more than half that torque, out to almost twice the speed (depending on batteries, most likely).

Edit: two very minor edits.
Last edited by coulomb on Sun, 24 Oct 2010, 16:47, 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
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 »

I should have added that for the minute (or was it 90 seconds? sorry, can't remember) of the high power test, the thermistor glued to the windings indicated an increase from ambient (guessing 25°C) to 90°C. Class F windings go to 130°C, and that's for the windings to last 20,000 hours.
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 »

Coulomb wrote: the power will peak at around 1800 RPM, giving us roughly constant power above that...
There are possibly two limits here. When we run out of voltage (say 1800 RPM), we should get roughly constant power and torque at 1/f. That's because flux seems to be proportional to V/f; with constant V, the flux and hence torque has to fall at 1/f.

At some other point, we may get to the maximum slip point, and can't go beyond that (this would be real breakdown torque, even for VFDs, I think). At that point, torque will fall at 1/f², and power at 1/f. Or maybe this maximum slip point is the one that TJ is saying does not exist for VFDs. That would fit what I seem to see with motor curves like the Tesla Roadster's, where power is constant (with a gentle fall possibly due to power factor decreasing) after voltage runs out ("nominal speed").

Perhaps when we get the data we might be able to see what is really happening.

Earlier discussion on this (thanks, Weber): http://www.aeva.asn.auhttp://forums.aev ... 970#p25371

http://www.aeva.asn.auhttp://forums.aev ... 984#p25392
Last edited by coulomb on Thu, 20 Mar 2014, 18:31, 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
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 »

coulomb wrote:Finally it (motor) went on the dynamometer last Wednesday, and it made 350 Nm. We have been working on 273 Nm (72 * 3.8; I hope those figures are right).
Nice. Good news.
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 »

This is the motor data:

Image

This is with a DC bus of 350/340/330 VDC for 100/200/300 A RMS to the motor (they have only a smallish pack on the dynamometer, and it's sagging even though it's only supplying the losses).

The 300 A curve droops at higher speed because voltage runs out, so we see the 1/f torque "constant power" region (even the power droops a little, as you can see).

The speed tops out at around 2000 RPM because the generator it was attached to is a low speed model that they don't like spinning too fast.

With our nominal 365 V pack, we might get the droop start very slightly later, and with it being stiffer, hopefully the droop will not be quite as pronounced. particularly in the power curve.

Of course, if we go with the 900 V controller and a nominal 730 V pack, things will be rather different, and will better match the original ICE engine (86 kW @ 6,800 rpm, and 136 N·m @ 5,500 rpm) (from the Wikipedia MX-5 NA page).

To save people looking up the details, this is our ABB 3GAA131008-ASE 22 kW nominal 2 pole 132 frame AC induction motor running on our Tritium WaveSculptor200 450 V controller.

Edit: Where are my manners? A big thanks is owed to James Kennedy (Tritium James, or TJ in other posts) and the rest of the Tritium staff for the data and the test. This goes way beyond ordinary customer support!
Last edited by coulomb on Sun, 08 Jul 2012, 09:31, 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
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 »

Battery layout.

This post is more or less for our own use, but others may find it amusing how hard we have to work to shoehorn 228 40 Ah CALB cells (splittable into two subpacks of 114) into a compact sports car.

Our original layout for some time (corresponds to the sketchup diagram here:

rear
2 14 2    (may not have to use 2s)
   23       Under boot = 64
   23

.   5   7
3 5    6 3    low = 29

   14
   14           mid = 52
   14
   10

   14       (maybe 15)
   18       neck-breakers = 32

   12
   14       bonnet = 40
   14

   13       (maybe 14) radiator = 13
front
This totals 64+29+52+32+40+13 = 230, two more than needed.

However, when we installed the rollbar, we lost a row of cells, and can only fit one row of 19 cells now. We made up the difference with an extra row under the bonnet. (I'm yet to be convinced that it will really fit.) Alas, another one was lost in the low boxes near the differential; the new diff is a bit larger.

So the latest layout is as follows:

rear
0 14 0    (four spare)
   23       Under boot = 60 (1)
   23

.   5   6
3 5   6 3    low = 28 (2)

   14
   14           mid = 54 (2)
   14
   12           (recently increased from 10)

   19       roll-bar = 19 (2)

   12
   14       bonnet = 54 (1)
   14
   14

   13       radiator = 13 (2)
front

So this gives totals of:
(1) 60+54 = 114
(2) 28+19+54+13 = 114

That looks like only 6 boxes, but the low box is split by the drivetrain, and the mid box is actually two boxes (14+14 and 14+12) bolted together slightly offset.

We want to split up the pack into ELV (<= 120 V) segments. If we take the maximum cell voltage as 4.2 V (we should not in reality exceed 3.65 V), that makes the longest string 120/4.2 = 28 cells long. So the boot box gets split into three segments, we could possibly combine the two low boxes into one segment of 28 cells, the mid and bonnet boxes split into two segments, and we have a total of
(1) 3+2 = 5 segments
(2) 1+1+2+1 = 5 segments (assuming we combine the two low boxes into one segment).

So that comes to 6 contactors per half-pack, with one contactor at the top and bottom of each sub-pack. When we go to the series pack for driving and paralleled pack for charging, we'll need one extra contactor to put the two subpacks in series (and some very carefully planned fool-proof logic so that the packs can never be shorted). There will probably be one more to bypass the precharge resistor.

Edit: removed the [ code ] tags; they just space the text out too much.
Edit2: increased mid box to 54 from 52; fixed some copy and paste errors; changed grouping.
Last edited by coulomb on Fri, 12 Nov 2010, 05:29, 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.
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 »

Wow, 228 cells in an MX-5! I squeezed 135 cells into my BMW and thought that was pretty good, although I could've gotten another 80 or so in if I sacrificed the entire boot space instead of just 1/3. But a boot is more useful than increased range.
I have about 16 units of electricity available (at 80% DOD) and I'm getting 90 - 120 kms range, depending on how much time I spend in the 70 zone!
I reckon you will have about 27 units available and that ought to give you about 160 kms. Maybe more as your MX-5 will be more slippery than the BMW, but you will be heavier too.
Its going to be really interesting to see what it actually does on the road.
Last edited by Nevilleh on Fri, 12 Nov 2010, 02:10, edited 1 time in total.
User avatar
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by weber »

Yes, the battery box that was marked "boot" on the diagram is really under the boot. We have not encroached significantly on the boot space of the MX-5, so you can fit a toothbrush and a change of undies in there just like before. Image
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
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 »

Your change of undies must be pretty "lacy"!
I can get my wife's suitcase in my boot and still have room for a toothbrush and some undies. That means I travel light of course.
What's your projected time for actually driving it now? And what's the weight?
Last edited by Nevilleh on Sun, 14 Nov 2010, 12:48, edited 1 time in total.
User avatar
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by weber »

I'm loath to have further predictions of completion recorded for the amusement of posterity. Predicted GVM is 1453 kg (Tare increased from 970 kg to 1290 kg). The increase is essentially the same as the mass of the batteries, as is commonly noted. This model MX-5 does not have a manufacturer's GVM. Engineer reckons it's OK (just barely). We are upgrading the brakes and will have to do braking tests and swerve tests.
Last edited by weber on Sun, 14 Nov 2010, 13:42, edited 1 time in total.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
User avatar
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Weber and Coulomb's MX-5

Post by weber »

We've come up with a second use for the optionally populated area on our cell-top battery monitoring unit (BMU) PCBs. The first use was for an isolated serial transmit stage for use at the end of a string of cells. The second use is for a differential amplifier to monitor the current through a shunt at the start of a string of cells.

Can anyone spot any potential problems or suggest improvements?

Image

I've left off parts of the schematic that haven't changed, such as the voltage sense, reverse polarity protection, voltage regulator and the bypass resistor circuitry. That can be seen here.

forum_posts.asp?TID=980&PID=25322#25322
Last edited by weber on Wed, 19 Mar 2014, 08:17, 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 »

I thought at first you where just adding the bits to a spare board as you only need one in the circuit.

But your Pin1 uTxm on IC1 has swapped with Byp on Pin6 and the designators are all different. (R8,9,10,11,15 .. and others)

Looks like your putting the CellV straight in with out divider.
(Sorry I'm nosey)

Is its based on this one Rev 32 in your linked post?
Ver 32 BMU

I take it the JP11 is not connected for Current Sense.
It must be for Cell No 1.

200A is 50mV on Shunt so thats 10x gain for 500mV

You could fit a cap on R8 and pull pin to GND for RC filter.
Oops your using VREF+ in measurement.
3.0V/2 =1.50V = +200A
2.5V/2 =1.25V =    0A (Ref+)/2
2.0V/2 =1.00V = -200A

[Edit: circuit looks ok -my stuff up and Eye for detail was fuzzed up]
Your Filter is at 1/T=f=1/RC = 1/(150k x 100n) = 66kHz @3dB
Not sure if my calc is right. Need small signal model.

You may need slower filtering for your AD rate.
I thought very slow like 10 samples a second with filter would work.

What happens if their is 2000A in shunt or wild voltages?
The is it worth clamping Op- and Op+ with diodes.
Probably not worth it with 15k R21 and R22 and MCP6001 has +/-2mA on the pins and ESD ratings on all pins.

So will this be a separate board or more parts as options on the ONE multi-purpose BMU-Link PCB?

Ohhh its just a separate board just for the SHUNT not cell connected.
Duh! [Edit - no, its a new PCB version]
Last edited by 7circle on Thu, 18 Nov 2010, 04:44, 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 »

7circle wrote: I thought at first you where just adding the bits to a spare board as you only need one in the circuit.
Actually, we need 2, one at the start of each sub-pack, and we want to use the link software to measure the shunt current. This circuitry will go on one or two boards per panel, so every 34th or every 17th board will have this circuitry in place of the end-of-row isolation circuit. End of row is only needed at ends of rows, never at the beginning, and the diffamp circuit will be at the start of a string of 8 or 9 BMUs (on one panel). For ~98% of the boards, there will be nothing populated in the diffamp/end-of-row area.
But your Pin1 uTxm on IC1 has swapped with Byp on Pin6 and the designators are all different. (R8,9,10,11,15 .. and others)
Fairly recently, we did a 3-wire rotation of the bypass and two transmit wires. This is so that we could take advantage of an internal inverter, so that the timer could generate differential drive without external components. Those boards are busy enough as it is.

The shunt diffamp output connects to LI, so we can use the link software to measure shunt current.
Looks like your putting the CellV straight in with out divider.
(Sorry I'm nosey)
? There has to be a divider; the reference voltage is 1.5 V, and we want to be able to measure up to 4.096 V.
Is its based on this one Rev 32 in your linked post?
Heh, close. We're on rev 45 now. In particular, the LI (link input) signal is a voltage divider to Vref+ so we can measure positive and negative link voltages. I think it's time for a (hopefully final!) circuit and layout.
I take it the JP11 is not connected for Current Sense.
It must be for Cell No 1.
It wasn't for a while, but Weber correctly insisted that it is, so that when the diffamp circuitry is there, the BMU can still be used as an ordinary BMU.
200A is 50mV on Shunt so thats 10x gain for 500mV
Believe it or not, those shunts are hard to source, at least at a reasonable price. Jaycar has some at a reasonable price, but they are 1mR, which means way too much dissipation (58W) at 240 A. There are 50 mR and 75 mR shunts, but its hard to find someone with them in stock. But we can easily juggle the resistor values for any particular shunt and maximum current. We need to get these things prototyped as soon as possible, since BatchPCB takes weeks. We could pay more for quicker turnaround, but we have so much else still to do that it isn't worth it.
You could fit a cap on R8 and pull pin to GND for RC filter.
Your Filter is at 1/T=f=1/RC = 1/(150k x 100n) = 66kHz @3dB
Not sure if my calc is right.
I think it's 66.7 Hz, not kHz.
The is it worth clamping Op- and Op+ with diodes.
I think that surely the op amp has diodes (possibly parasitic ones) already.
So will this be a separate board or more parts as options on the ONE multi-purpose BMU-Link PCB?
This is a circuit that replaces end-of-row circuitry on some boards of a panel. We have a new Protel database file now that has two schematics and two layouts. When we generate the pannelised artwork, we'll do the juggling to make some of the BMUs specialised for the shunt diffamp. We decided it was simpler that way, and we don't need to carry sensitive analogue inputs that are tied to the negative end of the string off the BMUs and isolate them specially. Well, we'll have to connect the shunt wires specially, and they will need their own connectors and are potentially lethal, but at least they're not MPU inputs. If there is an accident, the opamp will possibly protect the MPU. We hope.
Ohhh its just a separate board just for the SHUNT not cell connected.
As above, no, it's a regular, working BMU that happens to be at the start of a half-pack.
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 »

coulomb wrote: Heh, close. We're on rev 45 now. ... I think it's time for a (hopefully final!) circuit and layout.
Ok, here it is. Rev 45 was just created today by Weber; I haven't even reviewed it myself yet.

Schematics:

Image
   Image

Image

The boards without shunt diffamp circuitry:

Image

Now the rotated version, this time with shunt diffamp circuitry (we may never actually make this combination):

Image

Note the size of C20, C21 (0.1uF, 2%). They're size 1210 (resistors are 0805 by comparison; some designs use 0603 size).

Edit: minor change to the lower layout; I didn't reverse the arrows, and didn't re-pour the earth plane near the middle of the board.
Last edited by coulomb on Mon, 15 Nov 2010, 16:59, 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.
a4x4kiwi
Senior Member
Posts: 773
Joined: Thu, 03 Jan 2008, 19:04
Real Name: Malcolm Faed
Location: Australia
Contact:

Weber and Coulomb's MX-5

Post by a4x4kiwi »

One of these current transducers could simplify things a bit.
http://www.lem.com/docs/products/htfs_sp2_e.pdf

edit. But you would need 5v to supply it.
Last edited by a4x4kiwi on Mon, 15 Nov 2010, 17:34, edited 1 time in total.
Silicon is just sand with attitude.

Blog: http://malfunction.faed.name
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 »

Sorry just some thoughts I had, you might not have considered...

Why do you want so many current sensors are you thinking of charging in groups. Or Series parallel. Lots of options with the design. I like it.

Brushless Fans or a connector like used in PC's may be a cheap reliable option. So a Hbridge could be used.
(ImageAnd used later for 20W Link to AC Balance Power bus. But not fitted till you want it.... I love looking at all the flavours of geloto but never stray to far from my favourites)
[Edit cut my stuff ups to tidy post]
My memory for Small signal analysis is not there.
But if you allow for cap between R9 and R8 - To GND, its a good safe guard

Maybe Hall DC Current Sensor are cheap enough and temperature stale enough to do the job. Shunts are usually cheaper, but use power.

The Shunt is electrically connected to the Cell battery the Module is measuring. It might be worth showing this.

I can see that it can be floating to some extent due to the differential +/- connection. So it could be floating above or below the battery potential by 100V.

They usually show them in the Opamp model, like you I'm assuming they are part of the Die MCP6001 (oooh it's Microchip).
If shunt used. You could allow for Diode Clamping. But not fit.
Or perhaps you have tested it as a prototype already.

Strain Relief for Link wires when fitted

Also P2.6 is tied to oV. Why?
And Pd.2 [Edit P2.7] is NC (... No Connect) it might be worth adding CAP & res foot prints if they are needed in the future.

Also JTAG pins are not used for anything else ... I'm sure I could think of something....

Oh Noticed the Fan Net is only on the FET Q2 Gate

Could be used for Brushless Fan options or HF-Transformer to replace Resistors.

Allowance for External Temp sensor could be a good wish too.
Even as a digital signal not analogue could be valuable.

There's my rant hope its helps.
Image

[edit : comments on RC to freq ]
[Edit - Cut my mess out]
[Edit P2.7 added]
Last edited by 7circle on Sun, 21 Nov 2010, 06:38, edited 1 time in total.
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 »

a4x4kiwi wrote: One of these current transducers could simplify things a bit.
http://www.lem.com/docs/products/htfs_sp2_e.pdf

edit. But you would need 5v to supply it.


This one can run off 3.3V

Allegro ACS758. If they don't work I can use some external ones. They have 120 kHz typical bandwidth


Ricky-NZ is using them on his Awsome Vector Brushless Project

You could request 12 Samples from Allegro. Image
Not sure of price.
Post Reply