Antiscab's NWH10 Prius plugin Lithium conversion

Post up a thread for your EV. Progress pics, description and assorted alliteration
antiscab
Senior Member
Posts: 2700
Joined: Mon, 26 Nov 2007, 05:39
Real Name: Matthew Lacey
Location: Perth, WA

Antiscab's NWH10 Prius plugin Lithium conversion

Post by antiscab »

updates:

I'm going to start referring to the prius with the GBS 20Ah cells as prototype 1.

I'm having heat issues with the battery (mainly because it is too small)
I'm finding the car's original cut backs for temperature are actually about right for LiFePO4.

When the SOC is set at 100%,
at 40 deg C, regen gets cut back to basically 0
at 50 deg C, all discharge from the battery ceases

allowing for a 15 deg C temp gradient between cell core and temp sensor, that is cuting it a bit fine.

I am finding at 20 deg C temp rise when discharging 12Ah from the battery at 30A continuous.
So on a normal summers day, the battery starts at 30 deg C, but by the end the battery is up to 50 deg C and the car starts giving me errors and stops behaving.

Back in 2006 I did a test on a 15 cell 40Ah LiFePO4 battery
I used a 48 lead acid charger (10A till 57.6v, 57.6v (3.85v ave) till 3A, 54.4v float (3.65v ave)
I used a BMS that would disconnect the charger if any one cell went above 4v and had shunts on each cell that came on at 3.65v
And I discharged at 55A (1.4C) with 3 - 4 second peaks at 180A
I remember the cells getting hot, but I didn't think to measure temperature at the time, as the thermal limit was 65 Deg C, and they weren't getting *that* hot.
I got 4-500 cycles until capacity fell to ~21Ah (reasonably even across all cells)

by comparison the battery I have been using in another vehicle is up to 8-900 cycles with capacity coming down from the original 60Ah down to 53Ah.
That battery is discharged at mostly 40 - 50A continuous, charged at 10A until 3.65v ave is reached, and then discharge ceases. peak discharge current is 200A for 3 - 4 seconds.

Although in this car I don't float charge, I do know that a continuous discharge rate above 1C is very bad for cycle life. As is high temperature.

Matt
2017 Renault zoe - 25'000km
2007 vectrix - 156'000km
1998 prius - needs Batt
1999 Prius - needs batt
2000 prius - has 200 x headway 38120 cells
antiscab
Senior Member
Posts: 2700
Joined: Mon, 26 Nov 2007, 05:39
Real Name: Matthew Lacey
Location: Perth, WA

Antiscab's NWH10 Prius plugin Lithium conversion

Post by antiscab »

For prototype 2 - the prius with the A123 cells, there has been some slight progress.

My contact at A123 has rounded up some of the end plates used in the official A123 battery packs, along with some stainless steel strapping.

I'm also using a reinforced connection method.
Matt
2017 Renault zoe - 25'000km
2007 vectrix - 156'000km
1998 prius - needs Batt
1999 Prius - needs batt
2000 prius - has 200 x headway 38120 cells
T1 Terry
Senior Member
Posts: 1155
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Antiscab's NWH10 Prius plugin Lithium conversion

Post by T1 Terry »

As far as stainless steel strapping https://www.blackwoods.com.au/part/0424 ... 5mm-x-305m no idea what the different coloured plastic carrier means, but I use the yellow one.

T1 Terry
Green but want to learn
T1 Terry
Senior Member
Posts: 1155
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Antiscab's NWH10 Prius plugin Lithium conversion

Post by T1 Terry »

I wonder how the system Tesla uses for controlling their cell temps would go with prismatic cells? Or maybe Peltier effect modules used to cool the air in the battery box using them mounted through the lid and fan assisted to circulate the air. Cold plates between the cells would probably work best but quite a challenge to make, a definite CNC machine job making the snail track in 2 opposing aluminium plates with an inlet at one side and outlet the other. Maybe the aluminium cased lipo cells would be easier after all, but the A123 pouch cells would transfer heat just as well I would think.

T1 Terry
Green but want to learn
antiscab
Senior Member
Posts: 2700
Joined: Mon, 26 Nov 2007, 05:39
Real Name: Matthew Lacey
Location: Perth, WA

Antiscab's NWH10 Prius plugin Lithium conversion

Post by antiscab »

T1 Terry wrote: I wonder how the system Tesla uses for controlling their cell temps would go with prismatic cells? Or maybe Peltier effect modules used to cool the air in the battery box using them mounted through the lid and fan assisted to circulate the air.


Submerging the cells in mineral oil would just keep the battery temp even (an issue I'm not actually having at the moment)

The issue is the cells are getting hot internally, because they're just too small.

If I'm willing to invest that much mass into thermal management, I may as well just install a bigger battery

an 80kg battery has a lot of thermal mass - raising it by 20 Deg C in the space of 30-40 mins means there is a *lot* of heat to get out.
Ironically, if I halve the discharge rate, the heat rise more than halves

I think the solution here isn't more cooling, rather don't make so much heat - at least that's what I'm hoping :D


Matt
2017 Renault zoe - 25'000km
2007 vectrix - 156'000km
1998 prius - needs Batt
1999 Prius - needs batt
2000 prius - has 200 x headway 38120 cells
antiscab
Senior Member
Posts: 2700
Joined: Mon, 26 Nov 2007, 05:39
Real Name: Matthew Lacey
Location: Perth, WA

Antiscab's NWH10 Prius plugin Lithium conversion

Post by antiscab »

just had my first fill up since getting the first prototype sort of working.

I drove 1104.5km, used 99.82 kwh and 40.92L (so 3.7L + 9.0kwh/100km)

driving petrol only I used to get 5.95L/100km (average over 3000km)

Once I put a stronger battery in, I should be able to shift more of the usage to electricity

goes to show just how efficient the new prius are (my dad's gets 3.9L/100km on petrol only)
Matt
2017 Renault zoe - 25'000km
2007 vectrix - 156'000km
1998 prius - needs Batt
1999 Prius - needs batt
2000 prius - has 200 x headway 38120 cells
gtyler54
Noobie
Posts: 26
Joined: Mon, 13 Aug 2012, 00:54
Real Name: George Tyler
Location: Hamilton NZ

Antiscab's NWH10 Prius plugin Lithium conversion

Post by gtyler54 »

so 9 kwh = 11L of petrol in your car, more or less
gtyler
antiscab
Senior Member
Posts: 2700
Joined: Mon, 26 Nov 2007, 05:39
Real Name: Matthew Lacey
Location: Perth, WA

Antiscab's NWH10 Prius plugin Lithium conversion

Post by antiscab »

gtyler54 wrote: so 9 kwh = 11L of petrol in your car, more or less


it would appear 9kwh displaces 2.25L of petrol used

that should be improved upon, once I get the car to stop discharging in time , otherwise it faults and actively charges the battery using the engine (which isn't particularly efficient)

the atkin cycle engine is supposed to be 38% efficient - looks like that number is bang on
Matt
2017 Renault zoe - 25'000km
2007 vectrix - 156'000km
1998 prius - needs Batt
1999 Prius - needs batt
2000 prius - has 200 x headway 38120 cells
LightningEV
Noobie
Posts: 18
Joined: Thu, 24 Sep 2009, 05:07
Real Name: Robert James
Location: Tamworth
Contact:

Antiscab's NWH10 Prius plugin Lithium conversion

Post by LightningEV »

was wondering if there was more coming to this wonderful little goldmine of information.
Progress was never made by thinking Inside the box.
antiscab
Senior Member
Posts: 2700
Joined: Mon, 26 Nov 2007, 05:39
Real Name: Matthew Lacey
Location: Perth, WA

Antiscab's NWH10 Prius plugin Lithium conversion

Post by antiscab »

I haven't had much time to do more with this car, other than drive it

I have started to only charge the LiFePO4 battery to only 3.40v cell average (340v across whole pack), so that when you first start driving the car doesn't go crazy because the battery is too full.
Matt
2017 Renault zoe - 25'000km
2007 vectrix - 156'000km
1998 prius - needs Batt
1999 Prius - needs batt
2000 prius - has 200 x headway 38120 cells
antiscab
Senior Member
Posts: 2700
Joined: Mon, 26 Nov 2007, 05:39
Real Name: Matthew Lacey
Location: Perth, WA

Antiscab's NWH10 Prius plugin Lithium conversion

Post by antiscab »

one thing I have noticed with the colder weather

the battery ECU lowers the max discharge rate limit when the battery gets below 20 deg C

I think one of the ways to get around the lower power of the GBS cells is to fake the battery temp being 10 or 15 deg C below the actual temperature

so many things to try, so little time
Matt
2017 Renault zoe - 25'000km
2007 vectrix - 156'000km
1998 prius - needs Batt
1999 Prius - needs batt
2000 prius - has 200 x headway 38120 cells
LightningEV
Noobie
Posts: 18
Joined: Thu, 24 Sep 2009, 05:07
Real Name: Robert James
Location: Tamworth
Contact:

Antiscab's NWH10 Prius plugin Lithium conversion

Post by LightningEV »

the battery ECU lowers the max discharge rate limit

Wow, that would really suck, because the default is like only 10kw.

I've been thinking there is an advantage to the 1st gen over the 2nd gen, the physical capability to run full ev power from batter pack. As it doesnt have a multi stage inverter, the only question is, is the current limited at the battery monitor or the inverter?.

I know 30kw isnt alot, but it would mean staying in ev mode more often and should mean a little extra power from the ice is available to the wheels.

What are your thoughts?
Progress was never made by thinking Inside the box.
antiscab
Senior Member
Posts: 2700
Joined: Mon, 26 Nov 2007, 05:39
Real Name: Matthew Lacey
Location: Perth, WA

Antiscab's NWH10 Prius plugin Lithium conversion

Post by antiscab »

LightningEV wrote: the battery ECU lowers the max discharge rate limit

Wow, that would really suck, because the default is like only 10kw.

I've been thinking there is an advantage to the 1st gen over the 2nd gen, the physical capability to run full ev power from batter pack. As it doesnt have a multi stage inverter, the only question is, is the current limited at the battery monitor or the inverter?.

I know 30kw isnt alot, but it would mean staying in ev mode more often and should mean a little extra power from the ice is available to the wheels.


The default is actually 21kw at battery temperatures between ~22 deg C and ~32 deg C (I really need to compile a look up table relating charge/discharge limits, SOC and temp)

The battery ECU only reports to the M-OBD bus (no can on a car this old) what the charge/discharge limits are. The inverter mostly sticks to this (the discharge limits get exceeded some times during engine starting)

Staying in EV mode more often, but I'm finding the engine needs to run periodically to keep it at temperature

The inverter ECU will decide whether to use battery, engine or a combination in the moment based upon engine temp, battery SOC and charge/discharge limits.

Again, I need to compile another table that shows the relationship

upping the battery max discharge only increases the power at the wheels at fairly high speed, but it does allow a greater instantaneous shift of load from engine to battery
Matt
2017 Renault zoe - 25'000km
2007 vectrix - 156'000km
1998 prius - needs Batt
1999 Prius - needs batt
2000 prius - has 200 x headway 38120 cells
LightningEV
Noobie
Posts: 18
Joined: Thu, 24 Sep 2009, 05:07
Real Name: Robert James
Location: Tamworth
Contact:

Antiscab's NWH10 Prius plugin Lithium conversion

Post by LightningEV »

21kw is the current limmit in the nhw20, the nhw11 was limited to 10kw before asking petrol engine to switch on at which point it would supply up to 18kw from the battery and 15kw from the generator.

If the NHW10 is different, as it may well be, then this would make things even more interesting.

The reason behind this suggestion is that if the SOC is high and the temp sensor is faked the engine will stay off within the power capabilities of the pack/inverter. thus enabling an EV mode for a car which never had an ev mode.
Progress was never made by thinking Inside the box.
antiscab
Senior Member
Posts: 2700
Joined: Mon, 26 Nov 2007, 05:39
Real Name: Matthew Lacey
Location: Perth, WA

Antiscab's NWH10 Prius plugin Lithium conversion

Post by antiscab »

LightningEV wrote: 21kw is the current limmit in the nhw20, the nhw11 was limited to 10kw before asking petrol engine to switch on at which point it would supply up to 18kw from the battery and 15kw from the generator.


This is where having the look up table comes into it's own

the nhw10 and nhw11 are the same in this regard, more than 10kw demand with engine temp below a certain streshold (but above the not run when stoppped threshold) and battery SOC below a certain threshold (65% actual IIRC) will result in the engine running.

However, with the engine at full temperature, battery SOC above 70% and throttle application rate slow enough, the car will use up to 16kw before engine start (or rather spark and ignition since at speed the engine has to turn anyway)

I regularly cruise at 40A continuous (12kw) with no engine running (when the GBS cells are warm enough not to sag too far)
Matt
2017 Renault zoe - 25'000km
2007 vectrix - 156'000km
1998 prius - needs Batt
1999 Prius - needs batt
2000 prius - has 200 x headway 38120 cells
LightningEV
Noobie
Posts: 18
Joined: Thu, 24 Sep 2009, 05:07
Real Name: Robert James
Location: Tamworth
Contact:

Antiscab's NWH10 Prius plugin Lithium conversion

Post by LightningEV »

12kw is enough to reach 100km/h in the right conditions, what is the top speed you have reached whilst being able to shutdown the petrol engine? I know it spins at 41mph with no fuel adding about 1.5kw of parasitic drag.

I have a project in the works for which I wish to use the Hybrid power plant in a lighter kit car, still trying to work out if the NHW11 will suffice or if I need to use the more complicated NHW20.
Progress was never made by thinking Inside the box.
LightningEV
Noobie
Posts: 18
Joined: Thu, 24 Sep 2009, 05:07
Real Name: Robert James
Location: Tamworth
Contact:

Antiscab's NWH10 Prius plugin Lithium conversion

Post by LightningEV »

Tested nhw11 0-50kmh 7sec with fuel pump off. Which equates to 18.5kw ev mode. Based on 1325kg mass of car and driver.

One should be able to use a voltage modifier circuit to limit throttle output to EV only values, thus giving the car EV mode! A simple WOT switch could be used to disengage the voltage modifier circuit and reingage hybrid mode, so that the EV mode works exactly the same as in a nhw20.
Progress was never made by thinking Inside the box.
antiscab
Senior Member
Posts: 2700
Joined: Mon, 26 Nov 2007, 05:39
Real Name: Matthew Lacey
Location: Perth, WA

Antiscab's NWH10 Prius plugin Lithium conversion

Post by antiscab »

LightningEV wrote:
One should be able to use a voltage modifier circuit to limit throttle output to EV only values, thus giving the car EV mode! A simple WOT switch could be used to disengage the voltage modifier circuit and reingage hybrid mode, so that the EV mode works exactly the same as in a nhw20.


If only it were that simple :(

The throttle position for EV only varies with (in this order of effect):
Reported Battery SOC
Reported charge/discharge limits
speed
Engine temperature
throttle position change rate
Battery temperature

rather a complicated thing to fake out - I do it manually at the moment by driving with TECU running the whole time


Also, update time
I have found that if I discharge the battery at 20A continuous, actual capacity has come back to ~9Ah

over the past 4417.9km I have used 163.43L petrol and 444.32kwh (3.7L + 10.05kwh/100km)

I still need to find a battery that's light, power dense with a long service life.

I *might* get this from rewired leaf modules - when I get time I will test this with my 40A cycle tester
Matt
2017 Renault zoe - 25'000km
2007 vectrix - 156'000km
1998 prius - needs Batt
1999 Prius - needs batt
2000 prius - has 200 x headway 38120 cells
queenslandchris
Noobie
Posts: 1
Joined: Mon, 01 May 2017, 02:11
Real Name: Chris Michael
Location: Hobart
Contact:

Antiscab's NWH10 Prius plugin Lithium conversion

Post by queenslandchris »

Hi, just wondering if there has been any further development on your project. I have a mint condition NHW 10 in Hobart that I was using as my daily runabout up to 6 months ago. The main original battery was very pour but I could still get 700kms on 40ltrs of petrol. At the moment its sitting in my backyard while I decide what to do with it. I was wondering if I could use the engine as a generator and convert the output to 240v ac? Any one done this. Or anyone looking for a NHW 10 in mint body etc
hunty
antiscab
Senior Member
Posts: 2700
Joined: Mon, 26 Nov 2007, 05:39
Real Name: Matthew Lacey
Location: Perth, WA

Re: Antiscab's NWH10 Prius plugin Lithium conversion

Post by antiscab »

I'm now having a look at using 200 x headway 10Ah cells, 100s2p for 20Ah at 320v
I think the headways won't heat up as much, and should deal with the peak loads better

Against my better judgement, I've been given another nhw10, though this one is registered, and an old customer has heaps of headway cells from an abandoned project.
Matt
2017 Renault zoe - 25'000km
2007 vectrix - 156'000km
1998 prius - needs Batt
1999 Prius - needs batt
2000 prius - has 200 x headway 38120 cells
antiscab
Senior Member
Posts: 2700
Joined: Mon, 26 Nov 2007, 05:39
Real Name: Matthew Lacey
Location: Perth, WA

Re: Antiscab's NWH10 Prius plugin Lithium conversion

Post by antiscab »

a very long overdue update on the second prototype.
I'm using 200 x headway 38120s cells for 320v 20Ah nominal

The cells are built into 5 x 40 cell modules (64v 20Ah each) with wires for 2 x zeva EVMS 12 cell monitors each, and a separate set of wires to go back to the cars original battery ECU
IMAG1465.jpg
IMAG1465.jpg (2.32 MiB) Viewed 931 times
after this photo was taken, I realised had forgotten to drill the ventilation holes in the bottom of the polycarb case, so I had to pull it all apart again:
IMAG1468.jpg
IMAG1468.jpg (555.57 KiB) Viewed 931 times
Drilling the 36 holes took a coupe of hours, including de-burring.
I used aluminium m6 nut serts to bolt the poly carb case to the bottom of the original battery supports *and* to the bottom of the overhead brace
The case is hard up in both vertical directions, so it's not really for vertical restraint, so much as horizontal restraint.

A test fit without the batteries:
IMAG1460.jpg
IMAG1460.jpg (816.71 KiB) Viewed 931 times
By using two current shunts, (a 75mv 200A and a 75mv 100A) only 1/3rd of the current should go through the original battery ecu current sensor, meaning the original battery guage should be rescaled to 19.5Ah

That moment when you realise you need access to a pair of terminals :(
IMAG1467.jpg
IMAG1467.jpg (911.9 KiB) Viewed 931 times
Since I put the battery modules together a year ago, I thought I should check the voltages before reassembly, lest I have to pull it all apart again
IMAG1469.jpg
IMAG1469.jpg (1.3 MiB) Viewed 931 times
Rats, one cell pair is 3.4v
I have a coupe of dead cells - glad I didn't put it all together

I'm charging the dead pair now, but I think I'll replacement them anyway. Better safe than sorry
Matt
2017 Renault zoe - 25'000km
2007 vectrix - 156'000km
1998 prius - needs Batt
1999 Prius - needs batt
2000 prius - has 200 x headway 38120 cells
T1 Terry
Senior Member
Posts: 1155
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Antiscab's NWH10 Prius plugin Lithium conversion

Post by T1 Terry »

How would the Railway loco super caps be for replacing the traction battery or adding as an aux battery that lives in the spare wheel well the way Jason set my 2006 Prius up?

T1 Terry
Green but want to learn
antiscab
Senior Member
Posts: 2700
Joined: Mon, 26 Nov 2007, 05:39
Real Name: Matthew Lacey
Location: Perth, WA

Re: Antiscab's NWH10 Prius plugin Lithium conversion

Post by antiscab »

Caps in parrallel would help mask a bad traction battery.

Would be interesting to try though
Matt
2017 Renault zoe - 25'000km
2007 vectrix - 156'000km
1998 prius - needs Batt
1999 Prius - needs batt
2000 prius - has 200 x headway 38120 cells
T1 Terry
Senior Member
Posts: 1155
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Antiscab's NWH10 Prius plugin Lithium conversion

Post by T1 Terry »

Another angle, the caps to replace the traction battery and an aux battery built using the smaller capacity LFP or LTO cells that trickle charge from the cap pack at the safe charging rate for the LFP/LTO cells so they retain their cycle life. This combination would greatly enhance the electric assist or all electric range and greatly extend the power pack life. Weight and cost would be an issue though ...... still thinking about the possibility of using the Prius electrical part and either replacing the ICE with another electric motor or a much smaller ICE engine that could be spun up easily to allow the all electric drive to be more energy efficient and the hybrid drive to be more fuel efficient.

Would an increase in the traction pack/aux pack voltage improve the performance of the NWH10 set up?

T1 Terry
Green but want to learn
antiscab
Senior Member
Posts: 2700
Joined: Mon, 26 Nov 2007, 05:39
Real Name: Matthew Lacey
Location: Perth, WA

Re: Antiscab's NWH10 Prius plugin Lithium conversion

Post by antiscab »

optimal ICE sizing in a hybrid application is a bit of a balancing act. peak efficiency (which is what matters in a hybrid) improves as engine size goes up. energy lost to maintain temperature also goes up too.
That's one of the reasons the engine size in the prius has been getting bigger as time goes on

I would suggest that if keeping an ICE, keep the one that's already there.

from memory 380vdc is where regen starts getting curtailed or cut back completely. (I probably have it noted further back in this thread, it may actually be when any cell group goes above 19 or 20v)
performance likely wouldn't improve with a higher pack voltage, so much as one with lower internal resistance.
The battery discharge is power limited to 19kw (which the battery ecu can revise down), while in theory mg2 can take 30kw and mg1 can add another 10kw.

I'll be interested to see whether faking the battery ecu current sensor means a higher discharge limit.
I suspect when the battery ecu tells the inverter 19kw is the battery discharge limit, the inverter limits it using it's own current measurement.

The electric only running above 60kmh needs the engine to turn over due to mg1 reaching it's rpm limit

using super caps to buffer regen is an interesting thought.
using those 48v 83F cap banks, you would need at least 8 in series, though more likely for voltage balancing reasons it would be 10 (with a 83 - 100F range of actual capacitance)
so that gives 8.3F
the original 240 x nimh 6.5Ah cells nominally are similar to 100x LiFePO4 = 324v. voltage goes up to a max of 380vdc.
That gives a voltage window of 56v. with 8.3F that gives 464 A seconds
The internal resistance under regen is non-linear, but if we assume the caps take half the current till voltage limit, it gives 15.5 seconds where regen is at the 60A limit (which the prius only adheres to on average, it overshoots all the time).

adding the $3000 30kg of super caps drops the max regen to 15A per 320v 10Ah string
adding $2600 33kg of another 320v 10Ah string instead would drop the max regen to 20A per 320v 10Ah string, but also increase the capacity
Although going with a bigger battery, it would be cheaper to start with 40Ah cells, which I already know can take 60A of regen.

I like the idea of using super caps in place of the original traction battery - service life should be longer than the nimh, though not sure how the car would react to a battery with nearly no capacity. I suspect it would treat it as a bad battery.
Matt
2017 Renault zoe - 25'000km
2007 vectrix - 156'000km
1998 prius - needs Batt
1999 Prius - needs batt
2000 prius - has 200 x headway 38120 cells
Post Reply