Fantastic achievement. I wouldn't have thought it would take off that quickly with a Leaf motor? Are you giving the motor more power than the Leaf would have done? Or is it the case that the Leaf's traction control system would have just intervened to back the power off when wheelspin was detected? (I'm guessing the traction control on the Freelander (if it had one?) can't talk to the Leaf "throttle"?
Nodge68
Well-Known Member
A Nissan Leaf has an 80kW motor, so is quite capable of smoking the front tyres if too much throttle is used to pull away. An electric motor has maximum torque from the instant it starts turning, so breaking traction is pretty easy on a FWD car. What's impressive is the Leaf motor will spin a front and rear wheel on the Freelander when pulling away quickly.
The Freelander TC doesn't communicate with the ECM, it just applies the brakes to limit wheel spin. However when used in this current EV form, the ABS module doesn't know the throttle position, so doesn't know there's any wheel spin going on, so TC doesn't function.
That's what I meant. Had it only been driving the front wheels, I wouldn't have given it much thought, but driving all of them and STILL breaking traction was indeed the bit that impressed me. A Leaf isn't THAT powerful. I don't know which one forms the basis of this conversion, but the early ones were only 80 kW and newer ones were 110. I've been driving an EV with twice that, this year, and really didn't think the Freelander would be anything like that quick off the line with the Leaf motor.
Not sure I understand that? If a wheel is spinning, the ABS sensors will detect it's rotating significantly faster than the other wheels, so I don't see where the throttle position comes into it? It should be able to know that there's wheelspin going on, without a throttle input?
Alibro
Well-Known Member
Its all about gearing. I have the Leaf motor bolted to the Freelander gearbox so when I use 1st it is probably a lot lower geared than the Leaf would be. Normally I start off in 2nd which i think is closer to the original Leaf gearing but I then go up the gears as normal,
Last edited:
Nodge68
Well-Known Member
The Leaf motor at 80 kW has the same torque figure at zero RPM as the TD4 engine at 1800 RPM. So if you were to rev the TD4 to 1800 RPM and dump the clutch in first gear, it would still spin the wheels.
The ABS module can only apply TC when it can compare all data streams to a datum value. Yes the module applies brakes to a wheel or pair of wheels that it sees are spinning faster than the rest, but it can only do that if it knows that throttle is being applied requires the brakes to keep under control. On the standard ICE FL1, the TC doesn't function at idle speed, as the ABS module doesn't see excess throttle that needs to be kept in check.
I've not asked Ali, but suspect that the TC isn't functioning, although ABS is still working just fine.
Gosh! So it still drives through the original clutch? Years ago, (about 10!) we converted a number of diesel vans to electric. (Definitely not our finest hour, it has to be said)! We bolted the motor straight to the gearbox input shaft, so you couldn't change gear unless stationary and with no load on the drivetrain. Mostly they were in 2nd, some just about managed to start off in 3rd. They were front wheel dive, but any kind of wheelspin was a distant dream. I take it that video was starting off in 1st?
Understood, but I don't think I EVER got my TD4 to spin its wheels on damp tarmac, no matter where I dumped the clutch! I dd, once, get a wee chirp out of one (maybe both) front tyres when doing a hill start with a big trailer full of wood on the back, in the wet, but only for a fraction of a second. Traction on mine, was incredible. The X-Trail that it replaced, would give a brief bit of wheelspin from the front wheels under the same situation, before its (electronic) 4WD system kicked in (unless it was set to "lock", which keeps the clutch on the nose of the rear diff energised all the time).
OK, that makes sense, ta. Absolutely brilliant project. Hugely impressed! I'd certainly like to hear more as it progresses - particularly range and top speed?
Nodge68
Well-Known Member
That's already been covered in the EV thread.
It does about 100 miles per charge, working out at about 3 miles per kWhr of energy. Top speed would be similar to the TD4, however acceleration is much better, better than the V6 petrol version.
I'll have to have a look. I don't come on here so much since I got rid of the Freelander. Surprised it's as good as 3 miles to the kWh. What did the Leaf do?
Nodge68
Well-Known Member
Here's Ali's EV build thread. https://www.landyzone.co.uk/land-rover/freelander-ev.360880/
I thought I'd not seen you about for a bit.
The Leaf does about 3.5 miles per kWhr.
As Ali still has a manual box, he can select the best ratio for lowest current draw, helping keep the kWhr per mile high. It's not bad considering it's up at about 1900kg now.
Ali has done a lot of YouTube videos on the build too.
Alibro
Well-Known Member
No, I did what you said and with help from Nodge built a coupler to link the gearbox directly to the motor.
Changing gear is no problem though, maybe cause I don't have regen.
Everything Nodge said is correct although I won't be pushing the range again. I managed 110 miles then the battery voltage suddenly fell off a cliff and I had to limp it home, stopping at a friends house to recharge for a couple of hours. The voltage didn't fall below 3V per cell though and settled a little above that so I shouldn't have done any harm.
Last edited:
thefrog
Active Member
I did the test today because the car was resisting when reversing on full steering lock.
It looks good, I have 35 Seconds for 45° and 5 Kilos weight with a 1.20 M bar, just like the youtube video
I noticed that the rear tyres where at 2.1 Bars and the front at 2.4 Bars
So I pumped up the rear and it seems alot better
It looks good, I have 35 Seconds for 45° and 5 Kilos weight with a 1.20 M bar, just like the youtube video
I noticed that the rear tyres where at 2.1 Bars and the front at 2.4 Bars
So I pumped up the rear and it seems alot better
No, I've no real reason to pop in here much these days. The Freelander still lives (my dad has it), but his driving days are well and truly numbered now. I did a service on it for him just after Christmas. It still drives quite nicely, really. Not that it does many miles these days. He managed to stove-in both sills at the front, trying to jack it up. (Don't ask me how!) and there's a bit of rust on the bottom of the driver's door, but other than that, it's holding together quite well.
My congratulations. It's a hell of a project! I'm driving an EV as a company car at the moment, and really quite enjoying it, to be honest. Presumably, you have to feather the throttle to get pretty much no torque passing through the drivetrain before you can change gear? It'll be interesting to see how long the bearing inside the big gear in the IRD lives, I guess?
kernowsvenski
Well-Known Member
Retest done today. Its a bit warmer than the last test, and unfortunately I couldn't do it after the car had stood overnight. Had to do school run first thing. The car had stood for two hours.
1. 46s
2. 33s
3. 32s
4. 28s
5. 28s
6. 25s
7. 27s
8. 28s
9. 28s
10. 27s
11. 25s
12. 27s
13. 25s
14. 25s
15. 27s
This time the test was done forwards. 8kg @ 1.2m
Much more consistent this time.
1. 46s
2. 33s
3. 32s
4. 28s
5. 28s
6. 25s
7. 27s
8. 28s
9. 28s
10. 27s
11. 25s
12. 27s
13. 25s
14. 25s
15. 27s
This time the test was done forwards. 8kg @ 1.2m
Much more consistent this time.
Alibro
Well-Known Member
I just sorted out the cabin heater and tidied up the Low Voltage junction box so it was off the road for a couple of weeks.
I really missed it as I get a real kick out of driving it.
To change gear I just lift off the throttle completely and it changes gear no problem. I do try to be at an appropriate speed for the rpm when changing down gears but changing up is easy.
Pop over to my thread if you want to discuss it as this thread is for VCU testing.
Freelander 1 - Freelander EV
They are built by the same crowd that built the Jag I-Pace and the likes of these... They are indeed, along with many other cars, and the IRD on the FL1 too. ;)
www.landyzone.co.uk
Swl26
Member
Did the test today to check VCU
Supposedly a Bell rebuilt unit bought from a breakers 4 years ago.
1.2ish m with 5kg milkshake syrup jug
1:34 going forwards on car that had sat overnight.
Little long but good enough and hopefully confirms poorly diff mounts are due to Britpart cheapo's.
Supposedly a Bell rebuilt unit bought from a breakers 4 years ago.
1.2ish m with 5kg milkshake syrup jug
1:34 going forwards on car that had sat overnight.
Little long but good enough and hopefully confirms poorly diff mounts are due to Britpart cheapo's.
Last edited:
Bife
Active Member
Hi!
Has anyone tried tests to see how the time varies with weight and lever arm length?
From this it seems that Revs/min is simply proportional to torque (weight x lever arm)
I know its not the same geometry, but its an 'engineering' equivalent system and everyone's VCU is the same dimensions internally (I pressume), so in the equation
our equivalents of L, D & h are constant so we end up with:
T = μ x N x [Some constant] = Weight x Lever Arm
So, if someone did the test with the same VCU then since the viscosity is always the same (assuming the rate effects on viscosity are negligible given the relatively small differences in speed due to different lever arms and weights we would be talking about) then if you plotted time verses lever arm or weight you would get straight lines
Even if the plots of diff weights & levers wasn't exactly linear you could use the plots as 'calibration plots'
Given the variability in the times I have seen between various repetition of the test that you get anyway, any errors due to the assumptions in my approach wouldn't be significant anyway
If that was the case you could convert from times with different lever arms &/or weights
I think
Of course, the best way is if everyone just uses the 'ASTM standard FL1' of 5kg and 1.5 m ' or whatever, but I am just curious
PS I cant do the test myself as my VCU takes forever as it was fried in the PT sun by mistake - DOH! - and I am still waiting for a local guy to get mine refurbished
Has anyone tried tests to see how the time varies with weight and lever arm length?
From this it seems that Revs/min is simply proportional to torque (weight x lever arm)
I know its not the same geometry, but its an 'engineering' equivalent system and everyone's VCU is the same dimensions internally (I pressume), so in the equation
our equivalents of L, D & h are constant so we end up with:
T = μ x N x [Some constant] = Weight x Lever Arm
So, if someone did the test with the same VCU then since the viscosity is always the same (assuming the rate effects on viscosity are negligible given the relatively small differences in speed due to different lever arms and weights we would be talking about) then if you plotted time verses lever arm or weight you would get straight lines
Even if the plots of diff weights & levers wasn't exactly linear you could use the plots as 'calibration plots'
Given the variability in the times I have seen between various repetition of the test that you get anyway, any errors due to the assumptions in my approach wouldn't be significant anyway
If that was the case you could convert from times with different lever arms &/or weights
I think
Of course, the best way is if everyone just uses the 'ASTM standard FL1' of 5kg and 1.5 m ' or whatever, but I am just curious
PS I cant do the test myself as my VCU takes forever as it was fried in the PT sun by mistake - DOH! - and I am still waiting for a local guy to get mine refurbished
Hi!
Has anyone tried tests to see how the time varies with weight and lever arm length?
From this it seems that Revs/min is simply proportional to torque (weight x lever arm)
View attachment 320037
I know its not the same geometry, but its an 'engineering' equivalent system and everyone's VCU is the same dimensions internally (I pressume), so in the equation
View attachment 320036
our equivalents of L, D & h are constant so we end up with:
T = μ x N x [Some constant] = Weight x Lever Arm
So, if someone did the test with the same VCU then since the viscosity is always the same (assuming the rate effects on viscosity are negligible given the relatively small differences in speed due to different lever arms and weights we would be talking about) then if you plotted time verses lever arm or weight you would get straight lines
Even if the plots of diff weights & levers wasn't exactly linear you could use the plots as 'calibration plots'
Given the variability in the times I have seen between various repetition of the test that you get anyway, any errors due to the assumptions in my approach wouldn't be significant anyway
If that was the case you could convert from times with different lever arms &/or weights
I think
Of course, the best way is if everyone just uses the 'ASTM standard FL1' of 5kg and 1.5 m ' or whatever, but I am just curious
PS I cant do the test myself as my VCU takes forever as it was fried in the PT sun by mistake - DOH! - and I am still waiting for a local guy to get mine refurbished
Yes.
I’ve absolutely no idea what any of that means, but ‘yes’.
GrumpyGel
Well-Known Member
Hippo has charted 1WUT times using different weights on the same VCU.Hi!
Has anyone tried tests to see how the time varies with weight and lever arm length?
From this it seems that Revs/min is simply proportional to torque (weight x lever arm)
View attachment 320037
I know its not the same geometry, but its an 'engineering' equivalent system and everyone's VCU is the same dimensions internally (I pressume), so in the equation
View attachment 320036
our equivalents of L, D & h are constant so we end up with:
T = μ x N x [Some constant] = Weight x Lever Arm
So, if someone did the test with the same VCU then since the viscosity is always the same (assuming the rate effects on viscosity are negligible given the relatively small differences in speed due to different lever arms and weights we would be talking about) then if you plotted time verses lever arm or weight you would get straight lines
Even if the plots of diff weights & levers wasn't exactly linear you could use the plots as 'calibration plots'
Given the variability in the times I have seen between various repetition of the test that you get anyway, any errors due to the assumptions in my approach wouldn't be significant anyway
If that was the case you could convert from times with different lever arms &/or weights
I think
Of course, the best way is if everyone just uses the 'ASTM standard FL1' of 5kg and 1.5 m ' or whatever, but I am just curious
PS I cant do the test myself as my VCU takes forever as it was fried in the PT sun by mistake - DOH! - and I am still waiting for a local guy to get mine refurbished
The results are not linear or else you'd end up with negative times for heavier weights. Only the VCUs in DeLoreans can do that.
