Transmission losses

General BVS related area
User avatar
nino500
Posts: 63
Joined: Tue Jul 08, 2008 11:24 pm
Location: Isleworth, Middlesex

Transmission losses

Postby nino500 » Wed Aug 20, 2008 9:08 pm

Here's a question for the techies..........

Just how much power do we lose through various types of transmission in a coverted EV?

I've always been a biker and, over the years I've been led to believe (by bike mags, hear say and experience) that shaft drive bikes lose a lot of power in the bevel drive gears.

Experience has shown me that a 50bhp shaft drive bike is considerably slower than a chain drive bike with a similar engine. (Try driving a Honda NTV650 shaftie and then the Bros/Hawk 650 with virtually the same engine but chain drive)

BUT........How does this affect a 600kg car with a 10kw electric motor?

Will removing the trans and driving the halfshafts directly by belt be worthwhile?

Does a traditional rear axle with bevel driven diff cost power even if the motor drives direct to the shaft?

Just a few thoughts to get the grey matter working..........

Cheers guys,
Nino.

User avatar
Jeremy
Posts: 472
Joined: Thu Jul 31, 2008 6:35 pm
Location: Salisbury

Postby Jeremy » Thu Aug 21, 2008 6:13 am

Chain drive is surprisingly efficient, losses are only around 2% to 3% for a well-maintained chain. Gears can be efficient, but bevel gears in a bike drive shaft arrangement, particularly if cut so that they aren't excessively noisy, will probably lose around 4% to 5% or so.

The differences aren't that great, particularly as the losses from a poorly maintained chain, with worn sprockets, can rise quite dramatically.

Straight cut gears are more efficient, but are usually horribly noisy.

Gates claim that their GT2 toothed belts can be almost as efficient as chain drive, but with a longer life, no lubrication and less maintenance.

Although using the existing transmission is attractive when converting a light car, I've often wondered whether or not chain, or belt, driving the inboard ends of the drive shafts from a pair of motors might not be slightly more efficient, and perhaps lighter, way to go. Differential effect would come automatically as each motor experienced different torque demands.

Jeremy

User avatar
qdos
Posts: 2089
Joined: Fri Apr 27, 2007 10:26 pm
Location: Dorset
Contact:

Postby qdos » Thu Aug 21, 2008 9:04 am

Jeremy wrote:
Although using the existing transmission is attractive when converting a light car, I've often wondered whether or not chain, or belt, driving the inboard ends of the drive shafts from a pair of motors might not be slightly more efficient, and perhaps lighter, way to go. Differential effect would come automatically as each motor experienced different torque demands.

Jeremy


I have to say I tend to think the same Jeremy

User avatar
nino500
Posts: 63
Joined: Tue Jul 08, 2008 11:24 pm
Location: Isleworth, Middlesex

Postby nino500 » Thu Aug 21, 2008 9:24 am

Thanks Jeremy.

I had the feeling (and it is only by the scientific measuring system known as 'feel') that, in the bike case, the losses were greater but it occurs to me now that most shaft drive bikes have two sets of helical cut bevel gears as their crankshafts are accross the frame and the drive is turned through 90 degrees twice. By your figures this would double the loss to around 10% which coincides with my 'feeling'.

I am now thinking of the two motor and belt drive system on my Fiat 126 (using two eteks). Another concern is, if I use two controllers to provide enough peak current to both motors, is it likely that manufacturing differences will lead to one motor giving more torque than the other, casing a sideways 'pull' under power.

I guess only time will tell on that one and I'm planning to use the car as a sort of test bed for several different systems (a 'Travelling Science Project', but deliberately so).

MalcolmB
Posts: 423
Joined: Sat Jun 23, 2007 8:07 pm

Postby MalcolmB » Thu Aug 21, 2008 3:39 pm

Jeremy wrote: I've often wondered whether or not chain, or belt, driving the inboard ends of the drive shafts from a pair of motors might not be slightly more efficient, and perhaps lighter, way to go.

I really like the simplicity of this approach. I've been looking here http://www.fptgroup.com/products/pulleys.asp at belt drives, but I can't see any easy way to do it without a two-stage reduction. To get decent low-speed torque you're likely to need an overall reduction of around five to one. To get this with a single reduction using 8 mm pitch pulleys it would mean a final pulley diameter of around 350 mm, which is unlikely to give sufficient clearance under the vehicle. With a two-stage reduction you're talking four belts and eight pulleys which naturally adds to the inefficiency and cost.

P.S. Just checked my car. The largest pulley I could safely fit without raising the inboard ends of the drive shafts is about 200 mm.

User avatar
Jeremy
Posts: 472
Joined: Thu Jul 31, 2008 6:35 pm
Location: Salisbury

Postby Jeremy » Thu Aug 21, 2008 6:09 pm

I know it might sound a bit odd, but have you thought of using chain drive instead of belts?

A chain drive running in a sealed oil bath will be extremely long lived and reliable, it's only really the dirt, corrosion and lack of lubrication that makes them wear so quickly on bikes.

You may find that you can get sprockets that will allow the sort of reduction you need and that will fit in the space you have. To get the small diameter you may need to go to a smaller size chain, in which case you will probably need to go to a duplex or triplex chain to get the torque capacity.

HiVo inverted gear chains might be another possibility, but I've no direct experience of using them.

The Prius uses a chain final drive, BTW, as do some 4x4 transfer boxes, I believe.

Jeremy

User avatar
nino500
Posts: 63
Joined: Tue Jul 08, 2008 11:24 pm
Location: Isleworth, Middlesex

Postby nino500 » Thu Aug 21, 2008 8:56 pm

Yes, I have no problem with chains. They wouldn't even be too noisy if enclosed.

Interestingly though, the small wheels on the 126 mean the shafts would be spinning at 1100 rpm at 60mph and, as the eteks are at about 3500 at 72 volts (from memory) I would only need about 3.5:1 so the sprockets wouldn't need to be enormous.

User avatar
EVguru
Posts: 305
Joined: Fri May 25, 2007 5:17 pm
Location: Luton
Contact:

Postby EVguru » Thu Aug 21, 2008 10:19 pm

I've seen a Ford Fiesta with two Lynch/Lemco (not certain which version) motors in that kind of setup not be able to drive up onto a tilt bed trailer. It just didn't have enough torque.

The efficiency of the Etek seems to fall off rather badly over 350 amp and there is another consideration that seems to get overlooked a lot of the time. The continuous current rating of a Lynch/Lemco/Agni/Etek is rpm dependent (unless force cooled). The armature acts as it's own fan.

Eteks have a mixed reputation amoungst the motorcycle conversion crowd. Many many motors have been damaged/destroyed. When I took my Cagiva motorcycle to the first Sparkford Electrathon event, I'd rushed to get it ready and there was no additional motor cooling. I'd geared it for 60mph on 48volt, which should have been just about OK on the road with forced cooling. On the tight little Sparkford track where the highest speed achievable was about 40 mph, the motor overheated after about 8 laps (approx 8km) to the point that it spat all the solder out of the capping strips. Fortunately I had the equipment to sucessfully resolder all the joints and the motor worked well again.

The motor may be rated for 15hp peak, but that's from cold and it's at rated rpm. If it's already warmed up from running along the road, then the peak overload capability is reduced.
Paul

http://www.compton.vispa.com/scirocco/
http://www.morini-mania.co.uk
http://www.compton.vispa.com/the_named

User avatar
nino500
Posts: 63
Joined: Tue Jul 08, 2008 11:24 pm
Location: Isleworth, Middlesex

Postby nino500 » Thu Aug 21, 2008 11:25 pm

Interesting point about the fiesta. I wonder if that was using two controllers or one with enough power to provide maximum amps to both motors. Mind you, a tilt bed can be pretty steep.

The cooling will have to be investigated too. With full current available to both motors a few hill starts could lead to a similar scenario. Do you know of a practical motor temperature sensor system that doesn't get too complex?

User avatar
nino500
Posts: 63
Joined: Tue Jul 08, 2008 11:24 pm
Location: Isleworth, Middlesex

Postby nino500 » Fri Aug 22, 2008 12:07 am

I must apologise for starting this thread in the wrong area.

I guess it should really have gone in "conversions" or "cars, vans and other road vehicles".

I'd suggest a "general technical issues" area but I'd have to be in the "suggestions" area to do that.

The responses have been very helpful, all the same.

Nino.


Return to “General”

Who is online

Users browsing this forum: No registered users and 26 guests