I've been reading all the debate on transmissions with some interest, but let me throw another fly in the puch bowl...if you do need the variable ratios (which is another debate I will not venture into), why not do it like the prius does?
You would need a primary drive motor, an output shaft, and a smaller "ratio motor" interconnected through a differential. You could even use two identical motors, though you may not need to. The electronics could decide how to vary the ratios, and what motor needs to spin which direction, at what speed, etc., and if one or both needs to regenerate. >From a viewpoint of drive system design this is harder, and prevents using an entirely off the shelf unit. Although possibly Victor's drives could be configured with an additional "control box" to run in this setup...Victor? The advantage would be that the software would continually be determining what the "best" ratio would be and the driver wouldn't have to worry about it. You could even select a "best efficiency" mode or a "best performance" mode from a dash mounted selector switch. But from a viewpoint of assembly labor hours is where it really shines. These make the difference in cost to a production run of anything. You would have fewer parts, no clutches, (potentially) very easy attachment of the motors, and the ability to get "just the right" ratios at the right times. Just connect the wiring harnesses and forget it. I don't know if the weight would be a prohibitive factor, though. Anyway, just a thought. -----Original Message----- From: Peter VanDerWal [mailto:[EMAIL PROTECTED]] Sent: Sunday, September 08, 2002 10:15 PM To: [EMAIL PROTECTED] Subject: Re: [EVDL]Re: Volume build proposed for high performing EVs <Sigh> I'm comparing a fixed single speed transmission to a multispeed syncromesh, perhaps I haven't been exp-ressing it well but that's what I'm talking about. The synchromesh (as we agree) has all fwd gears meshed all the time. Assuming a 5-speed transmission that means 4 gears are idling on bearings (losses not present in a single speed) as well as 4 sets of gears "meshing" (more sliding surfaces and more friction, small but present) that don't exist in the single speed. Plus all of those extra gears spinning in the oil, etc. Actual measurements indicate 10% losses in a standard tranmission. Even using a two stage reduction you would only have 4 or 5 % in the single stage (unless you are willing to put up with the noise from spur gears, in which case 2-3%). That's a 5% advantage, and includes bearing and gear losses. Differentials have no looses unless you are cornering. Granted hypoid gears (usually found in front engine / rear wheel diffs) have looses but our single speed gear box has no need for hypoid gears. Even if it was only a 2% advantage, that's the same as the difference you estimate for different speeds on the motor so what is the point of the extra weight and complexity? "I" never said that AC was a cure all, just that they can do quite well with a single ratio. The EV1 has a single speed (spur cut IIRC) gear box and does 0-30 in 1.6 seconds and 0-60 in 7.4 seconds. Do you consider this too slow? >Virtually *all* syncromesh and non syncromesh gearboxes have all the >forward gears in mesh *all* the time. Reverse alone is usually on a >slider. Half the gears are fixed to shafts, the other half are on >bearings (in fixed and free pairs). Syncromesh or not. SAAB, VW, Webster >(non-syncro), Borg-Warner T-10, they all do it. > >Secondly any gearbox with a reduction of more than 5:1 will likely have >more than one stage. Partly to keep gear size under control. A 10:1 >reduction with a 2.25" pinion then has to have a 22.5" mating gear. >which means a gear case more than 24" in diameter once you include >bosses. That's 3X the diameter of a typical FWD car clutch and probably >twice the diameter of the bellhousing. A two stage reduction can get >you about as far as 25:1 before it becomes less practical. I have seen >crownwheel and pinion setups that go to 7.17:1, but they are not common. >The crownwheel is in excess of 15 inches in diameter. It goes in the >rear axle of a sixteen ton bus. Further reduction at the axle is done >with an epicyclic ahead of the pinion to make it a two speed rear axle. >Again, the exception, not the norm. > >Look at the Brusa AT 1200 for example. (www.brusa.li) from the >measurements you can see that it has to be at least a two stage >reduction to get the ratios they claim (7:1 to 14:1) and fit in the >shape with the center to center distance. That is an EV AC drive fixed >ratio gearbox. > >As for inefficiencies that are the same regardless of single or >multispeed, there would be bearing losses in input, intermediate and >output shafts, shaft oil seals, losses in the differential. Windage >losses from fewer rotating parts would be less, but many sources of loss >remain. So taking out a few gears would gain a bit, but there are may >losses that are not related to whether it is multispeed or not. > >I was using the VW golf as an example of a FWD car, they sold about 20 >million of them in gas and diesel models and the tire size and engine >speed is pretty common for that sort of car. I also happened to have the >factory workshop manual handy. Final drives vary from 3.67 to 4.25 and >first gear is about 3.45, second is between 1.8 and 2.2. Depending on >the engine and model. > >*My point* was that AC wasn't a cure-all and that although very >flexible, they still leave something to be desired as far as the >torque-speed envelope goes. Both mechainically and electrically. > >Seth > > > >Peter VanDerWal wrote: > >>> >>>First, what rubbing surfaces in a syncromesh? Syncromesh is engaged or >>>not, with a transition of sliding lasting perhaps a few hundred >>>milliseconds. >>> >><Sigh> I didn't say the sliding was in the syncro's . Metal sliding on >>metal is an inherent property of syncromesh transmissions. >>You said it yourself;" all forward gears in mesh all the time"....think >>about it. >> >>>Second, in a traction AC drive you are limited in how low you can gear. >>>At 13:1, a likely first gear (VW golfs came with 12.7 to 14.5 for first >>>gear depending on engine), the motor has to hit about 13,000 rpm for >>>highyway speeds, assuming a 22.8" wheel, >>> >>Ok so it's not exactly the same as first gear on your vehicle, call it >>first and half (more than first, less then second) >> >>-snip- >> >>>All my texts say 98% efficiency for helically cut gears, 99% for spur >>>cut. Which isn't that much for loss. If you picked up 2% by shuttling >>>around on the efficiency map for the AC motor (which is very doable) >>>then you could at least break even. And as a bonus, multispeed manual >>>gearboxes are quite common, unlike single speed. >>> >>Well thanks for proving my point. According to you a single reduction >>transmission would be about 98% efficient, whereas MEASURED efficiency >>on multispeed manual transmissions is around 90% give or take a percent >>or two. >> >>Actually 98% for helically cut gears assumes a ratio of 3:1 or less. >> Higher ratios tend to have slightly more loss. >> > IMPORTANT - THIS MESSAGE (INCLUDING ANY ATTACHMENTS) IS INTENDED ONLY FOR THE USE OF THE INDIVIDUAL OR ENTITY TO WHICH IT IS ADDRESSED, AND MAY CONTAIN INFORMATION THAT IS PRIVILEGED, CONFIDENTIAL AND EXEMPT FROM DISCLOSURE UNDER APPLICABLE LAW. IF YOU ARE NOT THE INTENDED RECIPIENT, YOU SHOULD DELETE THIS MESSAGE IMMEDIATELY AND YOU ARE HEREBY NOTIFIED THAT ANY READING, DISSEMINATION, DISTRIBUTION OR COPYING OF THIS MESSAGE, OR THE TAKING OF ANY ACTION BASED ON IT IS STRICTLY PROHIBITED. THANK YOU.
