PDA

View Full Version : AC 15 vs AC 20 on Curtis 7601



x88x
19 February 2012, 0420
Hi all, I'm in the planning phase for an upcoming motorcycle conversion, and I've been spending a lot of time researching motors lately. I've come to the conclusion that in order to get the performance I want, on the size and weight platform that I am expecting, I think that either the AC-15 or AC-20 will likely be my best bet.

For context, the donor I am looking at (hopefully picking it up this afternoon) is a Suzuki GS1100GL, so heavier than some but more space as well. This extra space comes in to the range I'll need; I'm currently estimating a 10-12kWh pack to get me the 75 mile safe (80% DOD) range I need. That'll be ~55mi at ~60mph, ~12mi at ~50mph or less, and a bit of buffer. This is for my daily commute, expecting to not be able to charge at work. I'm going to see if the building management company might be willing to work something out before I actually buy the batteries, but this is what I am going to base my calculations on for now. That size pack, if I can get my hands on Sinopoly's new 60Ah cells, should come to ~250lb, plus motor, controller, mounting hardware, and so on. My current rough estimate is that the electric components will likely come to ~400lb. The original bike is either ~550lb or ~630lb (the internet's a bit fuzzy on what an '82 GS1100GL weighs), so my goal is to not go more than 50lb over the original weight (GVWR's over 1,000lb, so I'm not worried about that ;) ). So I'll be looking at hopefully about a 600lb bike, but I want at least an 80mph top speed, and strong performance both at low speed and at mid-range speed.

...anyways, sorry if that got a bit too 'build planning' for the 'Motors' board, I just wanted to make sure I set as much of the context as possible.

Point of all that setup is, I haven't been able to find a comparison of the AC-15 and the AC-20. HPEVS has performance data for both, but the 15 is (presumably) on the 7501 (550A) at 72V, and the graph for the 20 shows current going over 700A, so I am guessing it's on the 7601 (650A), but only at 48V. The AC-20 torque drops off a lot faster, but I'm guessing that's largely because of the lower voltage?

So really, my question is, has anyone seen data on the two motors on the same voltage/current? Specifically on the 7601 at 90V+, but any controller, voltage, etc, would be fine; I'm trying to get a feel for the difference between the two.

Also, if anyone knows or can point me to an explanation of how they are actually different, that would be awesome.

podolefsky
19 February 2012, 0532
According to HPEVs, both the 15 and 20 do about 63 ft-lb at 550A and about 75 ft-lb at 650A. Normally you'd expect a longer motor to have more torque, but the 15 and 20 are wound differently so the 20 has the same torque, but to a higher RPM.

I haven't been able to find plots anywhere other than what's on HPEVs (doesn't mean they don't exist...but I've looked and looked and couldn't find them).

Yeah, lower voltage means torque drops at lower RPM.

You can get an estimate of what the curves will look like at different voltages by moving the "knee" to the right. That's the spot where the torque curve starts to drop. Move it proportionally to the voltage difference. So if you have a 72V curve and want to know what the 96V curve would look like, move the knee over by 96/72 = 133%. So on the AC-15 550A plot, the knee would be at about 3300 RPM instead of 2500.

Take a look at the difference between then 72 and 96V curves for the AC-31 to see what I mean.

x88x
19 February 2012, 0636
Hmm, ok. I am always hesitant applying blanket adjustments like that, but the AC-31 at least does fit that adjustment. So with that adjustment, it would appear that the AC-20 has a bit longer curve, but only by a few hundred RPM, and a sharper drop once it reaches that point. So that would put the AC-15 at 96V on the 7601 with a torque curve sitting at ~72ft-lb till ~3333 RPM, then dropping off. That behavior would explain the new winding for the AC-15 (I think it's newer....right?) as an attempt to flatten out the torque curve even more..or is that just a function of the higher current on the AC-20? I did notice that the drop-off on the AC-15 at 300A is shallower than the drop-off at 550A..hmm, but they both drop to ~17ft-lb at 6000RPM...which would imply that the higher voltage and higher current you put on, the sharper the drop-off will be as it drops to a certain point?

podolefsky
19 February 2012, 1714
I agree with your caution, it's just a rough estimate. But it should be pretty close for the same controller and motor. The 300A graph is for a different controller (1238-6301) than the 550 (1238-6501), so I'm not surprised the drop off is slightly different. The general curve is exactly what you'd expect though. Torque is basically proportional to current, so 550A/300A * 35 ft-lb = 64 ft-lb.

It's hard to compare all the graphs because they're all different motor/controller combinations. A few hundred RPM is about what I'd expect - since the AC-20 is about 10% longer, but wound to match the torque. So it makes about 10% more power by having about 10% higher RPM.

Roughly speaking...the AC-20 makes the same torque but about 10% more power. So you can gear it lower and get more rear wheel torque with the same top speed.

If your batteries can handle it, the extra current from the 650A controller will make more of a difference than the slightly higher RPM of the AC-20. I went with the AC-20 and 650A, best of both.

x88x
19 February 2012, 1949
Ah, yeah, I forgot they were running on different controllers; that explains the different drop-offs then.

If my plans come through as I'm hoping (wasn't able to get up with the guy about the roller today...hopefully will at some point this week), what I'm planning at the moment is a 96V 120Ah pack of (if I can get them) Sinopoly's LFP60AHA (B) cells. They appear to have the best energy density both by weight and by volume in the sub-100Ah market of large prismatic cells. Failing that, I'll be going with Winston LYP60AHA's, but those are 400g more per cell, or a good 24kg more for the pack. Either cell though should be able to push 650A briefly in a 120Ah configuration.

I'm hoping to be able to pick up a used motor and/or controller and save myself a bit of cash, so if I can find any used, that may dictate whether I go with the 15 or 20. If I buy new, the best price I have seen on them was...hmm, I'll have to find that link again...somewhere out on the west coast that I found a link to around here somewhere..and the 20 was only ~$50 more than the 15, both with the 7601. SO if I buy new, yeah, I'll just be going with the 20.

On another note, I know I've seen the AC-50 pushed up to ~130V; do you know if the 20 can also push that high? I know AC-induction motors are much more forgiving about high voltage than other motor types, so I would imagine that's more a limitation of the controller, and the AC-50 I saw that high was also on the 7601, so I'm guessing it would be fine but was wondering if you'd seen anyone do it. ;) I want to convert my truck at some point as well ('93 B2200), and I have this image in my head of building it in such a way that if I wanted to move my bike on the truck I could put it in the bed, secure it in place, and hook up a cable connecting it to the truck's power system to function as a booster pack to extend the range. I know I would need to plan the cells accordingly, so they could actually be paralleled and still work properly (ie, 100Ah2P instead of 200Ah1P on the truck if I do 100Ah1P on the bike), but if I'm building that in from the ground up it's less of a concern, so my main concern there would be bringing the bike voltage up to what could be a workable voltage for the truck...plus, it'd give me a higher top end on the bike. ;) It would also let me go to a higher energy density if I can't get a hold of the 60Ah (B) Sinopoly cells.

podolefsky
19 February 2012, 2057
Wow, that would be a big pack.

Yeah, it's the controller. The 1238 has an upper limit of 130V, and people have run it that high. The thing you need to keep in mind is that this is the maximum, not nominal voltage. On lifepo4, nominal is about 3.2V per cell, but fresh off the charger they'll be about 3.65. After they settle out more like 3.4.

So suppose you had a 36 cell pack. Nominal is 115.2V, max is 131.4, and after settling out they'll be about 122. That's probably the highest voltage you could run. If you tried for a 40s pack you'd be under on nominal (128V), but the pack voltage would sit more like 136V when charged and it wouldn't work.

x88x
20 February 2012, 0222
Good point on the nominal vs charge voltage; I'd forgotten about that. 36S it is then. ;) That would actually fit exactly into my calculations to match to 100Ah cells.

Yeah, it is going to be a big pack; I'm hoping that it will all fit in the 1100's engine area without sticking out too much. Unfortunately, I think I'm going to need that large a pack to get the range I need, so there's not a whole lot I can do to reduce the size. I'm considering looking at the A123 M20 cells if they continue to be available at such low prices out of China, but I would need to settle on a good packaging method for them before dropping the money on the cells. I've been watching Jack Rickard's attempts, and while they do continue to improve, I am still not quite sold on the resin method, just because I have this idea in my head that just in case, I would want to be able to replace a single cell from the pack. I've also looked at the Headway cells, but the cost of those cells is still just so much higher. And neither option gives me significant gains in volumetric energy density over the large format cells.

podolefsky
20 February 2012, 0849
Look at lipo. They're roughly 1/2 the size and weight of lifepo4 for the same Wh. If you get Turnigy packs they're about the same price, maybe a little cheaper.

Take a look at Ted's lipo thread (http://www.elmoto.net/showthread.php?1453-The-LiPo-Project-(The-R5e)&highlight=lipo+thread), and over on endless sphere. I honestly think there's no way you're going to fit 36S/100Ah of lifepo4 on the GS without it sticking out a foot on each side.

Richard230
20 February 2012, 1547
I can't help with technical stuff, but according to my 1983 Cycle World Road Test Annual, the Suzuki GS1100GL had a dry weight of 538 pounds. Of course, in those days, dry weights were very dry. The actual weight with a full tank of gas, oil, a battery, brake fluid and air in the tires was probably more like 580 pounds. :)

x88x
20 February 2012, 1829
Look at lipo. They're roughly 1/2 the size and weight of lifepo4 for the same Wh. If you get Turnigy packs they're about the same price, maybe a little cheaper.

Take a look at Ted's lipo thread (http://www.elmoto.net/showthread.php?1453-The-LiPo-Project-(The-R5e)&highlight=lipo+thread), and over on endless sphere. I honestly think there's no way you're going to fit 36S/100Ah of lifepo4 on the GS without it sticking out a foot on each side.
ES was actually the first place I hung out when I first started getting interested in this stuff. Not active over there any more, but yeah, I am familiar with Turnigy LiPo. I actually have some of the '20C' Zippy cells that I'm using to mess around with on a little scooter I picked up and run my lawnmower. My main concern there, though, is cycle life and stability. The cheap ones are also the least stable ones. ;) And last estimate I heard was only ~700 cycles for the non-nano-tech Turnigy/Zippy lipo at 80% DOD...which, granted, would give me over 50,000 miles on a 75 mile pack...and as long as I keep the temperature in a safe range, they should be ok...and I wouldn't be pushing them anywhere near their actual capabilities (figure ~50% of the rated for the non-nano-tech)...bah, idk; that'll be a decision for once I have the bike in hand and can measure out stuff. They are only ~30-60% better energy density by weight, but about 100% better by volume, which is the more important metric for these bikes...there's just still something that scares me a bit about having 11kWh of LiCoO cells between my legs, you know? There's also the additional weight and bulk of the mounting hardware to consider...geh, idk, like I said, it'll come down to what the actual measurements are once I get the bike.



I can't help with technical stuff, but according to my 1983 Cycle World Road Test Annual, the Suzuki GS1100GL had a dry weight of 538 pounds. Of course, in those days, dry weights were very dry. The actual weight with a full tank of gas, oil, a battery, brake fluid and air in the tires was probably more like 580 pounds. :)
Thanks, good to know. I wasn't able to get up with the guy for the bike this past weekend, but got up with him today and scheduled a time to come by next Saturday...and the CL ad expired and he says he doesn't really have anyone else interested in it, so I'm hopeful. :D

frodus
20 February 2012, 1908
The AC20 is a great motor. Everything Noah said is correct with respect to torque and RPM. Wound differently to give the same torque to a higher RPM. I'm very happy with the quality of my kit and Noah and a handful of others that have bought from me, seem to be pretty happy as well. In fact I bought this package and THEN became a dealer for HPEVS.

If you want to contact me (www.emf-power.com) I can try to answer some questions. Small website, nothing flashy and I'm straight about things with everyone. The I have is limited due to lack of more Dyno info. HPEVS gives graphs for some situations, but yours may vary a little. I also rent Programmers so you can change the battery and amperage parameters to match your pack and tweak regen and such.

I don't know of any used AC20 systems right now, but I do have someone who's got an AC15 system for sale for ~$2300 used. It has the 6501 controller so only 108V.

x88x
20 February 2012, 2006
Haha, funny story...that site I said I saw linked somewhere around here that had the best price I'd seen on an AC-20?

If you want to contact me (www.emf-power.com) I can try to answer some questions.
Found it! ;)

Hmmm, that AC15 is tempting... I want to have the roller in hand before I buy anything though, just in case. ;) The 6501 of course wouldn't be ideal (in my power-hungry brain), but would let me get rolling for a lot cheaper....and I could always upgrade later.

mpipes
21 February 2012, 0452
Haha, funny story...that site I said I saw linked somewhere around here that had the best price I'd seen on an AC-20?

Found it! ;)

Hmmm, that AC15 is tempting... I want to have the roller in hand before I buy anything though, just in case. ;) The 6501 of course wouldn't be ideal (in my power-hungry brain), but would let me get rolling for a lot cheaper....and I could always upgrade later.

You know, the difference in price between the kits only means some extra time to save up the money, and you KNOW you'll be happier with more power. ;) I bought an AC-18 kit from frodus, it's a combo of the AC-15 motor with the 7501 controller. It's all still in boxes because I still need to build a vehicle for it. :)

frodus
21 February 2012, 0911
Haha. Just keep my email handy and I can try to help when you're ready.

Sent from my SAMSUNG-SGH-I997 using Tapatalk

x88x
08 March 2012, 2218
Well, someone beat me to the GS1100, but I picked up an '85 Honda V30 Magna with a blown engine on Sunday. Should be very nice with the AC15 or 20 with 100 less pounds of bike. :D
EDIT: Build thread to be started soon.