Wait, I didn’t realize you were expecting me; I’m afraid I’m going to be late.
Just officially joined the forum after following your project. This is neat stuff
I have a few thoughts to share. I might be completely off here (I have not looked at every single detail of the past months), so just tell me if I am left field…
How much current Iin & Iout or you expecting in your motor controller? I saw 12s up there so you will be at around 50.4Vmax (@4.2V cell) and about 38.4Vmin (@3.2V cell). For a 31.5kW motor that will give you roughly (assuming eff = 1) → 31500W / 38.4V ~ 820 amps at the controller input. That is quite hardcore for a standard FR4 PCB, even on a multi oz copper… If you are planning to use FR4 you will want to make sure you do not get above around ~130C on the PCB. It will start delaminating at that point. Is this board going to be used as a first phase to test out the control and basic electronic design? Maybe this is the part I am missing…
Additional note, your initial FET selection (NVMTS0D7N06CLTXG) @ 60V Vds seems quite tight for a 50V input. You will certainly get some resonance on the full bridge due to parasitic inductance of your cables etc. Usually max input x2 is recommended to be on the safe side. You will pay a small price in Rdson increase on but that is usually a healthy compromise to avoid failures and the nasty repairs that come with it.
Check this chip this out if you have a chance: TMC4671 - Trinamic. It is quite a neat ASIC that implements full field oriented control that you might be able to utilize for your controller design. It packages a lot of the tricky embedded control stuff in one place. I know you are on another path but I thought I would share.
On a completely different note, do you have updates on the DHX motors? Would you know when they will be available and how much a P40 would got for (roughly).
Thanks for all the amazing work on this project. It is nice to follow. I thought I would share a few notes today that might be helpful. 100% trying to be constructive here
Hi… thanks for the questions and feedback.
First off DHX motors. They are making progress and have built a few of the P40s and P60s from what I have seen on their Instagram posts. They are now machine wound, but I’ve seen no indication yet as to when they will be available for general order at a much lower price than the hand-wound Hawk motors. I have hung onto my controller for the P60 hoping this would come through much sooner, but supply chain still seems to be somewhat of a constraint for them. Time to give them another phone call to see how it is going and if they have an ETA on when they will become generally available for purchase and at what price.
As far as my alternate SFPM motor design, yes those FETs and coils will need to take quite a bit of amperage - more than you would normally want to run as you state for use with PWM. Except that they are instead designed to be pulsed only once to enable the flux, and then again to disable the flux from the PM towards the rotor over a 20μs (or less) target switch time each. Even at high RPM (e.g. 40,000), the effective duty cycle is still quite low. Of course this “odd” coil use complicates the coil design balancing all the electrical factors required for a quick switch time, which is why I am still refining the PCB type design for a good compromise based on my (admittedly) limited magnetics knowledge. Every few weeks I test and then modify the design for (usually) better results. I’m not looking for “perfection”, just a reasonable POC motor design I can demonstrate and then go forward with starting a company and hiring mechanical, electronic, and magnetic specialists.
Good morning Bryan,
Thanks for the quick reply! Yeah, these DHX motors have fantastic specs… Let us know what you hear from them. I would love to get my hands on one of them. Power to weight is pretty amazing…
I just started a project myself, very similar to what you have done in the past. I just bought a 2019 Birel shifter chassis and looking to build and electric power system for it. I was initially contemplating buying a BSR 2.0 system… but given the cost… and the fact that it is fun to put together, I thought I would do it myself.
I actually have a ton of questions for you… I went through your gSheet in detail…
Looking at my different setup options, I found the ME1507 (air cooled) to be the most plausible candidate for my build. Simple goal would be to keep up with Rotax Max 125 on lapping sessions, have a battery swap system. The ME1507 is on the heavy side but I figured that with up to 44kW of peak power and 15kW continuous I would have plenty to work with to start. Price is decent, again, ~22kg is on the heavy’ish… What are your thoughts? The BSR setup seems to use a very similar motor (at least the housing is identical)…
I looked at others (like you did in your spreadsheet). The EMRAX are quite neat, but the 188 goes for 3.3kEuros… which is another level. Plus the out-runner design is a bit trickier to safely integrate in the build. I also heard they are quite a nightmare to tune and setup…
I found some chevy spark battery modules, 60aH, 100V, based on A123 lithium nanophosphate cells. Discharge current is plenty good and 60Ah seems about right, maybe slightly high…
Curious to hear from you what motor you would pick?
Were you able to finish the machining on the stuff I sent you?
Yes, as far as I can tell that is the motor that BSR is using. I’m not sure which controllers they use for the two variations the list, although many use the Curtis 1238-7971 controllers with it. Being commercial, they could be Sevcon units.
I agree there are other interesting motor choices - but yes, they are also pricey and as you note, some are awkward to use on a kart. Many of the FSAE teams run the EMRAX units for instance.
I’ve written DHX to ask about general availability and pricing today. I’ll let you know what they respond with. The P60 is what I wanted to use for sprint racing against shifter 125. But since that’s no longer a category at our local track, and going with something you can actually purchase using air cooling, either that or a Zero 75-7 would be good. The Zero motor is considerably lighter and would work good with those 100v modules, but unfortunately only available second hand from what I can tell unless you can find a Zero dealer that will sell a replacement motor direct.
Based on the below post, I added those modules to the spreadsheet under a new tab called Modules and uploaded it. What I don’t know is the drain or charge max C rates:
Do you have a tracking number? I never received anything.
I’ve been holding off sending out the machining order until I further nail down the coils. I don’t think it will require any change, but I want to make sure before spending any more.
My parents shipped it as I’m in college right now so I don’t have the tracking number. Maybe it got sent back?
@KMECHAD mentioned that the BSR system utilizes a Sevcon controller (Any questions about my ownership of the BSR2.0 Blue Shock Race fully electric powertrain? - #18 by KMECHAD).
I reached out to thunderstruck EV, they are recommending a Curtis 1238-7971 controller. I am yet to hear back form ElectricMotorSports. Sevcon and Curtis seem to be the front runners. Kelly and Fardriver seem like more affortable lower cost options… Any thoughts on these two?
The Zero 75-7 would indeed save me some kg’s and give me some extra power… but I am unable to find any for sale as you mentioned at least for now…
P60 would be amazing… Keep me posted. emrax is just super pricey and require specific hard to source controllers for their 188 low-voltage version… They do mention that Sevcon is in their recommended list but it does not show for the 188 LV…
The Spark modules are 28s3p A123 Lithium Nanophosphate cells rated at 20Ah (https://www.buya123products.com/uploads/vipcase/468623916e3ecc5b8a5f3d20825eb98d.pdf). From what I can imply from the nominal power rating these cells can do around 18C… which is a good bit. They are a bit on the heavy side though:
A123 20Ah Cell
20Ah * 3.6V (nominal) = 72W @ 0.496kg → ~145W/kg
4.2Ah * 3.6V (nominal) = 15.12W @ 0.0678kg → 223W/kg
Given the Molicel P42A is a pretty good cell (from what I gathered) the A123 cell is ok but on the heavy side… I am reconsidering whether or not it is a good choice… It is much cheaper for sure as I can have that 28s3p pack for 800CAD (640USD).
If I were to go for that option, I would buy two packs and reconfigure in three packs of 28s2p to get ~40Ah and shed some weight. That would run me around 70 lbs where a Molicel P42A would be around 36 lbs… 2x the weight does not sounds very good at all… But it would cost be 1125USD… just about twice the price…
I’m not familiar with Fardriver (QS controllers) myself. I have two Kelly controllers that are easy to program. They do require either a resolver or encoder depending on the model, and have no auto-learn. I have been told that they are a bit “soft” from a dead start, even when set to no delay, but I can’t confirm this. I was planning to use the Kelly KLS14401-8080IPS I bought second-hand with the Hawk 60 I purchased, or now a Peregrine 60.
DHX replied that they should be available for general orders “early next year”. No pricing yet.
I have a really stupid question about electric karts and decided to post it here since it seems like the most general thread about them:
Is there a reason that it seems no one has thought to pair a smaller electric motor with a gearbox so that the motor and batteries could be smaller/lighter/longer-lasting without sacrificing performance? I’m sure the spacing on a gearbox would need to be different than on a combustion engine (and maybe there could just be less gears altogether, 2-3 gears vs 5 or more), and I don’t know what the other challenges are, but it seems like it could solve a lot of issues/liabilities that popup with current configurations.
Like I said, probably a stupid question.
A gearbox does not give you more power, it only matches your mechanical load and motor. For example, you could take a super small motor and move a kart, but you would need a large reducer to get enough torque to move the thing. The result would be a super slow kart.
Power = torque * angular velocity
Adding a gearbox (typically reducers, increase torque and reduce speed) will give you more torque, but less angular velocity, the output power will be constant (ignoring efficiency for this conversation).
If you have a large powerful motor, then you can have both torque and speed.
Well said. The Dodge Viper of the 1990s is a good example…. Massive motor which allowed for a high top speed and pretty good acceleration. The large motor also helps “keep things simple”.
I could be wrong Gary, but I think what @Francois_Giguere is alluding to is that peak output power is not produced on a curve with an electric motor, whereas peak power for a combustion engines is. I think the choice to not pair a gearbox is less about simplification or ‘go big or go home’, than it is that because that the way peak power is produced for a motor reduces many/most/all of the advantages of using a gearbox that would occur for a combustion engine. At least that’s my understanding from his answer. Or maybe I’m completely lost.
Good point. I mis-read it.
That would be correct. Since you have high torque from 0 RPM you normally don’t need gears as such, but just a difference in the sprocket ratio to match the peak RPM power to the speed you need the wheels to turn. Of course depending on the motor, with a minimum front sprocket size it may require a rear sprocket that is larger than possible on a kart due the small diameter of the tires. That is where a reducing “gearbox” of some sort would be nice to have, but with added cost and complexity it is generally not done.
Pinged DHX again today, about 7 months since the last time since they had posted this on Instagram:
Hopefully yes. Maybe they’ll reply tomorrow.
I also pinged my local Zero Motorcycle dealer about selling me 75-7 motors. I mean, Zero isn’t really doing so great as a business and I would think they would want as many sales as possible. Nope, have to have a serial number for a failed unit and then send it back. No sales outside of that because… warranty and liability issues!
- It’s for racing, warranties aren’t a thing as long as it’s not DOA.
- As far as liability goes… I guess you can’t buy much of anything from US manufactures for the fear you might do something dumb, but no problems with Chinese ones because they’re harder to sue internationally? Is that screwed up or what? Sigh…
DHX P40 looks like it may be the winner. But in the mean time I also found this not yet in full production Saietta AFT140i 96v motor which could fit the bill. Although a little heavier than I’d like, it does include the inverter in the liquid cooled motor unit, which is a nice touch packaging wise. I talked with one of their salesmen that used to race karts in the UK, so he understood the needs and what I am trying to do, which was quite refreshing. For now they are produced in the UK, so pretty pricey, but they plan to mass produce them in India next year to substantially lower costs and serve that huge domestic market.
I wish we could have electric rentals with hot-swappable batteries and 15-18hp with 1h run time. Endurance racing is a big part of rental racing and eKarts would allow a lot of outdoor tracks to run 24h races
Me too… There is an old local track that now has apartments nearby. With electric karts it could still possibly be used.
But current batteries frankly aren’t that great. By my best area under the curve estimate (I could really use some dyno charts), to roughly match a IAME x30 125cc TAG time on an “average” sprint track for an hour would take a 12.6 kWh battery. At about 250 Wh/kg, that comes to 50.4 kg / 111 lbs, just for the cells. At least double that for a long track with more WOT. Compare that to about 1.5 US gallons of gasoline at about 9 lbs and you can see why having removable hot swap batteries is a requirement. Improvements like “solid state” batteries might double that Wh/kg figure. But it will still pale in comparison to hydrocarbons as far as energy density goes.
That said, the big pluses for electric karts are:
- Nearly silent. Listen for the tires loosing grip…
- Instant response and power. With no clutch or gears, the only slack in the system is the chain.
- Near Zero maintenance. Drain water coolant before winter storage. Batteries should last 5-10 years unless abused. Possibly a replacement of two shaft bearings after several hundred track days. Then just normal kart maintenance like chains and sprockets.
- No oil smoke!