The rears are only on to work on not run, normally it has slicks all round
The Sledgehammer - Shooting for 1,000hp
- Thread starter AlexLTDLX
- Start date
Good news, all. I got it wired up and running (on 8s, anyway). Since long cables have been such an issue, I decided to do my initial low-power testing with the actual cables I'll be using in the car. The motor cables are 4 gauge, 18.5 feet long each (combined, they weigh about 20 lbs (!)). The sensor cable is an extension I made up from a CAT 8 ethernet cable (they're shielded, and the MGM ESC has a connector for the sensor shield), and is comparably long.
Overall experience so far: setting up was much easier than the VESC; though the manual suffers from the typical, "written by the engineers who designed it vibe" - for example, there is no description up front of what the LEDs mean (that's all the way back on page 69), and there's no statement that says "you must do this first and see these LEDs for it to work" in the first few pages of the manual - that cost me probably about an hour and a half to figure things out. The other downside is the English translation is kind of poor - they're from the Czech Republic, so that's to be expected, but it can be frustrating when things don't make sense. It's not "chinglish" level bad, but it does add another level of challenge, especially since you want to do your best not to blow up a $2,000 ESC. Finally, there have been changes to the software since the manual was written, so certain things are labelled differently and other things don't even exist anymore.
But, all told, the experience wasn't too bad. So how does having a sensored motor change things? Well, you can start it super-smoothly and it can run smoothly at really low speeds (not necessary for our purposes, but it does make it easy to make sure it's spinning in the right direction.) The start-up has been the issue with the other setups - which the MGM/LMT setup has covered wonderfully in spades. Does it affect high-power running? Don't know yet. That'll be the next youtube video. Just running the bare motor it performed beautifully all the way to slamming the throttle. Once connected to the blower, I could only go about half throttle on 8s before it started blowing itself around the table.
Have a look see:
Overall experience so far: setting up was much easier than the VESC; though the manual suffers from the typical, "written by the engineers who designed it vibe" - for example, there is no description up front of what the LEDs mean (that's all the way back on page 69), and there's no statement that says "you must do this first and see these LEDs for it to work" in the first few pages of the manual - that cost me probably about an hour and a half to figure things out. The other downside is the English translation is kind of poor - they're from the Czech Republic, so that's to be expected, but it can be frustrating when things don't make sense. It's not "chinglish" level bad, but it does add another level of challenge, especially since you want to do your best not to blow up a $2,000 ESC. Finally, there have been changes to the software since the manual was written, so certain things are labelled differently and other things don't even exist anymore.
But, all told, the experience wasn't too bad. So how does having a sensored motor change things? Well, you can start it super-smoothly and it can run smoothly at really low speeds (not necessary for our purposes, but it does make it easy to make sure it's spinning in the right direction.) The start-up has been the issue with the other setups - which the MGM/LMT setup has covered wonderfully in spades. Does it affect high-power running? Don't know yet. That'll be the next youtube video. Just running the bare motor it performed beautifully all the way to slamming the throttle. Once connected to the blower, I could only go about half throttle on 8s before it started blowing itself around the table.
Have a look see:
Here's some data to look at:
So we hit about 2.16kw.
I hit 54% throttle; you can see the 8s pack wasn't fully charged, and sagged pretty badly - going from about 30 volts to about 27. And we hit about 78 amps.
And we got to an impeller speed (this was wide open; no restriction, but probably not fast enough to go into choke) of about 10,400 rpm.
As I said above, next up, high power tests - we'll do both with a restrictor (I'll probably limit max motor rpm to 50,000 so I don't hurt it - I have a feeling this thing could exceed that through a restrictor), and then fully loaded.
So we hit about 2.16kw.
I hit 54% throttle; you can see the 8s pack wasn't fully charged, and sagged pretty badly - going from about 30 volts to about 27. And we hit about 78 amps.
And we got to an impeller speed (this was wide open; no restriction, but probably not fast enough to go into choke) of about 10,400 rpm.
As I said above, next up, high power tests - we'll do both with a restrictor (I'll probably limit max motor rpm to 50,000 so I don't hurt it - I have a feeling this thing could exceed that through a restrictor), and then fully loaded.
WB projects
Active member
Niiiice
WB projects
Active member
Can't wait to watch the video! Is it clost to your expectation ? Or it go past your expectation?
I would say from an actual performance perspective, it's on the low side of what I expected (which is still good - 800hp should be obtainable on regular gas, 850+ on fuel with an oxygenate, like Q16, or even throwing some Klotz oxygenate into the tank). Electrically, it's pretty far beyond what I expected. We saw a peak of 675 amps and a continuous average of 650 amps running it wide-open into choke. This is practically double of what I've seen before.
The really nice thing is it just worked. No endless screwing around trying to get it to work reliably and consistently day to day. I left the same "tune" in the ESC from my low power tests inside the day before, took it out to the garage with the 5.6m (18.5 foot) motor cables & sensor cables, and it just worked. The only adjustment I made was motor timing, but that was to try to find a little more RPM - which it did, but not by much. Compared to the VESC, which would work one day, and the next day it wouldn't even spin with the exact same untouched settings. But I do believe that having a sensored motor makes all the difference, as does a motor with fewer poles (the LMT motor is only 2 poles, the TP Power is 6 and the sensored Castle Creations motor I just bought to replace the TP Power motor is 4 poles - I now expect the VESC/P2 setup to work without too much issue as well).
But going through all this empirically, I can now say I have learned more about brushless motors and controllers than I ever thought I would. And a lot of that has to be credited to the struggles with the VESC.
The really nice thing is it just worked. No endless screwing around trying to get it to work reliably and consistently day to day. I left the same "tune" in the ESC from my low power tests inside the day before, took it out to the garage with the 5.6m (18.5 foot) motor cables & sensor cables, and it just worked. The only adjustment I made was motor timing, but that was to try to find a little more RPM - which it did, but not by much. Compared to the VESC, which would work one day, and the next day it wouldn't even spin with the exact same untouched settings. But I do believe that having a sensored motor makes all the difference, as does a motor with fewer poles (the LMT motor is only 2 poles, the TP Power is 6 and the sensored Castle Creations motor I just bought to replace the TP Power motor is 4 poles - I now expect the VESC/P2 setup to work without too much issue as well).
But going through all this empirically, I can now say I have learned more about brushless motors and controllers than I ever thought I would. And a lot of that has to be credited to the struggles with the VESC.
TBrick - Through the restrictor? I don't know. To be honest, that's kind of a BS number anyway. It doesn't reflect the dynamic conditions it would see on the engine. That's why I didn't bother to hook up a boost gauge last time. It was taking people off topic (judging by the comments). I can tell you I expect to see about 8 psi or so at peak hp and about 10 psi peak off the trans brake. Give or take a couple of psi.
cmoalem - It's not just the direct drive that makes it smooth. The two poles help quite a bit as well. Seems the fewer poles you run, the less "magnetic cogging" you see. It's true the belt drive option does add a little noise, but I really think it's more from the motor itself. As for what more can I add - well, I've been already been thinking on that. We've tested both extremes - high RPM and high load (i.e. power). Voltage drop, as always, is our enemy (at least once stuff stops blowing up). Oddly, the long cables are only responsible for about 2 volts of that; the rest is batteries. I have more than enough Lishen cells to add another set of packs in parallel. That would, in theory, cut the voltage drop in half. So that would be the obvious thing to do.
Let's do some bench racing: in the tests in the video, the packs were charged to 61 volts; at 678 amps (that was the peak, but average was still 650), the pack voltage dropped from 61 to 48 volts, giving us 9 volts of pack drop (remember 2 volts from the cables - it's really a hair over 2 volts at 678, but close enough). The caps in the ESC are 63 volt caps - that would be the obvious limit there. So let's say we can safely charge the packs up to 62.5 volts; but let's call it 62. In the car, let's say the e-turbo will draw 550 amps; so pack drop (more or less linear) should be about 7.5 volts. By adding an additional volt by charging higher and 1.5 volts with less current draw, we'd be at 50.5 volts under load. KV isn't constant - under the max load in my tests, we were at about 585 KV. in the light load (i.e.restrictor) tests, KV was about 705. A safe bet for KV under the 550 amp load is 640. So we can expect 32,320 impeller rpm. that's almost 10,000 rpm less than I was hoping for. If we add a second set of packs, we should pick up another 3.5 volts, for 54 in total. That would still only get us up to about 34,560 rpm. BUT - and this is a big one - with the APD/TP Power setup, we only saw 26,000 impeller rpm and got 4 psi of boost. If we only see 32,320 rpm, we're only going to pick up a pound of boost at peak hp. But that's not the way it works with this setup for some reason. According to the Vortech Si compressor map, at 4 psi and 26,000 impeller rpm, we should've only made 600 hp. But we made more - 50 hp more, going by BSFC. That additional 50 hp likely comes from the fact that I'm not running the Vortech cast impeller, but a billet impeller - which gives a 5mm larger exducer diameter, is considerably lighter and has a much narrower center section for more flow. The compressor map tells us we should see 720 hp. But add in the additional 50 hp for the billet impeller, and you're at 775 hp. And just because I think it'll do a little better (I mean, c'mon - it's going to be right at double the power of the previous setup - right around 28 kW), give me 800 flywheel. Throw some oxygenate in the tank, and there's your 850. Oh, and it should be a little over 6.5 psi at peak hp and 8.5 at peak torque.
See - isn't bench racing fun? I just gave myself 75 hp more than the graph tells us it'll make! Lol. But the billet impeller gains are real. Actually, going through these numbers is helpful to me to zero in on what to reasonably expect. Assuming everything holds together on the dyno, I won't be happy until I'm over 600 rwhp through the glide. It'll take 800 flywheel hp to get there.
BTW - This is the kind of stuff that keeps me awake at night. Just not knowing what reality brings over theory...
This post got a bit out of hand. Sorry.
cmoalem - It's not just the direct drive that makes it smooth. The two poles help quite a bit as well. Seems the fewer poles you run, the less "magnetic cogging" you see. It's true the belt drive option does add a little noise, but I really think it's more from the motor itself. As for what more can I add - well, I've been already been thinking on that. We've tested both extremes - high RPM and high load (i.e. power). Voltage drop, as always, is our enemy (at least once stuff stops blowing up). Oddly, the long cables are only responsible for about 2 volts of that; the rest is batteries. I have more than enough Lishen cells to add another set of packs in parallel. That would, in theory, cut the voltage drop in half. So that would be the obvious thing to do.
Let's do some bench racing: in the tests in the video, the packs were charged to 61 volts; at 678 amps (that was the peak, but average was still 650), the pack voltage dropped from 61 to 48 volts, giving us 9 volts of pack drop (remember 2 volts from the cables - it's really a hair over 2 volts at 678, but close enough). The caps in the ESC are 63 volt caps - that would be the obvious limit there. So let's say we can safely charge the packs up to 62.5 volts; but let's call it 62. In the car, let's say the e-turbo will draw 550 amps; so pack drop (more or less linear) should be about 7.5 volts. By adding an additional volt by charging higher and 1.5 volts with less current draw, we'd be at 50.5 volts under load. KV isn't constant - under the max load in my tests, we were at about 585 KV. in the light load (i.e.restrictor) tests, KV was about 705. A safe bet for KV under the 550 amp load is 640. So we can expect 32,320 impeller rpm. that's almost 10,000 rpm less than I was hoping for. If we add a second set of packs, we should pick up another 3.5 volts, for 54 in total. That would still only get us up to about 34,560 rpm. BUT - and this is a big one - with the APD/TP Power setup, we only saw 26,000 impeller rpm and got 4 psi of boost. If we only see 32,320 rpm, we're only going to pick up a pound of boost at peak hp. But that's not the way it works with this setup for some reason. According to the Vortech Si compressor map, at 4 psi and 26,000 impeller rpm, we should've only made 600 hp. But we made more - 50 hp more, going by BSFC. That additional 50 hp likely comes from the fact that I'm not running the Vortech cast impeller, but a billet impeller - which gives a 5mm larger exducer diameter, is considerably lighter and has a much narrower center section for more flow. The compressor map tells us we should see 720 hp. But add in the additional 50 hp for the billet impeller, and you're at 775 hp. And just because I think it'll do a little better (I mean, c'mon - it's going to be right at double the power of the previous setup - right around 28 kW), give me 800 flywheel. Throw some oxygenate in the tank, and there's your 850. Oh, and it should be a little over 6.5 psi at peak hp and 8.5 at peak torque.
See - isn't bench racing fun? I just gave myself 75 hp more than the graph tells us it'll make! Lol. But the billet impeller gains are real. Actually, going through these numbers is helpful to me to zero in on what to reasonably expect. Assuming everything holds together on the dyno, I won't be happy until I'm over 600 rwhp through the glide. It'll take 800 flywheel hp to get there.
BTW - This is the kind of stuff that keeps me awake at night. Just not knowing what reality brings over theory...
This post got a bit out of hand. Sorry.
Chevibubba
New member
Nonsense! By all means, continue.
The bench racing tends to share the data & experience u have gained in a useful manner for all to see. I sooo wish I was in a better position to participate. I could sell one of my racebikes but that may contribute to my demise,(the wife is kinda emotionally attached to both). I am quite excited about the progress. I'm sure you've thought of this but; why not a voltage controlled relay circuit to latch additional batteries once the voltage drops? Maybe too much additional weight?
Installing the thing today. All the brackets, cables, etc are already made. I'm also buying a piece of property on the Texas Gulf Coast (was there last week), as well as fighting with lying russian trolls on news sites. And then there's the rest of life getting in the way. I wish this was l all I had to worry about.
Those black rubber round things are neodymium magnets to hold it to the underside of the package tray.
Those black rubber round things are neodymium magnets to hold it to the underside of the package tray.
win? of COURSE win!
Bit this violent jolt is scary sweet... pretty sure you want to reduce the lag ... and then control the hit as well!
Like this its like a "eject"
(closed loop to boost read out with a "target boost" table???
)
I do need to contact the MGM/LMT guys and see how comfortable they'd be getting the spool up time down. I'd like to get it to at least 1.5 seconds - realistically, that would still be a 2 second spool up time in the real world. In my case, the hit's not an issue - I want as much as I can get. But one of the comments on the video gave me an idea. How about a speed sensor on a front wheel and another on a rear wheel. When there's an rpm difference (under boost conditions), you could have the ESC throttle back the e turbo. Voila - traction control!
SGL12GUNNER
Member
That's not a bad idea at all!
Adding more mechanical grip would be preferable. However tires are a consumable performance modification. That may be the ticket! I know on your other setup you were running radials and they still broke loose a little bit.
It would be nice if you could make 8psi but I think that may be a stretch. It is exciting to see that it already makes more boost than the TP power set up With a little on the table.
I can't wait to see it at max power!
Adding more mechanical grip would be preferable. However tires are a consumable performance modification. That may be the ticket! I know on your other setup you were running radials and they still broke loose a little bit.
It would be nice if you could make 8psi but I think that may be a stretch. It is exciting to see that it already makes more boost than the TP power set up With a little on the table.
I can't wait to see it at max power!
On the track I'll go back to the slicks. The car does hook pretty hard - best 60' so far has been a 1.41 on a small 26 x 8.5 slick. I'm with you - I'm not sure we'll see 8 psi at peak engine hp, but I'm pretty sure we'll see more than that a lower rpm - like on the transbrake. I think most likely we're looking at around 7 psi or so. But then again, there are still things I can do to get more out of it. For example, I have enough Lishen LTO cells to make a 2p24s pack - right now it's just a 24s pack. That should give us a few more volts to work with and at least another 1,200 impeller rpm. That might be enough to hit 8+ psi. And that should be more powerful than the Whipple I had on before that hit 15 psi, and went 9.83 @ 140.07 mph. Plus, the weight distribution is better with this setup. Almost all the supercharger-related weight was in the engine compartment. Now, 75% of that weight is directly over the rear axle.
We can also play with motor (not engine) timing - that might buy us a little more impeller RPM too. If I really want to optimize to the max, I could shorten the motor cables by about a foot to 18"; I'm already planning on taking about a foot out of the power cables - but that's mostly just to neaten things up.
I need to make a proper battery box mount first. Probably out of bed frame angle iron, bolted to the rear seatbelt mounts. And then I'll probably have to re-make it to go to 2p24s packs... I seem to do a lot of re-making... lol.
We can also play with motor (not engine) timing - that might buy us a little more impeller RPM too. If I really want to optimize to the max, I could shorten the motor cables by about a foot to 18"; I'm already planning on taking about a foot out of the power cables - but that's mostly just to neaten things up.
I need to make a proper battery box mount first. Probably out of bed frame angle iron, bolted to the rear seatbelt mounts. And then I'll probably have to re-make it to go to 2p24s packs... I seem to do a lot of re-making... lol.
BTW - The MGM guys said I should be fine with dropping down to a 1.5 second spool up time; and then after they look at the logs again (I assume they're looking for current spikes), they think we could still go quicker - like 1 second or so.
There is an issue with MSA's arduino code when using the potentiometer - when the pot's set to max, the ramp up time appears to go away - it just snaps from throttle at zero to throttle at 100% after the first time it's triggered. I've reached out to him, but haven't hear back. Anyone want to look at the code who knows what they're looking at? He did a great job; it's above my head. Software isn't my strength.
There is an issue with MSA's arduino code when using the potentiometer - when the pot's set to max, the ramp up time appears to go away - it just snaps from throttle at zero to throttle at 100% after the first time it's triggered. I've reached out to him, but haven't hear back. Anyone want to look at the code who knows what they're looking at? He did a great job; it's above my head. Software isn't my strength.