This was a good week. Kind of strange, but very good.
The good part was that Sebastien Bourgois, the head of EVnetics, travelled at his own expense from St. Petersburg Florida to our location in Cape Girardeau Missouri to address our Soilton1 weeney power problem. I would guess he had it fixed in about 20 minutes.
[jwplayer file=”news040111 – iPhone.m4v” hd.file=”news040111-1280.mov” image=”http://media2.ev-tv.me/news040111.jpg” streamer=”rtmp://s2v8uso6bi7t47.cloudfront.net/cfx/st” provider=”rtmp” html5_file=”http://media2.ev-tv.me/news040111 – iPhone.m4v” download_file=”http://media2.ev-tv.me/news040111-1280.mov”]
The problem, from what I understand, is that the Soliton1 was reading it’s output current at 1000 amps and of course dutifully limiting it to that.
It is safe to assume that the Soliton1 uses a Hall-effect current sensor, and since we use those for many, but not all current sensing in electric cars, it might bear some examination.
Edwin Herbert Hall discovered the phenomenon while he was working on his doctoral degree at Johns Hopkins University in Baltimore, Maryland in 1879. Hall noticed that if he applied a small magnetic field to a section of conductor that was carrying current, an electrostatic field would be caused traversing the conductor. Essentially, the magnetic field caused all the free flowing electrons to move to one side of the conductor and the positively charged “holes” to the other. This voltage was both measurable and proportional to the level of current in the conductor.
And so a hall effect current sensor has a small loop enclosing the conductor and a small magnetic field is generated by the loop. Then the voltage from one side of the loop to the other is measured and output as a signal.
Typically, this signal is very low level. The LEM HAS sensors we use, for example, are bidirectional. The power we provide to it is 5v and ground and the resulting output signal is 2.5v with no current through the conductor. Current in one direction will cause this to rise to as high as 5v, and current in the other direction will cause it to fall to as little as a bit over 0v.
Normally, we would take that signal and apply it to an analog to digital converter to convert this value to a digital value that we can use for intelligent purposes. A 10 bit A/D converter would then output 1024 possible values with 512 representing the 2.5v and 1024 representing 5 volts for example. In the case of a 16 bit A/D, you would have a better resolution with 65535 possible values with 32767 representing zero current and 65535 representing 5 volts for example.
There are two problems with this. The digital output is quite precise, but the analog output of the sensor is somewhat less so. The 2.5 volts wanders around a bit with temperature, and so your 32767 value for zero amps is a little squishy. Normally, you would temperature compensate this numeric value by having another A/D input for temperature. You might also give it a little room and declare a plus or minus 25 count as still representing zero amps, in software.
From there, you get to scaling. If the sensor is a 200 amp sensor for example, 5 volts might represent 200 amps. Unfortunately, the inexact nature of the sensors means they might build ti for 200 amps and it actually indicates 186 amps or 214 amps.
And so the device has to be calibrated. Fortunately, this is quite easy. Let’s say 1000 amps is nominally 5v and we have a very nice zero point at 32767. If we divide the remaining 32762 value by 1000 we see we have 32.762, or very nearly 33 as the value of ONE amp. We can calibrate this by actually measuring a known current value, and changing the 33 to 32 or 31 or maybe 34 or 35 to get a digital output corresponding with our known current level.
The astute among you will immediately see the problem here. Where do we get an exact 1000 amp current? Ok, we can calibrate it at 200 amps. That moves the problem to the still difficult, where do we get a precise 200 amp current to use to calibrate.
The problem in our Soliton1 was that its calibration was off, and at about 700 real amperes, it read 1000 amps. The iGBT’s in the Soliton1 are undoubtedly good for a higher current level – you would typically do this to provide a little “headroom” from the spec so you don’t easily blow up the component. I’m guessing 1400A IGBT’s. But then in software you “limit” the current. In this case, the current limit is of course set to 1000. We can in fact, in the configuration screen, limit it to a lesser value. But we cannot limit it to a greater value.
The fix is simply to recalibrate the Soliton1. IN our case, it was easier for Seb to swap it out.
Since we already had three other Soliton1’s sitting around, why didn’t we just swap one out. Well, duh, because I thought it was working. It operated fully and smoothly, and 722 amps at 160 volts is not an irrationally minute amount of power in itself. That the Soliton1 was reporting 1000 amps while I can’t measure 1000 amps made me somewhat skeptical of their advertised claims. If they claim 1000 amps, and report a 1000 amps on the log, if you never measured you would of course believe you were getting 1000 amps. Not that any vendor would do such a thing………
As it turns out, I do not believe they were doing it on purpose. When he swapped the Soliton1, it immediately made 1000 amps that we COULD measure externally as 1000 amps. There’s no purpose in rigging such a thing if you really can do 1000 amps. ANd indeed, as Yogi notes, “It aint’ braggin if you can do it.’
So I’m persuaded they simply have a calibration issue and further that it’s probably just procedural – going back to where do you get a good 1000 amp calibration “signal” and a lab certified ampmeter as well. Or even a 200 amp source. Well, you see the problem. I have no idea what the calibration procedure is and Seb didn’t share, but he did state in a clearly annoyed fashion that they were going to revisit the issue.
Meanwhile, he and Brian took a test drive and immediately burned up a Stage II Kennedy clutch. Brian had CB Performance overnight a stage iV Kennedy clutch and 3000 lb pressure plate. Man are THOSE guys good. It was hear the next AM. He and Matt pulled the motor, swapped the clutch and pressure plate, and had it all back in by noon. THEN we blew the 400amp fuse on the system. We had simply forgot to upgrade it when we rebuilt Redux.
As George Hamstra notes in his ongoing redesign of the Warp series of motors – “If you beef up one thing that fails under extreme load, you simply move the problem. Next time then something ELSE will blow.”
But by Friday afternoon, Brain had done a test run with a zero to sixty time of 6.96 seconds. He relates that at 6.98 in the video.
As our shift points will have shifted given the higher power, we expect to improve that very slightly on the dynomometer this week. And it will be interesting to graph the results there. I’ll work on improving the graphs. Now that the reported amps and the measured amps are pretty close, I can probably use the log file to back up our video technique of matching amps/volts to the dynomometers output hp/torque. I’ll work on it.
Motor amps vs Battery Amps.
This is entirely forum speak. And I’ve avoided it. Enraging both the forumites and forumicators. From my perspective, everyone who raised their hand and brought this up is a moron in public. I understand that that IS the minority view.
If we treat the controller, motor, and transmission as a black box, we don’t need any test equipment really. Many of our viewers have EV’s, and many of them do NOT have $20,000 in test equipment laying around on the off chance that they MIGHT want to measure something like that someday. I not only show what we do on the video, I am very careful to TRY to do it in such a way that YOU can do it yourself and either get the same result, or get a different one. In this way, you can determine if what we say is real, or if we have made an error. And you do not have to depend on the overabundance of self appointed experts on the forums to “interpret it” for you.
Almost all EV’s have instrumentation to show battery amps and pack voltage.
A dynomometer is a heady investment. Fortunately, so heady that almost every town (except Cape Girardeau) has one where you can rent some time for a hundred bucks or so. They will give you precisely what your output horsepower is and what your output torque is by RPM.
IF you have the ability to note the current and voltage at those RPM, you can correlate this very nicely. You will put IN an mount of power calculated by multiplying the voltage times the current and expressed as watts. You can even calculate this as horsepower by dividing the watt value by 746.
You will note a disparity between what you put in and what you take out – that is the efficiency of your drive train. You will typically get OUT 80 to 85% of what you put in.
And all that is related to the limits of your controller, motor, and in this case clutch, but primarily your controller.
The difference between battery amps and motor amps doesn’t matter because they are going to be the same. When you get on the dynomometer, you’re going to mash that accelerator and it will accelerate as best it can right up through your maximum RPM. And at all those values, you are going to match on both sides of the controller the same amp rating to within less than 1%.
The ENTIRE purpose of this PWM waveform is to provide LESSER amounts of power for lesser throttle positions, and you’re not doing any of those on the dynomometer. BACK EMF, power peaks, etc etc ad nauseum do NOT have any effect on this. Your efficiency actually varies a little across the RPM band. And both the current and voltage can vary, but the relationship between battery and motor amps won’t and that is why I have said they are the same and they do not matter. And they do not.
Dragging it off into discussions of PWM waveform analysis was not where I wanted to go and I refused to do so. It had NOTHING to do with whether or not the Soliton could put out 1000 amps ever.
Ensued a big discussion of what Jack knows and what Jack doesn’t know. It isn’t, and never was about Jack. I don’t even know how to address all the things I know and don’t know to this group. It would astound you in both directions. I’ve lived a very large life on a very wide front and to a very deep depth. And at 55 years old, I’m just really really comfortable with it. In almost all cases, I attune my output to what I think is useful to the person I’m talking to. In thiis case, it’s a rather wide group with doctorate level viewers in materials science and battery research, and guys who can’t put batteries in a flashlight and get it right before the first try.
HERE is a huge line of distinction. MOST of what I have read and you might read on the forums has little do to with anything beyond the personal self aggrandizement of the posters. They are very insecure in their lives and what they know and will seize any opportunity to demonstrate any knowledge they think they have, and in most cases have part of and have that part misunderstood, in an effort to have YOU believe they are an expert in the field. 90% are outright poseurs, but even those that are doing it I would observe are MOSTLY doing it for the personal satisfaction of personal self aggrandizement.
I don’t live in that world. I don’t need to. That’s not why we have an EVTV. I already thought I was a thorough genius and fabulously wealthy and successful person on such a wide array of fronts you don’t even know about that I have ZERO interest in it at all. Actually probably negative numbers. Here in Cape Girardeau I live very quietly in very modest digs and spend most of my time trying NOT to call attention to myself.
EVTV is not about me and what I know or don’t know. I just can’t share all that with you.
I see a “perfect storm” of issues and problems that encompass our air, our money, our kids, our troops, our relationships with other countries, our idiot politicians, and a very real potential for total economic collapse that will effect us all. It has a lot of aspects, but an UNUSUALLY NEAT little Nexis centered on about 1/3 of our energy use for personal transportation purposes. And we don’t need the whole hog to win. if we translated 20% of just that 1/3, using a readily available efficiency multiplier of about 8 that we get by using electric motors instead of internal combustion engines to drive our personal cars, MOST of it all moves out 50 years. That’s plenty of time to convert everything ELSE to electric and make the whole thing eight times more efficient.
And to do THAT, I am absolutely certain of what I need and it would surprise you no end. It’s not 300 milliion people, nor General Motors, nor Barrack Obama.
If we could get 10,000 guys to build an electric car, and use it to each persuade 10 other guys to do likewise, we have 100,000 people intimately familiar with electric drive automobiles. The advantages are more persuasive than having a global Internet. And I watched just that happen in those numbers on my watch.
And it will happen here as well. I get up at 5 AM and go at it till midnight and I do that seven days a week until it is done or the big heart attack removes the problem for me. It’s not the 100,000 that’s hard, it’s the first 1000. Once it hits critical mass, I’m sipping whiskey and looking at the river.
AND YOU’RE A LONG WAY FROM THERE YET. You don’t have probably 300 LiFePo4 builds worldwide.
In the meantime, I’ll do whatever I can to persuade you, to enable you, and to goad you into action. But it never was about me. And I have no need to be politically correct, give up smoking, give up whiskey, flex my muscles, do my hair, wear a necktie, hide my microphone, or demonstrate any particular knowledge or expertise so anyone thinks well of me. We seek advertisers and sponsors and we don’t need ANY of them. THEY will be better off for getting on the train and they can enable YOU. If I want to call you all Nappy Headed Whores, you’ll hear it and CBS hasn’t got shit to say about it. If you try to put me in a box like CNN or Mythbusters or Senior/Junior or whatever, there’s a REASON we’re not on cable. It would be a mismatch from the get go. The cable networks would never do precisely what I told them to do 12 minutes before I thought of it and they would be immediately dismissed from my own little insular world for even thinking of such irrational insubordination.
So EVTV is never going to look like television you are accustomed to watching. And I’m not a forum guru. And there are reasons for all of that. But I am single mindedly devoted to the notion of persuading YOU to CEASE being a VICTIM of oil companies, governments, global economics and the uncertain future. I happen to know it really IS what you make of it. You have no CONCEPT of how powerful you already are. Go to your garage NOW and sweep out a space about the size of a car. That’s the first step….. Now picture a car in your mind…. that’s the second step. The rest of the steps are so easy we can do a video while we’re doing it.
Boy, rainy morning has me all smokey.