Hello to All,
We had left off:
After the power wheelstand at speed…I had seen enough and made the decision we would not be going back out. There was a new forecast calling for mid 70s on Saturday, and with the prospect of a warmer and hopefully stickier track, we were determined to return the next night to get the job done.
Saturday arrived with more ocean clouds blanketing the Portland metro area, but it was still warmer than Friday’s unfriendly weather. We all gathered at the Village Inn for the traditional Electric Breakfast show & shine EVent, and it was as usual, a great time for all. As the hours clicked by, the clouds gave way to warm sunshine and clear skies. We had all hoped for this so the track could be heated up all day for a better night of racing.
Back at the Wayland EV Juice Bar, everyone did their typical fix and repair thing before the night of racing. We were still perplexed as to why the Zilla in the Zombie was being so unruly with it’s not-so-smooth throttle making for jerky take-offs from rest in mild street driving, and of course its insane power bursts whenever pressing hard on the Go pedal. It was also weird to see those red low Volt set point LEDs lit up on the battery pack modules, even after brief throttle jabs on the street – accompanied by instant twin black marks on the road! The pack was still measuring 4.05V per cell even with the tell-tale LEDs glowing. What the heck?
Once again, in a non-typical mode, we all left for the track on time and we caravaned down to the track. It was now in the mid-70s and it was windows down driving. Upon arrival, we could tell this night was much busier and had to wait in line for nearly 1/2 hour. After tech-in and a recharge to top the pack off, we were ready to see what the night had in store for us. There was a group of bracket racers running 9-11 second cars, so lots of rubber was being laid down and the track was the opposite of Friday night’s horrible conditions, with a hot sticky surface to give good launches…we would soon explore this situation like never before!
Tim and I discussed the technique to hopefully get the car to hook up, which included dropping the tire pressure in back and doing extended burnouts to ensure hot sticky tires. It worked, but just a tad bit too well! The car had crazy wheel stand launches that continued even after leaving the 60ft. area and the nose popped up multiple times as the Zombie went down the track. A longtime drag racing friend, Bob, was on hand to give his always respected analysis of what was going on. He helped us adjust the pre-load of the CalTrac bars, but that made the car hook up even harder! Turning down the battery and motor amps in the Hairball had no effect, and each time Tim returned after a run, he had that look on his face of ‘HOLY sh#%&*tt!!!’ We ran high 10s and an 11 flat, not bad but not as good as in late July when we ran 10.4. The cell low set point LEDs were always lit after the pass, yet the pack seemed unaffected and was at very good cell EQ and SOC. Doc Sherry was pointing out the lowest sag levels per cell that we had never seen before and suggested that we were somehow pulling really BIG currents to make then dip to around 2.5V per cell. Meanwhile poor Mark Farver was still trying to get me to pay attention to his warnings that something was not right in the Zilla’s data that contained no amp readings. I had both Bruce Sherry and Mark Farver with their respective laptops connected to all the Zombie’s electronics pulling data and helping us try to solve the problem at hand…note to self: ‘listen’ to them!
Otmar joined in with us while we were all trying to figure things out, as we decided to simply pull my Zilla and swap it out for the one from Otmar’s 914 Porsche…and that’s when he noticed that some idiot had reversed the B- and M- power cables at the controller’s lug terminals! Remember that bit about how I had removed those annoying white marker dots? From my pt.1 story:
I took the time to clean off the power cables to make them conform to my wiring neatness standards, as someone had marked them with goofy white dots from a white-out pen…note to self: sometimes having cables clearly labeled is a good thing.
This ‘unique’ way of connecting a Zilla resulted in an interesting situation to where the controller still worked and still varied the speed to the motor, but it was running blind as to how much current it was making. Here’s the deal…the B- and M- power studs enter the Zilla and are bolted directly to the two ends of an internal shunt. If you were to measure this chunk of metal with a simply DVM it would be pretty much at 0 ohms, but it is of course, it’s not exactly 0, and from one end to the other it generates a varying amount of mv in proportion to the amps flowing through it to tell the Zilla’s brain what’s going on. With the two power cables reversed, the controller still had a B- input to operate on, and the motor as well, had its B- to operate on. The Zilla could still output the pulsed B+ juice to the motor, but it had no idea as to how much juice it was sending. In fact, as it was sending pulses out, it was probably saying ‘Huh? No amps yet? OK, I’ll send more!”…reminds me of ‘My Big Fat Greek Wedding’…”Mom, he’s a vegetarian” – (Greek mother’s reaction)…”What, he don’t eat meat? – OK, I make you lamb!” Good thing Otmar over-built his creation, because absent feedback from its internal shunt it just kept cranking up the amps to the tune of more than 3000 amps! How do we know this? OK, we really don’t know the exact figure right now, but we do have a pretty good idea based on the cell volt sag we saw in the BMS data, the lowest being 2.44V under the highest loads – thus those 2.8V low set point LEDs lighting up!
Here’s how we have figured things. Note that we have also talked with our good friend and engineer at Dow Kokam on this and he too, is working on the data we provided him to see just how much juice we actually pulled from those ultra high power cells. At the near-400V pack level, the Zilla will limit its max battery input to 1800 amps, and we have very accurate data on cell sag at 1800 amps, that being 3.12V for a paralleled pair or 900 amps per cell. The cells sit at 4.13V or so off of charge, and settle in to 4.05V after pulling them down a bit, then just stay there. The cells sag from 4.05V to 3.12V @ 900 amps. From what we’ve seen all along as we’ve gradually pulled increasing amps from the battery pack (that is until the 9-10 thru 9-11 weekend’s crazy stunt), the cells seem to be very linear and predictable. I saw this trait last Winter when initially testing the cells. We saw it as we started using the pack in the car, and we continue to see it. Subtracting 3.12V from 4.05V, we get a .93V drop, so that is .103V per 100 amps pulled. Now, knowing the cells sagged to 2.44V with the unlimited Zilla doing its thing, we see the cells dipped 1.61V…divide that by .103 and you get 15.63 times, or 1563 amps per cell! In the paralleled pairs then, this was 3126 amps! There are 96 pairs of 3.7V cells to get the pack’s 355V nominal rating, but at the 2.44V per cell low point this was 234.2V @ 3126 amps, or a whopping 732 kW of power for 981 hp! I had written a long time ago as we were creating the pack for the Zombie, that it would make 926 hp and it seemed that some thought that might have been an exaggeration. From my 11-28-09 post entitled ‘Dow/Kokam Powered Zombie…10s in 2010! (pt. 3)’:
I wanted to assemble these cells into modules. I worked closely with Rich Rudman on this with many brainstorming sessions over pie and scribbled-on napkins, and also with him back in Missouri where we ran the concept past the Kokam engineering team. The idea was to keep the design clean, simple, and accessible. It’s the accessible part that dictated a modular design, because having a large assembly of cells all ganged together in the trunk space of the car, makes a package that although small compared to the lead acid pack, is still too heavy and bulky to work on. It also makes it difficult to quickly get to cells if there’s a problem. With a possible TV show in the works (more on this in pt.4) and with Kokam interested in being a supplier of cells for that project, I wanted the modular design so other packs could be configured by adjusting the numbers of and the placement of modules for a given vehicle. Other factors that shaped the module’s design were weight, physical size, shape, current carrying ability, and cell numbers per module. I wanted each module to not be too heavy, so 35-37 lbs. was the goal. Rudman’s newest BMS board is an 8 channel unit, meaning it can keep track of 8 cells (or 8 paralleled groups of cells). At 1.8 lbs. per cell, and needing to have pairs of cells in parallel, a 2P X 8S, 16 cell module made sense. Each module would be made of tough clear Lexan, and the cell’s output tabs would be tied together with high current nickel plated copper buss and clamp bars. With just shy of 29 lbs. of active material (cells) and the heavy 3-4 lbs. of copper interconnects (needed to pass 2.4 kiloamps), hitting that 35-37 lb. goal would be a challenge. At 29.6V, 64 ahr @ C2, and ~36 lbs. per module, and with pack voltage, space constraints, and a pack target weight including cabling and hold-downs of 450-460 lbs., I went with a 12 module, 192 cell design for a 355V nominal, 22.7 kWhr @ C2 power package capable of outputting 2.4 kiloamps for 10 seconds! The very low voltage sag at high currents is very impressive with these particular cells. Graphs provided by Kokam reveal that for every 5C rate of discharge, the cell sags ~.1V, so beginning at 3.8V if one were to extract 150 amps, the cell drops to 3.7V, and at 10C or 300 amps, it goes to 3.6V, so at its continuous rating of 20C or 600 amps, the cell drops and stays at 3.4V…this is very impressive stuff! In theory, at the 10 second rate of 40C -1200 amps, the cell still hangs at 3V! Do the math for our 2P96S pack, and this equates to a staggering 691 kW! It’s amazing, that 345 lbs. of Kokam cells will generate 926 battery hp! This is terrific power density.
Well, as the bad toys out for a night on the town say in that great Kia commercial (music by The Heavy) say, ‘How you like me now?” As is typical from what we’ve seen with Dow Kokam’s cells, they have once again exceeded their specs, in that the Sept. ’09 predicted 926 hp actually turned out to be 981 hp! Note that a year ago I had said “Graphs provided by Kokam reveal that for every 5C rate of discharge, the cell sags ~.1V”…looks like these guys know their cells, as we saw .103V per 100 amps pulled – lining right up to that ~.1V per 100 amps from the Dow Kokam specs! Of course, the Zombie’s motor didn’t make 981 hp, but it’s a good guess it was well over 700 hp at times. No wonder the car went airborne at 80 mph on Friday night and did monster wheel stands on Saturday night! 700+ hp in a 2348 lb. car is for now, too much power. I know, I know, I’ve always subscribed to the notion that ‘too much is just right’, but watching Tim save the car numerous times (and himself) plus seeing much worse ETs with this level confirms it was way too much.
Getting back to the story…once we had the controller swap done, with the Zombie up on blocks in the pits the more typical butter-smooth throttle response was back. Having survived crazy power levels, we were no longer too afraid to crank the Zilla all the way up, which we did with 2000 battery amps and the full 2000 motor amps. As noted earlier, the Zilla would only pull 1800 battery amps due to our pack’s near 400V level. The next trip to the line was entirely different. The Zombie simply stuck, raised up the body level, then shot straight and true to the tune of a 10.258 @ 123.58 mph establishing the best ET ever for a street legal electric car…cheers from the bleachers and lots of sighs of relief!
Back in the pit, Tim said the car was boring now Not a single red LED could be found, and the pack had stayed right near 300V for the run… if my memory serves me correctly, Doc Sherry figured it went to 305V @ 1800 amps…incredible cells! The car didn’t spin much, and simply took care of business, knocking down run after run in the 10.2-10.3 range, all at 122-123 mph. With it’s 110-120 mile range per charge energy density, the last two runs on of 10.261 @ 123.54 mph and 10.287 @ 123.38 mph were done without prior recharges, then the car was driven a bit hard leaving the track, then driven 16 miles uphill back home. The next morning the 3.7V nominal cells were at 3.87V, and not a single red LED could be found!
It was a weekend we’ll all remember for a long time, and though we didn’t make it to the 9s, none of us are too upset about only running 10.2 with a street legal electric car! We had gone to the dark side of electric power, but lived to tell about it. That being said and strange as it seems right now, Otmar and I have been scheming together again just as it was all those years ago during the skunkworks days. With some new ideas about Zilla current control mods, and with more suspension work to the car, we’ll return to the dark side in 2011 as we head into the single digit ETs! A note on safety here…my accidental mod to my Zilla is not something anyone should purposefully do in terms of safety and in terms of possible damage to an expensive Zilla Z2K. The controller was not intended to run this way, and I in no way am recommending it. It does however underscore the high degree of ruggedness Otmar designed into this mighty controller!
Thanks go out to our sponsors, the Plasma Boy Team, and all our friends who were there to cheer us on.
See Ya…John ‘Plasma Boy’ Wayland