DIY Electric Car Forums banner
23K views 49 replies 8 participants last post by  danh818 
#1 · (Edited)
I'm a bit late getting this started, but here goes.
I've always liked the body on the Porsche 944 so when I found one about six months ago on ebay only 30 miles away from me I jumped on it. I intend for this car to be driven daily with a 80 or so mile range, but with the ability to give so great stoplight to stoplight performance.
To date I am about 40% complete converting my 1987 Porsche 944. Here is a quick rundown of most of the hardware going into the car:

NetGain Transwarp 11HV
Gear Vendors Overdrive (no transmission)
WarP-Drive Industrial 360v 1400A Controller
384 A123 20ah cells
Full Roll Cage
Currie custom 9" IRS Housing with High Pinion 3rd Member 4.86 gears and and LSD

I have posted more info and tons of pictures on the blog at http://infa944.blogspot.com

I'm looking forward to all your feedback.

I would like to thank everyone who has helped out up to this point for your advice, encouragement, and expertise:

The No Agenda Show (In The Morning Slaves) Stephen Johnsen (HPEV in Washington), John Wayland (Plasma boy Racing), John Metric (DC Plasma), Aaron Sacks (Electrical Engineer a.k.a. Homer Simpson), George Hamstra (NetGain Motors), Jack Rickard (EVTV), RWaudio (electricporsche.ca), David Katz (for encouraging me to pull the trigger on buying the donor), Stuart Horwitz (my dad and hero), Bill Cloutier (BC Elecric - Master Electrician) Jonathan Stilz (Suspension Guru), Larry "Bear" Watson (Master Welder), Roger Derryberry (Miata Autocross Conversion), Richard Rodriguez (Porsche 914 Conversion) Onegreenev (blog advice) Northridge Tire Pros (Tires) Wenco (Drive Shaft), Sway Away (Custom axles), a BIG Shout Out to Bowser330! Oh and don't let me forget all the OEM's (without all your crap cars I would not have been motivated to do this for myself.
 
See less See more
1
#2 ·
but with the ability to give so great stoplight to stoplight performance.
:eek::eek:!!! With this kind of conversion, you will reach the city speed limit under a second...:D

Please continue to share development. The final result will be impressive!
 
#4 · (Edited)
Nice to see a thread and blog for this build, great start! Your car is going to be a silent rocket!

I just got my test pack of A123's in for some real world driving. Took my wife for her first drive in it tonight (she drives an '83 944 with the engine out of my '86) she was impressed to say the least, quiet and so much more power off the line it puts her car to shame. Full throttle with a 2nd gear start ~7:1 overall ratio and it's quick on 1000A with 1400A you would do well with taller ratio's. I haven't tried 1st but I'm sure I could light up the tires ~11.8:1 ratio.

Is that a 2 speed? What will your overall ratio(s) be? The LSD should help a lot!

Since you don't have your cells yet, I assume you haven't settled on a connection method, I would strongly urge you to have a look at some testing that Andre Ferron did with his A123's. I gave him some ideas based on my experience with the cells from the results of my testing and he did all the dirty work testing different methods.



He was going for the lowest mv drop in a series connection for a 3P module, the left meter is measuring the left connection which is made from flattened copper tubing and some inexpensive hardware. The right meter is measuring an earlier attempt which is made from an aluminium tube and round bar of suitable diameters to form a half moon clamp but without the tabs overlapping.

This test was at 618A as you can see on the clamp meter.
The secret to it's success is alternating the tab layers, so in the 3P grouping the sandwich is formed from the following layers:

copper bar
group1 cell 1 +
group2 cell 1 -
group1 cell 2 +
group2 cell 2 -
group1 cell 3 +
group2 cell 3 -
copper bar

By alternating the layers with the most practical contact area and not using any separator blocks between the series connections you get the lowest possible connection resistance for a pack of this style.

I assume your pack will be 96S4P so you could expect LESS than a 1.5v drop at 618A due to connection resistance from your entire pack (excluding cell sag of course). Or less than 3.4v at 1400A.... this compares to almost 20v using the same cells with the slightly different but good technique shown on the right.

Good luck with your build!
 
#5 · (Edited)
Thanks for jumping in RW, I have been admiring your build from afar for awhile now. It's really great to hear that your car is on the road.

I did just receive conformation that our cells have cleared customs and will be delivered in a few days. I've got two PL6's that I plan to use for capacity testing, and following your lead I will be grouping cells accordingly. You are right to assume that I will be going with a 4p96s design, although from the get-go I have been planning on slightly undercharging, so its a possibility that I may add one or two more to each parallel string.

I have not seen this "half moon" connection method before, it certainly gives me some new ideas to ponder. Off the top of my head though I think I would machine flats on the top for better terminal contact. This method of alternating polarity from cells of different groups warrants much thought as well, as it is clearly superior in terms of voltage drop. A large concern of mine, like Jack Rickard of EVTV has stated, is movement of the tabs independent of the cell pouch. This will eventually lead to fatigue, connection failure, possible cell damage and/or fire. I have been thinking long and hard bout this and will finally have the ability to experiment when my cells arrive.

I am curious to know how your charger is working out. I have also chosen to go with the DIY 10k, but I have ordered mine built from EMW. Valery emailed me Tuesday to let me know that the build was complete and is now being tested.
 
#6 ·
Since I'm not running a traditional transmission my first gear is 1:1 with the motor. The Gear Vendors overdrive unit is a .78 so with a 4.86 rear end that gives me a 1st gear of 4.86 and a 2nd will be 3.79. I talked to George Hamstra of NetGain about this, and we both agreed this would be a good starting point, the rear end gearing is by no means set in stone.
 
#10 ·
Actually Joey it is my first extensive car build. I have a good amount of general technical knowledge and and experience (I own a small contract manufacturing and prototype company), but as far as cars go, my experience stops with bolt on parts for my ICE cars. I have done quite a few high wattage stereo and subwoofer installs in my cars over the years but that is child's play compared to this. The only other experience that might help me out is that I have flown R/C airplanes since I was a kid and have used LiPo's for the last 3 or so years. That also led me to dabble in building some submersible HID flashlights used for scuba diving, but it both cases nothing over 12 volt nominal and a couple amp hours.
 
#14 ·
I'm actually not sure. We gave Advan the original parts to make molds from since they have never done anything carbon for the 944 before. I should have weighed them before sending off the originals. I will find a way to weight the new carbon parts, but I'm not sure if we are actually going to get the originals back. Although it's my car one of my good friends had a contact at Advan and worked out the deal. The parts were given to us gratis as they are prototypes and in return we agreed to do some promotional photo shoots when this car is finished.


RWaudio, what are the specs of your traction pack for the drive with your wife? Amp draw?

Now that I have had a chance to physically get my hands on the A123's I'm fairly impressed with the robustness of the tabs. I'll be spending the weekend thinking and tinkering intently on my cell connection and packaging method.
 
#15 ·
I'm using a 44S3P pack right now which is a bit low on voltage for the 11HV.
Battery and Motor amps set at 1000, clamp on meter on different runs showed 949 motor amps and 809 battery amps. I did notice from the log that I wasn't getting full throttle so I'll have to recalibrate that.

Torque is dropping off by 2500rpm but that seems normal since I'm running 1/2 the rated motor voltage.
 
#16 · (Edited)
Thank you for sharing your specs RW. This is promising news for both of us, and anyone else using an 11HV I think that in the end we will both have plenty of power at our disposal.

I have ordered some toggle clamps and nylon thumb screws from http://mcmastercarr.com to build out a cell testing board. I've also ordered another two PL6's, so I will now have four total. This should speed up capacity testing considerably. I should have the 'board' setup in a couple days.

I have now firmly decided that I will only be charging my cells to 3.6v so I will be doing a 4p98s design. I like what Kerry Mann has done with his battery design, and will be doing something very similar, but for me there is a bit too much exposed conductive material. I'm fortunate to have a laser cutter at work and we do a lot of work with Acetal POM sheet. I'm looking into using POM or a smiler engineering plastic as an external insulator for my battery groups. I am also thinking about potting the the connections to help reduce flex. While doing research last night I stumbled upon http://www.bricolagefaidate.com/english/magicjoint.php
Looks promising and warrants more research. Interestingly this product is "re-enterable" so it can be torn apart pretty easily and a particular cell could be accessed with relative ease. Seems as thought some other products from this company may be applicable to EV'ers as well, such as their self regulating heating cables.
 
#19 ·
Not quite as much progress today as I had hoped for, but if its worth doing, its worth doing right. No cell testing today, but I did grab some 1/0 stranded cable and nip four, 6 volt batteries from our Sports Rider golf cart. By the end of the day I had the lead acids wired up in a 2p2s configuration to feed my four PowerLab6 chargers, but alas, I needed a 12v load to draw them down as there were already topped off. My go to guy David reminded me that we could use the Transwarp 11HV and kill two birds with one stone, breaking in the motor and using it to draw down the golf cart batteries. Cell testing to commence tomorrow morning.
 
#22 · (Edited)
I've made some changes to my cell testing parameters, so I thought I would post the whole procedure from beginning to end.

After a couple of trial runs, I found that my lead acid supply bank just wouldn't take a 4x 40amp regenerative discharge, so I had to drop down to 35amps.

To start off the process, I number each cell with a paint marker as well as its corresponding plastic tray. The sticker from the negative tab of two cells is then removed by hand and I place those two cells side by side with the tabs resting on a 3/32" thick piece of Acetron GP (aka Delrin). I then use a shop rag with a bit of WD40 to remove the leftover adhesive from the sticker. Next a different rag with rubbing alcohol is used to remove the oil and ensure the contacts are spotless.

My cell testing station is setup with four 6volt deep cycle golf cart batteries arranged in a 2p2s group using larger copper ring terminals and 1/0 copper stranded cable. All four PowerLab6 chargers are daisy chained using servo extensions and 'y' harnesses with a single USB interface connecting them to a PC for control and logging. I have wired the PL6's in a 2x 2p arrangement by removing the factory bullet connectors and using copper crimps to to attach the input leads to some 4awg stranded copper cable I had around the shop. I have chosen to terminate the positive end of the chargers with a 'jumper cable' clamp, so that I can easily disconnect them when not in use, this also lets me power up each set independently if necessary.

Cycling parameters are as follows:
Charge - 3.6v 38amps C.C.C.V C/20 I.R. is marked on each cell at the end of the charge cycle
Regenerative Discharge - 2.5v 35amps C.C. Amps out are marked on each cell

I run only one cycle beginning with charge and ending in discharge. Cells will be left at discharge voltage until I have tested them all this way, then grouped by capacity and bottom balanced using C.C.C.V discharge to 2.5v before assembly into modules.

The purpose for this procedure is simply to weed out weak/damages cells and begin to break the cells in without fully taxing them.
 
#24 ·
They really do not have any issues with self discharge, although in my opinion long term storage should be somewhere around half state of charge. Also, since I am drawing the cells down to 2.5v with a constant current output, they bounce back up to between 2.9-3v almost instantly and after a few minutes, some are reading over 3.1v
 
#26 · (Edited)
Feel free to stop by tonight Bowser. I'll be here late testing. You can have a look at the data for yourself, it's always good to have a second set of eyes. So far I've got 44 cells done, I'm shooting for at least 50 by tonight.

If any other forum members that are in the LA area want to come by and check out the build progress feel free to send me a PM .
 
#27 ·
Not much to report in the last week. The car is out at my fabricators shop getting the rear end installed. They will also be welding in some gusset plates into the front end to strengthen the strut towers and act as a attachment point for the front most motor mount.

I have now tested 102 cells, all with very good results, even the few so far that have slightly wrinkled tabs, or a bit of curvature to the pouch. Even with my very conservative testing settings, the lowest capacity I have come across has been 18.3aH.
 
#28 ·
New Blog post is up!
Sorry to all you that have been waiting with baited breathe.

My laptop decided it had taken enough abuse and gave up on life. After a couple trips to the Apple store they decided to replace the hard drive cable, but alas that wasn't quite it. Since I had replaced the hard drive with a non-apple one, I was on my own at that point. Although I back up all my data, starting from scratch and transferring files was and absolute last resort. I used Spin Rite, a hard drive recovery tool that I have had great success with in the past, and after eight straight days of running, my machine was again bootable, not having lost a single file. I immediately cloned the drive installed a new, solid-state hard drive and transferred all the data. Finally back up and running.

Throughout the ordeal we became swamped at work and I was left with zero free time and no computer to dump all the new fabrication photos onto. But all the while Empire Fabrication of Simi Valley Ca. was hard at work bending and welding, getting us a few steps closer to a finished car.
Without further ado:


Check out my blog for more pictures of the whole fabrication and mounting process.

Rear end mounting is complete! - Now for the motor...
 
#29 ·
Awesome progress, the rear end looks sweet!! It's odd not seeing the transmission there. I assume you are going to put some of the batteries back there similar to what I did? What kind of weight balance are you looking for? Keep the stock 50/50? Mine has turned out a bit rear heavy and although that works very well for traction I'm not sure how it will affect the handling in the end. I'm going to move the charger, dc/dc converter and 12v battery up in front of the motor as low as possible to redistribute some weight.
 
#30 ·
Trial by fire at the track

It's been almost 2 years since my last update to the forum here, although I had made some updates to my blog about a year ago. Sorry to everyone if you were wondering what happened with the car.

In short - complete success!!

The car is up and running with a 4p88s pack of a123's. I literally got the wheels turning July 17. With less than 1 mile on the odometer and no charging system hooked up, loaded it on the trailer and towed it 300 miles from LA to Mazda Raceway Laguna Seca in Monterey California for Refuel 2014. Got the J1772 wired to my charger on the 19th and had a completely trouble free day yesterday July 20, running the car on the track.

Although the Model S is certainly a faster car on the first lap once they became power limited due to heat build up, we were able to give them a run for their money. I know the 944 will out brake the S any day of the week and we are better in the corners.

Anyway just a quick brain dump for you guys until I have a chance to get some rest, clean the horrendous mess I left in the shop before leaving, get the car off the trailer and back inside, gather my thoughts and post some pictures and get you guys up to speed.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top