Headway Pack Performance Analysis

ahambone

10 W
Joined
Oct 6, 2008
Messages
71
Time to share some Headway data porn. The 120 cell Headway pack has been in "testing" use in the car for a few months now. I've determined that the 50-AHr (5p)x24s pack gives me just about 15.2 miles of range, taking the cells down to 64V total (~2.6V/cell) where my controller cuts me off. I've never tripped the BMS LVC on the road (fortunately). The 15.2 mile range isn't far from the 16.x miles I expected based on various EV calculators. It's almost a really usable range, which makes me happy.

Over the last couple weeks I have finally connected up all the feeds for my Cycle Analyst which allows me to collect data from my drives. Hopefully sharing the data with the community will spread more insight and discussion about the cells and what they are capable of. The data here represents about a 15 minute drive around the neighborhood. Like most neighborhood drives it starts with backing out of the driveway, accelerating down the street, and then stopping at a stop sign. This accelerate/stop at a sign pattern repeats several times throughout the drive. Should be very much like data from a neighborhood drive because, well, it is excatly that. :wink:

Most of the streets around my house are 35mph, with one 40 and one 45 mph street. The CA's speedometer is very accurate with respect to the car's speedometer which leads me to think I got the algebra right when computing the circumference of the car's tires. There are some flaky spots where the CA's speed signal flaked because I have a sub-optimal connector attaching it to the speed sensor. I had to extend the speed sensor wire on the CA to route it through the firewall. The magnet for the sensor is attached to passenger drive shaft of the front wheel drive and the stock CA cable didn't quite reach.

Fun stats about the drive:

Amp Hours Consumed: ~20 AHr
Peak Amp Draw: ~360 Amps
Min Voltage Valley: ~68 Volts
Peak Speed: ~ 43 MPH.

Anyway, on to the artwork. I've graphed the entire drive in GNUPlot. Here's the graph for the entire drive.

View attachment fullDriveGraph.png

The only thing significant here is that the pack is about 40% depleted at the end of the drive and the voltage still bounces back to over 78 volts at the end of the drive. I love how the pack "feels fresh" for the entire useful life of the drive.

I've "zoomed in" and created graphs for both the first real acceleration and the biggest acceleration (which happens at about sample 4000 in the graph). Here's the fresh-out-of-the-driveway acceleration.

View attachment 2

The pack's voltage really drops off when I start pulling between 3 and 6 C out of the pack (300 A = 6C on this pack). I try to keep the over 200A draws to short bursts and the 300A+ draws to really short bursts. For most of the test drive I was showing off by driving not-conservatively in order to show off how the pack performs.

Here's the data from the acceleration where I let the car rip on a 45 mph road.

View attachment accelerationGraph.png

The vehicle is literally floored during the upward arc of the blue speed curve. When the CA got over 350A I tapered off, and was basically already up to speed with traffic. That's almost 400A - which is more load than I think should be routinely placed on a 5p setup of 38120S cells, but it is nice to see they can handle it.

Next week I have another block of 72 cells arriving that will get integrated to make the pack an (8p) x 24s setup. I'll run another set of data with 60% more cells and we will see if the voltage dips are as severe (of course they won't be - but the fun is in collecting the data!). :mrgreen:

I've attached the original data files in case anyone wants to steal the GNUPlot commands used here for their own purposes.

Cheers,
--Adam
 

Attachments

  • ev-data-logs.zip
    74.3 KB · Views: 20
Those graphs are totally cool 8)

Have never seen voltage, amperage, and vehicle speed so nicely superimposed in that way. Such a nice "big picture" seeing it like that :mrgreen: .

Based on what I have seen with my S version Headways (mere 8s :oops: ) using my rudimentary hobby tools, it intuitively seems to me like your P versions really hold up very well under load (recovery too)... but that's just a novice opinion based on what I think I can see in the graphs :roll:
 
scoot said:
your P versions really hold up very well under load (recovery too)...

If it's 5p and 50Ah, they are not "P" versions.

The P versions would have half the voltage drop. But 5p of them would only make 40Ah.
 
liveforphysics said:
scoot said:
your P versions really hold up very well under load (recovery too)...

If it's 5p and 50Ah, they are not "P" versions.

The P versions would have half the voltage drop. But 5p of them would only make 40Ah.

Yes Luke... :oops: Thank you for the correction. I missed that obvious detail :roll: . I seem to be having problems lately with confusing members and their various builds etc. I hate getting old :cry:

With the p versions having half the sag as you say, I am very tempted to seek some out for the next time. Hill climbing is becoming a priority for me lately :wink:
 
MrKang - thanks for the link to the program. That certainly will make things easier!

Cheers,
--Adam
 
Back
Top