Cooling Modifications Experiment for the Track: more air flow through radiators to keep battery temperature down

[bsclywilly]

For anyone hitting thermal limits on the track, I did some experiments at a second outing at Laguna Seca with my CT4.

I was interested to see whether improved air flow through the radiators could help extract more heat and keep battery temperatures down, allowing us more laps before reaching the 132F thermal limit and the car cutting power. The summary is that I was able to test 1 out of the 3 modifications I had planned and the gurney flap added to the leading edge of the wheel well resulted in a 20% decrease in battery temp ramp rate. That equates to about 2 extra hot laps around Laguna Seca.

The premise is that by reducing the front wheel well air pressure, we get more airflow and heat transfer through the radiators which vent into the wheel wells. The wheel wells see higher pressures since the wheel and tire creates turbulence, so the rad exit location is not ideal but gives some opportunities to improve cooling. Planned tests consisted of the following:
1. Extended wheel deflectors. These reduce the pressure in front of the front tires by directing air around the tire. You instead see higher pressure in front of the deflector and the pressure behind, in the wheel well is reduced. Note the pressure plots below with and without deflectors. The wheel house pressure further reduces the longer the deflector is, at a small detriment to overall drag. See attached for image reference, a good read!

I had made 1" extensions from the existing wheel deflectors but they didn't last 1 lap since I was already bottoming out on the stock deflectors. I would revisit this again with flexible rubber deflectors.

2. Gurney flap to direct air outwards from the wheel well. This feature has been well documented in SAE papers to help extract hot air from the wheel well for brake cooling effect and reduce lift. Penalty again is an increase in drag as the side wake is increased. I made a test prototype out of 1/16" thick PC that extended out 3/4".

3. Taping up the air curtain ports on the sides of the front bumper and also partially closing off the brake ducts were the other planned modifications that would reduce airflow from the front of the car entering the wheel well. I was testing a new set of track pads but on my initial session, I was worried that overheating the brake fluid would be a potential risk so decided against the brake duct blockers. The air curtains, I just forgot to tape up, doh! But you can see the Taycan GT spy pics had the same idea.

So here's the results that the gurney flap made to cooling. Measurements were logged using the CarScanner app. Both with and without the flap were run in the same weather conditions, partly cloudy 59F in 15 minute sessions. Raw temperature data presented below. With flap on left, without on right.

In and out laps are more variable so I truncated the data to include only the hot laps, 7 laps each session:

x-axis in seconds.

The yellow line represents battery temp and its immediately obvious that the slope, or rate of increase is higher without the flap. In fact, without the flap, the starting battery temp is lower but ends with a higher temp than with the flap.

Battery temp w/ flap: 87->115F​

Battery temp w/o flap: 85->121F​

To make sure we have an equal comparison, I confirmed that the cooling pump was running equally for both sessions, both got up to 90% by the start of the second hot lap. I also wanted to normalize the battery temp ramp according to the energy used. The session without the flap used more energy at 2% higher rate. I added that 2% to the battery temp ramp rate of the no-flap trial. The normalized difference: 20%.

Battery temp increase w/flap: 2.40 degF/minute​

Battery temp increase w/o flap: 2.99 degF/minute​

​

What does that ramp rate mean? I can extrapolate that to see how much longer (in minutes or laps) it would take for me to hit 132F battery temp, at which the car cuts power. With the gurney flap, I'd be good for another 4 laps, but without, I'd be done at another 1.8 laps. That's a difference of about 3.6 minutes. With the gurney flap I'd be looking at 11 full hot laps in these conditions, a bit more than 23 minutes lapping at full speed with a proper warm up and cool down lap. Perhaps even longer in reality as there's another 10% capacity in the cooling pump that hadn't kicked in yet.

This is good to know because next spring, with higher ambient temps, I'd like to get those other cooling mods ready so that battery temps will not be a limiting factor. If anyone else is interested in testing some of these things out, let me know. I'm particularly interested to see how well they work with a turbo or turbo S, as @daveo4EV found his limit was only about 6 laps, and that's no fun. We need more Taycans on the track!

Sponsored

 

[RAHRCR]

Excellent write up. Does the rest of the car hold up given the additional battery temp headroom (tires, brakes, suspension)? Curious about the next weakest link if you can extend the track time of the battery. Thx

 

[Vercingetorix]

Test with gurney starts at 100%battery charge and without starts at 50%? If so that alone may explain any difference.

 

[daveo4EV]

this is super interesting data - it was super fun chasing you on track

what was your tired setup again?

 

[daveo4EV]

temp rate rise for the turbo would also be interesting - as the difference between the 4 and Turbo is mostly max KW fed to the EV motors which means higher discharge rate from the battery which means more heat generated in the battery…so the issue is can the cooling system keep up for full power in sports plus mode

perhaps worth testing at a private track day at Thunderhill sometime before spring since Laguna is close for remodeling until May 2023

 

[bsclywilly]

Excellent write up. Does the rest of the car hold up given the additional battery temp headroom (tires, brakes, suspension)? Curious about the next weakest link if you can extend the track time of the battery. Thx

I had tires, brakes and brake fluid upgrades. It was great lapping 20min sessions at full pace! Tires would probably be the next thing, for such a heavy car and extended sessions >20min you still have to be careful of tire temps. And driver fatigue. Also, if you don’t have fast chargers nearby, you’re also limited to a battery budget before you have to leave to get a charge. Currently, running two 20 minute sessions is not a concern you use up about 50% each. Longer sessions and you’ll need a top up after the first session.

Test with gurney starts at 100%battery charge and without starts at 50%? If so that alone may explain any difference.

Please explain why!

what was your tired setup again?

Bridgestone RE71RS 285/35-20 all around.

 

[Vercingetorix]

Please explain why!

This is all speculation, but to keep an equal amount of power to the motors requires higher amperage as the battery voltage drops. The battery voltage drops no matter what, it’s what they do to provide energy. Higher amperage means more heat. Also notice the cooling pump seems to not be at 100%, indicating there was still cooling capacity remaining. Another variable would be ambient temperature. Was it warmer later?

 

[Neptune]

Apparantly turning off regen completely stops the battery from overheating on the track.

 

[bsclywilly]

This is all speculation, but to keep an equal amount of power to the motors requires higher amperage as the battery voltage drops. The battery voltage drops no matter what, it’s what they do to provide energy. Higher amperage means more heat. Also notice the cooling pump seems to not be at 100%, indicating there was still cooling capacity remaining. Another variable would be ambient temperature. Was it warmer later?

This makes sense and some quick research supports this where an increase in cell heat gen increases with DoD during discharge. Based on the linearity of the battery temperature rise for both tests, I suspect that effect is not great, otherwise we would see a slightly progressive curve given that other cooling variables are essentially equal. For the data I collected, I just further truncated it so comparing a closer range of 75-55% SoC (w/gurney) to 53-34% SoC (w/o gurney) and the difference actually further supports the effectiveness of the gurney. I would say this phenomenon is likely in the noise. But future testing, I'll be sure to reverse the order. Here's a reference to put this concern in context:

This is for a LiFePo pouch battery (fig.5) with passive cooling so I'm sure there are some chemistry differences. But for the sake of this argument, We're discharging the Taycan battery at a rate of about 1.6C. The heat generation in the graph for 1C and 2C discharge rates are still relatively flat and constant. So I think this further supports it's not a huge factor. Perhaps the slightly higher discharge rates of a Turbo would see a greater demand on cooling at lower SoC, though.

I noted in the original post that ambient temps were identical. The cloud cover also would have maintained consistent track temps. The runs were made about 1.5 h apart , both in the afternoon. Also noted the cooling pump has more capacity so the actual run time may be more than I extrapolated.

 

[daveo4EV]

Apparantly turning off regen completely stops the battery from overheating on the track.

I know of no way to 'turn off' regen.'

how is that accomplished?

 

[grahamsimmonds]

Interesting. I track my Turbo S regularly here in the UK at Goodwood race circuit. Some thoughts to throw into the mix:

1. Letting the car cool down. If you start a session at 45 deg C, it does not take very long to reach 55 deg C (the Celsius limit). I have found the quickest way to cool the battery is to charge it. This will get it down to 30 deg C in about 15 minutes. We are fortunate to have 50 kW chargers opposite the paddock so this is easy for me.

2. Tyre pressures - it is essential to check your pressures constantly. This substantially reduces tyre wear. I check mine after the first session and let them down to the standard pressures. Usually about 4 psi. I then check them after every session and don't let them above 40 psi.

3. Ceramics - They are amazing. Obviously the 265 kW of regen does most of the work around the circuit but we have a 155 - 70 mph braking zone where they come in useful! They never overheat and are still 7mm all round after 2 years (20+ track days).

4. Garmin Catalyst - what a piece of kit! As you go around the track, it offers advice about turn in and apexes, braking etc by looking at how you drive a lap. At the end of a session it constructs an optimal lap which is made up of the "best bits" of your session. Usually about a second quicker than your best session time.

5. Driving in the wet - simply one of the best cars to drive in rain around a track (this is the UK after all). You can stick your Caterhams, Aerial Atoms or any other RWD car. The Taycan pees all over them.

6. Tyres - because I have 21s, tyre choice is very limited. Slicks/Semi slicks are not an option. The best tyres I have found are the Michelin Pilot Sport 4S. Loads of grip and last for ages.

 

[bsclywilly]

1. Letting the car cool down. If you start a session at 45 deg C, it does not take very long to reach 55 deg C (the Celsius limit). I have found the quickest way to cool the battery is to charge it. This will get it down to 30 deg C in about 15 minutes. We are fortunate to have 50 kW chargers opposite the paddock so this is easy for me.

Have you experienced hitting that 55C limit and have power cut like daveo4ev has in his Turbo? How long are your sessions and how many in a day? May I ask what typical lap times around Silverstone you get?

I've also heard keeping it in Sports Plus mode keeps the cooling system at full power which during charging helps cool even further. Probably most beneficial if you have trackside charging and can get it after each session. Our sessions are usually spaced about an hour or more apart so plenty of time to cool down in between.

 

[grahamsimmonds]

Have you experienced hitting that 55C limit and have power cut like daveo4ev has in his Turbo? How long are your sessions and how many in a day? May I ask what typical lap times around Silverstone you get?

I've also heard keeping it in Sports Plus mode keeps the cooling system at full power which during charging helps cool even further. Probably most beneficial if you have trackside charging and can get it after each session. Our sessions are usually spaced about an hour or more apart so plenty of time to cool down in between.

Yes I have reached the temperature limit on many occasions. I have learnt to manage it by starting the session at 30 deg C. The problem with it cooling itin Sports plus is it uses battery. Cooling it by charging obviously gets around this. I will also do cool down laps.

 

[rs38]

great stuff!
were you really able to keep the same lap times during the 2nd stint with battery down to ~15% SoC?
this would help to compare them on the same x-axis to see the essence of all the data in one diagram comparing only few essential values (batt temp and maybe power or amp hours) over time (or even better: over distance).

I do most of my car scanner data analysis with Python to automate export workflows and align data.

Did you log with SportChrono active as well?

The problem with it cooling it in Sports plus is it uses battery. Cooling it by charging obviously gets around this.

cooling by charging eats the same amount of battery that is missing in your battery but you pay for

 

[WasserGKuehlt]

This is all speculation, but to keep an equal amount of power to the motors requires higher amperage as the battery voltage drops. The battery voltage drops no matter what, it’s what they do to provide energy.

Sorry, help me understand. The battery would not be one contiguous 'element' that is depleted gradually and uniformly - I'd expect it is a set of "modules" comprised of serialized pouches, and which are selectively used in parallel as needed. That is, there will always be a module delivering 800V for as long as "the battery's" SoC > min. Is that not the case, or am I missing completely how this works? Thanks.

(edited for syntax and clarity)

Sponsored