ICE TANK TESTING RESULTS:

I finished my ice tank project and took it to the drag strip for some testing. It definitely works! The ice tank dropped the intake air temperature 50* F!

The tank holds two “large” bags of ice (22 lb) plus water. I made two passes with just water in the tank (full, 7-8 gallons) and then drained most of the water, added both bags of ice, and made two more passes. There were a lot of cars there, so I only got four passes all day. After two passes and about 1.5 hours of downtime, about half of the ice remained. So I’m thinking 2-4 bags of ice for a day of racing.

The picture below is a typical ¼ mile log for me before the ice tank. The IAT (top graph, purple) was 107* F before the run, and climbed to 132* F during the run. It was 67* F outside. This is with stock pulleys, Johnson CM30 pump, and the cooling systems separated. I’ve seen as high as 150* F IATs on hotter days. (!!)

5-9-121-HX.jpg

The log below is with the tank with water only. Same mods as above plus a 2nd OEM heat exchanger (Thanks 32crazy!) in series. The IAT was 95* F before the run, and rose to 125* F during the run. It was 61* F outside.

Noicerun2.jpg

This log is with ice added to the tank. The IC had actually heat soaked more and was 113* F before the run. When the IC pump kicked on it dropped the IAT to 48* F! The IAT dropped 65 degrees in 24 seconds. The IAT rose from 48* F to 75* F during the run. That is a solid 50* F reduction, and possibly quite a bit more on hotter days.

Withicerun4.jpg

Here are pictures of the tank. It fits on the parcel shelf and under the tarp for the roof (SLK32 AMG). This part of the trunk is hard to access with the roof down anyway, so it is a very convenient spot. I also wanted to retain the deeper portion of the trunk for cargo.

IMG_1091.jpg

IMG_1095.jpg

I MIG welded the tank from aluminum “5-bar” plate. It is like diamond plate, but looks a little nicer in my book. I put a giant clear lid on it with a 10” opening. There are two ¾” NPT fittings on the lower right side for supply and drain lines, one ¾” fitting on the upper left for the return, and mounting tabs on both sides.

The welds are FAR from perfect and beautiful – I’m a hobbyist, not a pro with a TIG budget – but I did pressure test the entire tank and it IS air tight.

I wanted to try this new PEX tubing they’re building everything out of, so I made the lines out of ¾” PEX instead of braided hose. I actually like it a lot, and it is EASY.

I thought that the warmer water exiting the intercooler might still be colder than the outside air temperature and that running it through the heat exchanger(s) would only warm it up and melt more ice. So I also installed a 3-way valve that can bypass the heat exchangers and route the return from the IC directly back to the ice tank. I do not have any intercooler water discharge temperature readings to prove if this is helpful or not yet.

I was considering installing a 2nd Johnson CM30 pump to maintain flow through the extra length of tubing, but after observing the flow rate back into the tank I don’t think that’s necessary.


I have enough material to build two more tanks. So I am going to go ahead and weld them up sometime in the next few weekends and post them for sale. I have to add up my material cost to figure out the price, but I’ll give you guys first dibs.

 

GREAT WORK!
Love the fact that's it's an SLK32 as well......
With your runs - is the water/meth off or is it running in combo?
Is the box insulated, I'm guessing so.
So - looking at 10kgs of ice to water (335kJ/kg) say over 120seconds of "need" = approx 30kW or 100,000BTU/hr....

Coolant temp into the charge air cooler would be good - then you could just switch your 3 way from IC to no IC.....and [the temp diff between the 2] x [flow rate] x [cp] being your ambient ingress to the IC.

Going by your AIT in graph 3 it going beow ambient - so no brainer....

 

Thanks! I don't have meth/water injection. So there is no cooling contribution there.

I have not insulated the ice tank, but I may. It would be more to prevent condensation than to keep the ice from melting. I didn't get any meaningful condensation on the tank, but the weather was mild. We'll see next summer. It should be easy to insulate.

 

Those are INSANE (!!!!!!) IAT's... very well done sir.

 

Very impressive. Racing isn't my cup of tea.. but do like to follow along threads like these. Only one thought here on it.. Will the IC hold up without cracking it's welds with those rapid temperature swings over a length of time?

 

Could you post the end results of the lowered IATs? The actual comparisons of the runs of stock vs water-only vs ice? The 1/4 time is not as important in this case as there are a lot of other factors that come into play.. I'd be more interested in the trap speeds. Thanks.

 

Very good point, I wasn't even thinking about this but in fact this is quite a drastic drop in temps that the IC would not experience in any other conditions and possibly might not be designed to withstand such (especially the sharp drops in temperature versus the sharp increase which it should be designed for) I was actually thinking that this was partially the reason that the ICs do fail while reading the 'broken IC' thread. I don't get on it whatsoever when I drive for at least 10 minutes after the car has fully warmed up per temp gauge.

 

Great questions.

I had to have a look.
Below are the same three runs, but rescaled and with all graphs removed except IAT.

The first one is back when I just had the one OEM heat exchanger, plus a Johnson pump and separated cooling systems.
The IAT increased at a rate of +1.90 deg/sec during the run, and decreased at a rate of -1.29 deg/sec while decelerating (1/4 mile racing, and 1/4 mile braking).

5-9-12TT1notank1HXjpg.jpg

The second graph below is with two OEM heat exchangers in series and the tank, but no ice.
The IAT increased at a rate of +2.48 deg/sec during the run, and decreased at a rate of -1.97 deg/sec while decelerating.
Those cooling rates alone indicate a 53% increase in cooling by adding the 2nd stock heat exchanger. The cooling rate would probably be even higher with a large supercooler HX.

10-21-12TT2tankwateronly2HX.jpg

The third graph is with ice in the tank and the heat exchangers bypassed.
The IAT decreased at a rate of -2.73 deg/sec while cooling down the initial heatsoak.
The IAT increased at a rate of +2.45 deg/sec during the run, which is nearly identical to the rate with no ice.
The IAT decreased at a rate of -1.58 deg/sec while decelerating, which is actually slower than the previous run with no ice, but very close.

10-21-12TT4icetank.jpg

So the rate of temperature change doesn't change much with the ice tank. But the temperature can be cooled MUCH lower.

 

Sure can. There was no improvement. I only got 2 runs with and 2 runs without the ice. Neither of the runs with the ice were very good. I spun the tires on both launches and had slower 60' times. I was hoping to get some more test runs last Sunday, but they cancelled the test & tune due to rain. That was my last shot until spring So I don't have a good feel yet for how much faster it'll run with the ice.


You'll notice the runs below are very inconsistent. I am not happy about that. One thing I am noticing is that the timing advance varies considerably and might be a big factor. On the faster 13.04 run the timing advance stayed pretty consistent around 17 degrees BTDC. On the slow 13.20 run it was bouncing around between 11 and 17 degrees. I see this inconsistency in a lot of my runs this summer. Can a tune help with this?

runs.jpg

 

The higher rate of temperature increase with the 2nd HE makes me think you have them in series, adding restriction. If you were to have them is parallel, your flow would increase and both the temp rise rates would drop and the cooling rate would increase.

Woody

 

Yes, they are in series.

I could definitely be talked in to trying them in parallel. Something else to do this winter...

 

I tried 2 heat exchangers in series and saw the same negative results. I've run 3 different heat exchangers and they were all within 5 to 10º of one another in peak IAT's. The larger the heat exchanger the faster the temperatures fell. Ambient was within 5º of one another as well. Unfortunately my intercooler has a leak and I don't trust my data completely but it seems to jive with others experiences.

Les

 

Well all your runs have a nearly identical 60' and the trap speeds are within 1 mph of each other so way below any statistically significant difference. That pretty much says that no timing was retarded due to heatsoak with the stock cooling system and the car did not advance the timing any with the ice. So this brings me to my original point of methanol injection, not only it lowers the intake charge by anywhere from 60 - 100 degrees, it also makes fuel burn a lot more controllable sort of like running race gas so the car can be tuned for methanol and run a crap load of timing advance. Furthermore it's feasible to always run methanol if it's not your DD as in one tank of gas you would probably go through 3-4qts of the methanol mixture and could have a 5-6qt tank. Most cars running meth have a fail-safe that instantly retards the timing back to normal once it senses that you're about to run out of methanol so there is absolutely no risk to the engine... methanol gives actual whp and very notable gains without the inconvenience of carrying ice, having a giant tank, yada yada. I'm not knocking your project but I just don't see any real benefit or point to it, otherwise everybody would be doing it and they would sell ice at the drag strip lol.

 

Well, I'm spraying water/methanol before the throttle body and I haven't seen lower IAT's. The supercharger is cool to the touch after spraying! For our configuration, after the intercooler is necessary. The intercooler sucks all the cooling from the spray.

Les

 

Well even if you don't see lower IATs which is odd, you should be able to tune for meth and advance the timing nevertheless which should give you ponies as lower AITs don't seem to do so in our application. What meth concentration % are you running by the way?

 

lower iats dont give you ponies, it allows you to keep the ponies you have. by not retarding the timing you keep the power longer. the ecu wont advance the timing anymore than the tune is set for.
tune for meth and have a leakdown, run out because your failsafes didnt work and you could pop a 36,000 $ engine. this is a non forgiving way to advance timing and it only takes 1 issue to give you a huge paperweight

 

I've varied it but at the moment I'm using 60% meth and 40% water by volume. I was surprised too. Obviously the intercooler is sucking all the benefit out before reaching the IAT sensor. Oledoc is setting his up to spray after the intercooler and I'm awaiting the results. I'm not tuned for meth. My reason for spraying was to lower ECT's as well as IAT's and it hasn't lowered either one. I ordered a new intercooler from Rob because my original one is leaking. It's possible that the leak is impeding cooling. Time will tell.

Les

 

Yes and mostly no. Factory intakes on most cars suck in warmer air and also heatsoak so whenever the car warms up, you never have the horsepower that you potentially could. Cold air intakes do provide gains due to better flow but mostly due to much lower AITs which would not be there with a factory intake in the first place so yes you gain horsepower because the horsepower that the car is rated at takes into consideration the factory 'hot air intake'. You can also have a huge paperweight due to increased boost so why do we bother adding pullies? It's always about horsepower gain versus longevity and stress on the engine. Methanol systems are extremely reliable and have been around for so long, if it's done correctly there will be no issues. Look at Mitsubishi Evos, so many of those guys run meth with great results and no issues whatsoever. A conservative tune on meth will actually be easier on the engine, fuel burn is greatly improved without having have to resort to race gas. Why knock on something that so many people run in their cars with absolutely no problems.. in much more expensive cars than SRT or SLK for that matter.

 

im not knocking it . you perceive this to be a easy mod for the xfire and its not. tuning is critical and having an encrypted ecu that is programmed in german doesnt help. iats dont add power believe it or not. i have modded on these cars for yrs and never once saw a dyno improve power thru ice or lower iat. tune for the iat and power may inc. thats the tune not the iat.

btw look under your hood. you have a factory cai. the aftermarket cai that is used on these cars is for increased airflow not colder air flowing in. once the air from any intake you put on this car hits the s/c its superheated before the i/c AND the iat sensor.

 

I know lots of people who've used it or are using it. Some have multiple tunes and can switch back and forth depending on what's in the tank. For many people it's an economic decision. Meth is cheaper than race gas over the long run. They're the ones that blow up. With an aggressive tune it can happen fast. I use it as a margin of safety. I installed it to lower ECT's and IAT's but now use it to vary my AFR's since tuning is such a pain for me. I may invest in multiple tunes in the future.

Les