Yet another cooling (IC/HX) thread....kind of nerdy

[qcrazy]

They also sell a product named "TinyCWA", an add-on controller module for any of the pumps. Are you familiar with this? If so, how useful is it?

It is just a small controller that produces the PWM signal to control the CWA pumps. It essentially does what our ECU does, just gives the user more control over setting the operational parameters.

Can be used with or without a coolant temp sensor. Can also run some fans. Gives more control then most can get with our ECU. But obviously our ECU has way more parameters to work with.

I personally don't see the point, since I think they should always run 100%.

[IHave2Turbos]

yeah i dont think this is necessary especially when DS1 hard codes the pump at 95% duty cycle (correct me if that value is wrong)

https://tecomotive.com/download/manu...ntercooler.pdf

Apparently the CWA200 was also used by BMW not sure exactly how the flow/amp draw compares to the 100

https://mbworld.org/forums/attachmen...-waterpump.pdf

[qcrazy]

Yep, DS1 is 95%. It wouldn't change when I cut the wire either. It appears 95% is full speed on the CWA's anyway.

The CWA200 is a 15A+ pump. To be clear, the CWA200 is a low pressure pump. It is terribly suited for our application. The CWA400 (35A+) is a bit better, but it is huge pump. It is not a good fit for our system either...we can do much better.

CWA50, 100 and 150 match our system pretty well. Even better in combination. The other CWA options are not well suited.

[rodent]

Yep, DS1 is 95%. It wouldn't change when I cut the wire either. It appears 95% is full speed on the CWA's anyway.

The CWA200 is a 15A+ pump. To be clear, the CWA200 is a low pressure pump. It is terribly suited for our application. The CWA400 (35A+) is a bit better, but it is huge pump. It is not a good fit for our system either...we can do much better.

CWA50, 100 and 150 match our system pretty well. Even better in combination. The other CWA options are not well suited.

I got my CWA150 and Merc HX installed. There was a little confusion on which tube went where on the HX since I didn't mark the tubes taking off the old one but I think I got them correct. The relay power was tapped into the 12v bus bar under the strut bar in the middle. There is no good area on the jump start terminal. The black plastic cover there is not easy to remove without taking off the brace but can be done. There is a spare post there so you just need a nut or just put under an existing terminal. There is also a convenient threaded bolt hole to the left of that for the ground wire. For a relay harness I used a pre-made 40 amp that my local auto parts store had. It originally came with a switch which I removed and used the switch wire to tap into the old 12v pump wire. The wires going to the pump are pretty thick and should easily handle 15amps. It also came with a built in 40 amp fuse which I swapped for a 20 amp. It also came with protective sheathing around the wires. There's two output wires and you only need one so I just cut off the extra wires along with the white and black wire going to the switch which are not needed. Blue is the only one you need to trigger the relay.
https://www.oreillyauto.com/detail/c...it/hop3/nv3015

My biggest question to you is what is your secret way to burp/bleed the coolant system? I used this coolant system filler but not sure it got all the air out. The pump sounds like it may have air bubbles still so I'm hesitant on driving the car until then. This is the tool I used to fill:
https://www.amazon.com/gp/product/B01BW39HJS

Thanks again for all the testing you have done!

[qcrazy]

For wiring, I bent a connector over and put it below the jump post. Then lightly sanded down the cover so it sits flush. The hole next to my jump post is not tapped.
20210627_204625.jpg
20210627_204927.jpg

I haven't received my Merc yet...does it have a bleeder?

If so, the first step is getting that to bleed. I put in an auto bleeder, but I realize most people can't/won't do that. You don't need auto bleeders, they just speed it up a bit.

Then take off one of the hoses to the IC. If the system is air locked you won't have any coolant at the IC. If you have coolant start leaking when you go to take off the IC hose...you are not air locked. You can skip below.

If you are air locked, take off one IC hose. I prefer taking off the IC inlet hose (the one closest to the radiator)...but this may vary based on where you are air locked. Once off, pour coolant into the IC hose. It won't take much at a time...pour a little, wait, repeat. When you pour the coolant in you will hear air burp into the reservoir. Repeat many times...make sure coolant doesn't start coming out of the IC port.

You do not have to fully bleed it, just get enough air out to get it flowing. Hook the IC hose back up, and cycle the pump a few times and see if it is flowing. I pinch the IC outlet hose to see how well it is flowing. You will feel the flow in the hose...and the easier it is to squeeze, the worse it is flowing. If you can crush the hose easily, you likely aren't flowing much at all.

Repeat as necessary until your IC hoses are firm. Make sure you bleed the HX occasionally as you run the pump cycles. If you don't know, opening the door will cycle the IC pump. I cycle it until the hose is firm and doesn't change for a few cycles (and I feel no air flowing). Then I cycle it about 5 times every 5 minutes to get the entrained air out. Again, keep burping the HX.

[rodent]

For wiring, I bent a connector over and put it below the jump post. Then lightly sanded down the cover so it sits flush. The hole next to my jump post is not tapped.
20210627_204625.jpg
20210627_204927.jpg

I haven't received my Merc yet...does it have a bleeder?

If so, the first step is getting that to bleed. I put in an auto bleeder, but I realize most people can't/won't do that. You don't need auto bleeders, they just speed it up a bit.

Then take off one of the hoses to the IC. If the system is air locked you won't have any coolant at the IC. If you have coolant start leaking when you go to take off the IC hose...you are not air locked. You can skip below.

If you are air locked, take off one IC hose. I prefer taking off the IC inlet hose (the one closest to the radiator)...but this may vary based on where you are air locked. Once off, pour coolant into the IC hose. It won't take much at a time...pour a little, wait, repeat. When you pour the coolant in you will hear air burp into the reservoir. Repeat many times...make sure coolant doesn't start coming out of the IC port.

You do not have to fully bleed it, just get enough air out to get it flowing. Hook the IC hose back up, and cycle the pump a few times and see if it is flowing. I pinch the IC outlet hose to see how well it is flowing. You will feel the flow in the hose...and the easier it is to squeeze, the worse it is flowing. If you can crush the hose easily, you likely aren't flowing much at all.

Repeat as necessary until your IC hoses are firm. Make sure you bleed the HX occasionally as you run the pump cycles. If you don't know, opening the door will cycle the IC pump. I cycle it until the hose is firm and doesn't change for a few cycles (and I feel no air flowing). Then I cycle it about 5 times every 5 minutes to get the entrained air out. Again, keep burping the HX.

Thank you SO much for the info. I will try the bleeding procedure this weekend and see if that helps. I do have a Merc HX with the bleed gate valve on the left. I considered attaching a 12" hose with a bleeder valve at the end. Where are you getting your auto bleeder? Yeah I never realized how much that pump runs when you open the door. Afraid its going to kill the battery doing work on the car (opening/closing door).

Just to make sure I'm correct... Where should the bottom HX hose route to? Is this diagram accurate?

[qcrazy]

Yes, that diagram is correct for S6/7. Should come out the IC outlet (top port, furthest away from the radaitor), then down to the bottom port on the HX.

I just use a flow orifice and tie it into the radiator vent hose. You can use one of the really small orifices, don't need much flow to keep it bled.

You'll have to check what type of threads are in the Merc IC. The JHM had M12x1.25, so I just had to order an adapter to 1/4" NPT.

[rodent]

Yes, that diagram is correct for S6/7. Should come out the IC outlet (top port, furthest away from the radaitor), then down to the bottom port on the HX.

I just use a flow orifice and tie it into the radiator vent hose. You can use one of the really small orifices, don't need much flow to keep it bled.

You'll have to check what type of threads are in the Merc IC. The JHM had M12x1.25, so I just had to order an adapter to 1/4" NPT.

Thanks! Your info REALLY helped me get rid of the air lock I had. The pump definitely did not sound right until it was bled correctly. The Merc has 1/4" NPT. There's not much info out there on how to properly bleed these systems.

[sepheroth86]

qcrazy,

Would you be willing to work through these tests on a local 3.0t car if one were available?
Many of us A6/7 and S4 3.0t people are following these results closely.

[qcrazy]

Gladly. Need a car and a few gallons of coolant. Flow testing would be easy.

The logging might be hard though, since the 3.0's don't have access to DS2.

[qcrazy]

Been a while since I posted. Still waiting for the Merc Hx. But just a quick PSA...

Anyone looking at at running these large pumps, beware as these are demanding pieces of equipment. I am running an EMP. It is a loud pump, so it is pretty obvious when it is running. I ran some errands last night, heard the pump come on when I started the car. When I got home, I realized the EMP was not running. Checked the fuse quick, it was fine.

Took some time tonight to pull the bumper and take a look...
20211108_193444.jpg
20211108_194208.jpg

That is (was) the ground wire! It burned all the way through the ground wire. I even took time to size the wires for this too. Using a 30A fuse and a 40A relay. Not sure if I had a bad ground selected or what. But I'll take some time to re-check everything now that it is apart again.


Anyway, if anyone starts playing with these large pumps, do NOT cut corners. Luckily this was fused, but if I cut a corner and this goes just a little different, this could have turned out much worse. It is reasons like this that I recommend running new wiring for any pump larger than a CWA100.

[Dasquade]

Been a while since I posted. Still waiting for the Merc Hx. But just a quick PSA...

Anyone looking at at running these large pumps, beware as these are demanding pieces of equipment. I am running an EMP. It is a loud pump, so it is pretty obvious when it is running. I ran some errands last night, heard the pump come on when I started the car. When I got home, I realized the EMP was not running. Checked the fuse quick, it was fine.

Took some time tonight to pull the bumper and take a look...
20211108_193444.jpg
20211108_194208.jpg

That is (was) the ground wire! It burned all the way through the ground wire. I even took time to size the wires for this too. Using a 30A fuse and a 40A relay. Not sure if I had a bad ground selected or what. But I'll take some time to re-check everything now that it is apart again.


Anyway, if anyone starts playing with these large pumps, do NOT cut corners. Luckily this was fused, but if I cut a corner and this goes just a little different, this could have turned out much worse. It is reasons like this that I recommend running new wiring for any pump larger than a CWA100.

....or just an other good excuse to simply dump this water/air path and go a2a (it ain't all happy joy joy but less mechanical parts and potential failures with hopefully better results). Never the less, still praise you for doing all this testing!!
*offtopic: but yesterday discovered my fast acting valve wires corroded somehow and broke off...luckly i didn't used the kit and was just installed for potential later use).

[nthusiastt]

Best thread I've read in a very long time. Well done qcrazy, really well done sir!

One thing to add is the small turbos add a lot of heat because we're running a lot of pressure. If we ran really large turbos on lower boost with higher CFM's, it might help. This was my thinking way 20 years ago when I built a big turbo honda before that became mainstream.

[qcrazy]

Thanks!

Yeah, boost is certainly a key player. But it is much easier to say to use a larger turbo than it is to actually fit one. The packaging constraints on these engines is pretty crazy.

That said, I saw some data recently on the 3.0's, and from what I saw they actually look hotter than the 4.0T. A little apples to oranges, but I was bit surprised by the data. Maybe I shouldn't be surprised though...a roots SC is really just a heat pump at the end of the day.

[nthusiastt]

Big single with a hood Cowell. Someone needs to make a twin scroll that rotates the split turbine housing exhaust inlets 45 degrees from eachother. Custom exhaust manifolds obviously because the packaging is so slim. Oh yeah 10k later but we can rid ourselves of a good bit of plumbing dropping down to 1.

[qcrazy]

To finalize the wiring portion of this, I am going to post a quick update on the repair.

20211108_193444.jpg
Above I posted that the picture of my ground wire burned through. The EMP manual indicates to use 12 AWG wire minimum. I was using 12 AWG wire. I am not sure if the ground wire was a little under-sized or if was just bad wire, but it was shot. The wire burned through in one spot, but it was severely overheated over a foot away. It was gone!

The positive (feed) wire itself looked fine. No issues were visible. It was the "same" wire, but it was from a different spool since I used a different color. I was also using a 40 amp relay. The relay itself looked good, but the harness plug was clearly overheating. A portion of the plastic was melted, it was clearly getting a bit toasty! The EMP manual says this pump will use 25 amps, so I would expect the 40 amp relay to be fine. But the harness clearly was not up to the job. With that, I re-wired the whole thing.

I used 10 AWG wire when I re-wired it. On the EMP pump you can wire the main feed wire direct to the battery (with a fuse of course) and then just turn the pump on and off as needed with a second wire, so I utilized that to help my relay out.

I took the 10 AWG direct from the jump port (through a 30 amp fuse) and then directly to the pump.
20210627_204625.jpg
Previously, I ran this wire through the relay, but I changed that so not all of the amps will go through the relay. I also used the same 10 AWG wire for the ground, and I used a slightly different grounding point this time.

To activate the EMP pump I tee'd into the CWA pump power wire (red arrow below). I use that wire to activate the relay coil itself. Relay location shown with the orange arrow.
20211114_104515-Wiring.jpg

The relay coil is grounded to the relay mount (green arrow). The actual "on" signal to the EMP pump comes off the main feed wire that is running directly to the pump (yellow arrow-bigger black wire). The red wire (blue arrow) is tee'd into the main feed wire, and this runs through the relay and goes directly to the pump as the "on" signal. This causes the EMP to be on whenever the CWA is activated, ensuring both pumps run together. This also reduces the draw through the relay significantly. As stated previously, the pump motor takes 25 amps. While the "on" signal takes about 10 milliamps, quite a drop.
20211114_104502-Wiring.jpg

For those wondering, the other two black wires shown above are for my switch to the CWA PWM signal. I don't need it with DS1, but I left them anyway. The yellow wire is the relay 'NC' contact, and is not used, it is just taped up out of the way.

This is what I ended up with. We will see how it ages. A lot of the above is specific to the EMP itself, but some will still be applicable to other pump installations.

Update:
After running the EMP for a few months I no longer recommend running it on a relay triggered by the factory pump.

The EMP is just so big, it is really hard on the battery. Everytime you open the door, the factory pump cycles. Everytime you shut the car off warm, the factory pump after-runs for a few minutes. Over time the EMP can really wear on the battery. The more short trips the worse it gets.

While not ideal to un-synchronize the two pumps; with the EMP that is the better option based on battery life. SRM recommends using the fog lights or running lights as the trigger, and based on real-world experience with the EMP I agree. For all other pumps I still recommend running off the relay from the factory pump.

[IHave2Turbos]

so after all this work and 15 pages of a thread are we comfortable saying that most people should just buy the AMS kit? this seems like a very tortuous journey to prove their approach without saying so. asked another way, is there any potential upside to your configuration vs just buying from them?

[qcrazy]

Tortuous? This is just the scientific method, which rarely works in a straight line. Labor of love for me anyway.

Also, an interesting conclusion to read pages of data, and conclude one should buy items that were never tested.

But...I know what you mean. The big issue I have with AMS is I have been unable to figure out what pump they have. I have closely reviewed the higher head pumps that work well on our system, and to date I have not been able to match it up. If they used a low head pump, it would be marginally helpful. Since I don't know the pump, I cannot possibly say that it makes sense.

Also, I think it is over-priced. I can have a reservior fabricated, buy a JHM Hx, and install the EMP for about the same money. And while I don't know what pump AMS uses, I know for sure the EMP is larger.

Finally, my later testing with more flow indicated that the benefit from the reservoir is marginal, if it even exists at the limit. I have shown that the charge cooler is the limiting factor in this system, and sucking water out of a swimming pool isn't going to fix that. If a reservoir improves your temps, your pump(s) are too small.

I need to get a pump summary post up to tie some of this together. Maybe I'll have some time this weekend. But I have one more test I am going to try tomorrow...while I keep waiting for the Merc.

[KenworthT680]

@qcrazy, was one of the IC/HX you tested the CTS Brand?

[qcrazy]

@qcrazy, was one of the IC/HX you tested the CTS Brand?

I did not test the CTS.

So far I have tested:
--Stock S6/S7 Hx
--CSF #8111 Hx (e.g. No Name)
--JHM HX

Hopefully before I die, I'll get to test the Merc Hx too. At this point every Hx I have tested significantly out performs the stock Hx.

One thing I will note here is that the stock S6/S7 (4.0) Hx, is also the same (3.0) Hx used in the 2013+ S4/Q5 and the 2016+ A6.

[Colter84]

@qcrazy, I just got done reading this from start to finish! Man you are doing Gods work for SURE!!! I am beyond appreciative of the ridiculous amount of time and money that you've spent to do all of this research, and then take the time put the detailed data together and share it with all of us!

With that said, I have some questions that I did not see directly discussed.

1. What would your opinion be on running a CWA150, instead of the CWA100, along with the OEM CWA50. Basically I'm looking at either that setup, or the EMP29/32. If I went with the EMP, could it be run with the stock CWA50, or would it be too small and hurt rather than help?

2. Had you given any thought to running two OEM HXs, or do you think the aftermarket options would outperform that setup? If someone was to run two OEM HXs, would they be best suited in parallel, or series?

Looking forward to your opinion on this! Thanks!

[qcrazy]

Since I had to pull the bumper off again to fix the EMP wiring, I figured I would take a look at everything and inspect my previous installation.

Previously I had put a large hose (1") in from the EMP up to the IC. Since I had access again, I went ahead and inspected the hoses to see how they looked, as some areas had pretty tight fits. I installed "softeners" to protect the hoses, but I wanted to check anyway. The hoses were all technically okay, but in a few places they had some pretty large creases. And in a few areas the "softeners" were really getting some pressure applied. Since I don't really want to have a sudden failure, and I didn't really have space to relocate the hoses, I decided to pull out the 1" hose and install a 3/4" hose. And if I was changing the hoses, I had to drain the system...again. So...I didn't want to waste the opportunity to try and find some more cooling, I went ahead and looked at the hose routing to see if we could improve anything.

Hose routing had previously been discussed a few months ago on here, but re-routing is not easy so I kind of skipped passed it. But since I have now already re-run the hoses, it wasn't as big of a deal. Anyway...this is really stealing an Audi idea.

On the S6/S7 the routing goes: Pump --> IC --> HX. This is shown in the SSP:
S6-S7Cooling.jpg

However, on the A8/S8/RS7, Audi changes this and it goes: Pump --> HX --> IC. This is also in the SSP, but not as obvious...so I added some edits....
S8-RS7Cooling.jpg

The advantage here is simple, running it through the HX after the pump allows the HX to take all of the (as much as possible) heat before the coolant hits the IC. Again, all HX's (IC's included) work best with larger temperature differentials. The larger the temperature delta, the easier it is to move the heat out. This re-routing takes advantage of this. The other advantage that I have on my set-up, is I am running a lot of "pump", WAY more than the stock set-up. I am just pouring energy in to the system trying to maximize the flow. Some of this energy turns into heat in the coolant...since I have more energy input, I will have more heat as well. On the typical S6/S7 routing, this pump heat gets added to the coolant right before it hits the IC...squeezing that coolant temp differential just a bit for the IC.

Using the factory hard pipes and hoses on the S6/S7 would not be possible to make this re-route. The hard pipes have to come out. One hose, can be re-used, but relocated a bit. The others all need to be re-run. On the EMP set-up I needed a 180 hose and a few 45 hoses to make it work. After that it was pretty easy. I used pretty much all 3/4 hoses, expect for right at the EMP which takes 1" hose.

Here is the side by side with the other JHM run...
SbyS - JHMRerouteIAT.JPG
A couple minor points on this one. First, we improved the recovery with this, the blue arrow is showing better recovery between runs. It improved faster, and gets to a little bit lower temperature. Interestingly, you'll see the yellow arrows shows LOWER heat removal from the IC. Lower temp removal on the one with the lower IAT's...how is that possible? It had lower temps (heat soak) to start with, which can be seen more below. Finally, you can kind of ignore the first pull on the re-route chart...my tires were cold and I could not get traction on the first run until I got into third gear. Not super fun.

SbyS -JHMRerouteTemps.JPG
Here the yellow arrow shows improved pre-turbo (heat soak) recovery. Meaning we are reducing heat soak with this change. How much? The blue arrow shows significantly less PEAK heat soak with the new routing...about 15F less. This means there is less work for the IC to do, so the overall IAT's are reduced. Obviously, not by all 15 degrees, but a portion of that is gained overall. Why is potential heat (heat soak) coming down? I'll cover that in another posts about the pumps, but this change makes a step change in this area.

Here are the current reductions....
JHMHX-IATbar3.JPG

With that change we got another ~3F. We went from 66.24F to 63.52F. Not a bad gain late in the game, considering this all started at 93F.


I have been working on some different runs lately, before it gets any colder. This is something I am preparing for the 3.0 guys, getting a baseline set (tease). I set the 4.0T baseline with the set-up outlined above this weekend. However, in setting the baseline I decided to get my B5 S4 out as a comparison...just to see what a2a looked like. My B5 is a Stage 3+, just about everything is modified. Needs a little love this winter, but I wanted a benchmark for the 4.0T. This is a w2a vs an a2a, same run, for benchmarking. Here is what I got....
40vS4.JPG

I left the stock run on the chart above, but it was for a different pull. Stock is really apples to oranges here. In reality the stock run would have been higher, most likely above 100F, but I left it on here to provide some context on where this all started. As far as the a2a, I thought the 4.0T held up pretty well here. 4.0T still isn't where I want it, but a lot more comparable to an a2a than I anticipated. The actual "heat gain" between the 4.0 (40 to 70F) and the a2a (10 to 40F) is about the same. The 4.0 just starts with so much more heat soak. That is really the difference between the two here.

That about covers it. Still no Merc HX here. I need to get a pump summary post up. And I'll be starting a new 3.0 cooling thread one of these days. If you think the 4.0 is bad, wait until you see the data on the 3.0's.

[qcrazy]

@qcrazy, I just got done reading this from start to finish! Man you are doing Gods work for SURE!!! I am beyond appreciative of the ridiculous amount of time and money that you've spent to do all of this research, and then take the time put the detailed data together and share it with all of us!

With that said, I have some questions that I did not see directly discussed.

1. What would your opinion be on running a CWA150, instead of the CWA100, along with the OEM CWA50. Basically I'm looking at either that setup, or the EMP29/32. If I went with the EMP, could it be run with the stock CWA50, or would it be too small and hurt rather than help?

2. Had you given any thought to running two OEM HXs, or do you think the aftermarket options would outperform that setup? If someone was to run two OEM HXs, would they be best suited in parallel, or series?

Looking forward to your opinion on this! Thanks!

You can run a CWA50 and a CWA150 no problem. On the stock system that would increase your flow about 80%. I have the graph, but it is kind of messy so I am not going to post it. You can also run a CWA50 with an EMP. The CWA100 just gains you more with the EMP, and since I had it from my testing I used the CWA100 instead. But the CWA50 would work just fine as well.

The OEM HX is crap. Get it out of there. Running two of them would actually make the cooling worse. Someone bought a JHM HX and then took the time to plumb up the JHM and the OEM HX, and his temperatures got worse so he pulled out the OEM. These cars are not HX limited anyway, when cruising you have more than enough air flow of cool down the coolant. My NN HX was smaller than the OEM HX and it performed significantly better.

[sepheroth86]

That about covers it. Still no Merc HX here. I need to get a pump summary post up. And I'll be starting a new 3.0 cooling thread one of these days. If you think the 4.0 is bad, wait until you see the data on the 3.0's.

I am very much looking forward to this.

I have been eyeing the AMS HX kit because it seems to tick most of the boxes you had covered here (better HX, additional pump) while being very plug and play.

[Colter84]

You can run a CWA50 and a CWA150 no problem. On the stock system that would increase your flow about 80%. I have the graph, but it is kind of messy so I am not going to post it. You can also run a CWA50 with an EMP. The CWA100 just gains you more with the EMP, and since I had it from my testing I used the CWA100 instead. But the CWA50 would work just fine as well.

The OEM HX is crap. Get it out of there. Running two of them would actually make the cooling worse. Someone bought a JHM HX and then took the time to plumb up the JHM and the OEM HX, and his temperatures got worse so he pulled out the OEM. These cars are not HX limited anyway, when cruising you have more than enough air flow of cool down the coolant. My NN HX was smaller than the OEM HX and it performed significantly better.

Thanks for the reply! After pouring over all of your very well laid out graphs and data, I finally decided to go with running a CWA150 along with the CWA50. I'll make the plumbing change that you discovered works a little better also. I also ordered a CTS HX as they're on sale for a decent price at the moment, and in stock. I figure it'll probably perform about the same as the JHM. I really wanted the new Mercracing one, but I'm not sure how long I'd have to wait, and I'm assuming it's probably not going to be game changing compared to the JHM data you already have. If it turns out that it is, I can always upgrade later.

From what I could tell, the difference between running the CWA150/50 combo vs the EMP wouldn't be a huge improvement in performance, and I couldn't find the EMP in stock at the moment anyways. Basically the cost vs performance and ease of install sweet spot kind of seemed to be the CWA150/50 combo. Especially since I could find the 150 for cheaper than the 100 a lot of places. I'm going to wire up the 150 like it's an EMP in case I want to upgrade later of course.

I think I can safely speak for the community
by saying thanks again for all of your hard work on this subject so that we can make informed decisions on improving our cooling systems! Can't wait to see more data as it comes out!

[qcrazy]

I ran another comparison to measure the progress on the 4.0T.

Right now I am warming the car up, cruising on the highway to get everything well up to temperature. Once the IAT's stabilize, I am doing a 0-124 mph (200kph) pull, trying to build some heat and see what we get. I have ran my C7 S6, my B5 S4, and my B9 S4. All these logs are being run on VCDS.

The cars are set-up as follows:
C7 S6:
Boost: Stage 4, DS1 OTS, TS1 turbo's, 91 octane map.
Cooling: w2a - CWA100&EMP Pumps, JHM HX, Stock IC, and the hose routing changes.

B5 S4:
Boost: Stage 3+, 996 compressor wheels in K04's, Downpipe and exhaust and all that stuff. Running 91 with a kick of E.
Cooling: a2a - ER SMIC's

B9 S4:
Boost: Stage 1, 91 octane map, stock turbo
Cooling: a2a - Wagner FMIC

40vS4(x2).JPG

Now clearly the 4.0T is running warmer. But looking closer, pretty pleased with the results. The main thing I look at is the heat rise during the actual pull, which are as follows:
C7 S6: 69-38.4 = 30.6F
B5 S4:40.6-10 = 30.6F
B9 S4: 29.6-4.4 = 25.2F

Interestingly, the B5 and the C7 have the exact same heat rise. The B9 is a little bit lower, but that it not surprising since that is still running the stock turbo...and just a stage 1 tune. In my mind, the C7 IC is performing pretty well at this point, it is hanging right with the a2a's as far as heat rise goes.

The issue that is still visible in the 4.0T is the heat soak. An interesting point here is that the B9 is also a hot vee, but's its recovery temps are so much better. The B9 is seeing temps that are less than 10 degrees over ambient, that is pretty impressive.

The 4.0T IC can keep up pretty well at this point, but it is the the heat soak that is ultimately running the temperatures up. Increasing flow and changing HX's has helped the heat soak some, but not enough. The next item I need to look at is a fan. That will be the next item I look into closely...I am just waiting to select a HX at this point. In reality, this probably isn't going to happen until Spring. I don't drive the C7 in the winter.

I am going to try hard to get a pump summary up over this long weekend. In that, there will be some data regarding heat soak effects, and the role the reservoirs make on all of this. I'll do my best to get something up soon. Still trying to figure out how to best package the information.

[irish07]

I ran another comparison to measure the progress on the 4.0T.

Right now I am warming the car up, cruising on the highway to get everything well up to temperature. Once the IAT's stabilize, I am doing a 0-124 mph (200kph) pull, trying to build some heat and see what we get. I have ran my C7 S6, my B5 S4, and my B9 S4. All these logs are being run on VCDS.

The cars are set-up as follows:
C7 S6:
Boost: Stage 4, DS1 OTS, TS1 turbo's, 91 octane map.
Cooling: w2a - CWA100&EMP Pumps, JHM HX, Stock IC, and the hose routing changes.

B5 S4:
Boost: Stage 3+, 996 compressor wheels in K04's, Downpipe and exhaust and all that stuff. Running 91 with a kick of E.
Cooling: a2a - ER SMIC's

B9 S4:
Boost: Stage 1, 91 octane map, stock turbo
Cooling: a2a - Wagner FMIC

40vS4(x2).JPG

Now clearly the 4.0T is running warmer. But looking closer, pretty pleased with the results. The main thing I look at is the heat rise during the actual pull, which are as follows:
C7 S6: 69-38.4 = 30.6F
B5 S4:40.6-10 = 30.6F
B9 S4: 29.6-4.4 = 25.2F

Interestingly, the B5 and the C7 have the exact same heat rise. The B9 is a little bit lower, but that it not surprising since that is still running the stock turbo...and just a stage 1 tune. In my mind, the C7 IC is performing pretty well at this point, it is hanging right with the a2a's as far as heat rise goes.

The issue that is still visible in the 4.0T is the heat soak. An interesting point here is that the B9 is also a hot vee, but's its recovery temps are so much better. The B9 is seeing temps that are less than 10 degrees over ambient, that is pretty impressive.

The 4.0T IC can keep up pretty well at this point, but it is the the heat soak that is ultimately running the temperatures up. Increasing flow and changing HX's has helped the heat soak some, but not enough. The next item I need to look at is a fan. That will be the next item I look into closely...I am just waiting to select a HX at this point. In reality, this probably isn't going to happen until Spring. I don't drive the C7 in the winter.

I am going to try hard to get a pump summary up over this long weekend. In that, there will be some data regarding heat soak effects, and the role the reservoirs make on all of this. I'll do my best to get something up soon. Still trying to figure out how to best package the information.

Could it be, the 4.0T IAT sensor itself is heat soaking more than the actual air temp? I’ve seen this on other vehicles, the air at the sensor location was nowhere near the temp of the sensor. The sensor metallic body would act as a heat element and false the sensor to read higher(aka..just poor design of sensor). On this particular vehicle(Jaguar F-Type); Blowing heat at the exterior sensor location(post blower temp, pre IC), would massively raise the post blower IAT reading, but the post IC readings where the same.


Sent from my iPhone using Tapatalk

[michael.kaemmer]

I ran another comparison to measure the progress on the 4.0T.
...

The 4.0T IC can keep up pretty well at this point, but it is the the heat soak that is ultimately running the temperatures up. Increasing flow and changing HX's has helped the heat soak some, but not enough. The next item I need to look at is a fan. That will be the next item I look into closely...I am just waiting to select a HX at this point.

Good idea!
You are running DS1. You may use both fans removing heat. I am using switch 6 pretty often especially after highway rides. Further I altered some temperature setting in section with tuner pro helping C7.5 to run fans more often depending on temps for coolant, oil, trans oil. Helps a lot!

[qcrazy]

Could it be, the 4.0T IAT sensor itself is heat soaking more than the actual air temp? I’ve seen this on other vehicles, the air at the sensor location was nowhere near the temp of the sensor. The sensor metallic body would act as a heat element and false the sensor to read higher(aka..just poor design of sensor). On this particular vehicle(Jaguar F-Type); Blowing heat at the exterior sensor location(post blower temp, pre IC), would massively raise the post blower IAT reading, but the post IC readings where the same.

I looked at that in Post 49, and then expanded on it later. But the short answer is that is not the issue here. I wish!

The longer answer is that the 4.0T has a couple of temp sensors so we know the numbers are real. It has the G42 sensor, which is the intake air sensor. This sits on the drivers side intake manifold down below the coolant reservior. Not a great place, but not terrible either considering this is a hot vee and the main heat is all up high.

The 4.0T also has two temp sensors in the throttle-body, pre-IC, G673 and G674. These sit right in the turbo oven, nice hot location!

When you make multiple runs, and pull after pull, you can look at the relationship between these sensors. I have about 8GB of data now to look at, and when going over this data these sensors line up all the time. What the pre-IC temps show aligns with the expectations on the post IC temp, and vice versa. One would hope you would see some sort of anomaly over time in certain cases, but you just don't. I even tried blowing air on the sensors while driving.

Of all three cars, the B5 intake temp sensor is in the worst location. Buried between the engine and the IM, on top of the engine.

[qcrazy]

Good idea!
You are running DS1. You may use both fans removing heat. I am using switch 6 pretty often especially after highway rides. Further I altered some temperature setting in section with tuner pro helping C7.5 to run fans more often depending on temps for coolant, oil, trans oil. Helps a lot!

Good call out Michael. Not sure I love switch/map 6, since it is a stock map. Maybe okay for short term.

But the TunerPro may be a great option for some! I don't have access right now, and not sure I would have the time to learn it. But for those already in that space it would be a step in the right direction for sure!!

I also like the idea of adding a fan to the HX, since in my case I increased the HX size. Since the HX is now bigger, so is the airflow restriction through the HX stack. If the restriction is worse, adding a fan to deal with that would be appropriate.

Speeding up the existing fans would be good too, but in my case I think more fans would also be needed to get where I want to go. If I get ambitious I may try and do both. I may be reaching out for help this winter if so...

[michael.kaemmer]

Good call out Michael. Not sure I love switch/map 6, since it is a stock map. Maybe okay for short term.

But the TunerPro may be a great option for some! I don't have access right now, and not sure I would have the time to learn it. But for those already in that space it would be a step in the right direction for sure!!
..

Axxxxx,

You are right! BUT:

When do you need extra cooling? Low speed or standing still? 90% fan dc reduces heat in minutes!

You may alter switch 6 data with data of you switch 0 or 1 or alter min fan speeds to whatever is needed. Already couple of minutes reduces the heat; you see this when logging. You may also do this in your original switch. Every switch section has its own fan min speed settings. Normally starting at 10%. I have changed that starting with 20%. Made good results with this.
Michael

[qcrazy]

You can use my name on here Michael. That's fine. Appreciate you pausing.

I hear you, and I think I am going to follow your lead here. I still think I need to increase the energy input due to the increased restriction. However, I ran the air flow calculations last night and the required air flow does not match the fans for the area. Said another way, I need larger fans than I have space available.

So I see three options:
-Edit maps with TunerPro.
-Swap out a factory fan with a high flow fan (this is what I did on my B5).
-Drop the HX lower so I have room for a larger fan.

Of those three, TP is the cheapest and easiest. So I will probably start there...and may bug you at some point if I get stuck. The nice thing is, since this is hest soak testing, I will be able to test this over the winter.

I may look into a reposition of the hx too, but that would be next year when I select a final hx. The JHM is great, but I don't love how it affected the recovery. All the big hx's may do that, but I don't know that yet.

[qcrazy]

This post has been overdue for a while. TBH, I am not sure how to write it. I have so much data, but it is hard to boil down and put into small digestible pieces. This post is going to be a high level overview, should be relatable to the most people. I will dig in deeper with one or two later posts, with more of the nerdy data.

The simple thing I would like people to take away from these 300+ posts, is that if you modify your 4.0T, you need to do something with your cooling.

Upgrading your car and not addressing your cooling, is not too far away from throwing your money away on these. You don't have to drag race or track the car. If you upgrade and just simple drive the car, you really need to address the IC cooling. This post is going to focus mainly on pumps. I will deal with hx more in a later post when I have more data. But both the pump and hx play a role here. Relying on one will only get you so far. You really should address both to get to a better place.

Let's look at IAT's, that is what matters at the end of the day. As always, I am looking at the delta...
IATsStockHX.JPG
Above you will see the effect of both pumps and hx's. This data is from my last pull (2-3-4), so keep in mind this is from a warmed system. But this is more realistic to every day driving. All of my testing also has the pumps running full speed at all times.

The grey line above is the stock pump and stock hx (S6). This is showing 93F of heat rise. The 4.0T at about 140F (60C) will start reducing boost to lower heat to protect the engine. What this 93F heat rise means to us, is that if it is 50F outside, your temperatures will start to get into this 140F window. Meaning, you will start losing boost. If it is 80F out on the stock set-up, your car will really start pulling boost. This is why I say not upgrading the cooling system is pretty close to throwing money away if you upgrade w/o it. For those of you running stock cooling systems that can do a pull or two and then the car falls apart...this is the reason why. Your IC cooling system gets warm, and then the ECU pulls boost.

With that being said, if you look at the grey, red and olive line, you will see the direct effect of pumps when working with the stock hx. Just using a CWA100 on the stock hx gets you the red line, which is better (85F), but still pretty warm. The CWA100/EMP (olive line), is starting to get quite a bit lower temps (73F).

Now, if you compare the grey to orange line, that is the same pump with a change in HX (stock to NN). The orange line is much better than the gray. But there is still room for improvement.

Comparing the red to blue line, shows us the difference in using the CWA100 with a different HX. Interestingly, here you can see that the blue line is now the same as the olive line. Or in other words, a CWA100 matched with the right HX performs about the same as the CWA100/EMP pump with the stock HX. The thing to note here is that the stock HX (olive line) is about as far as the stock hx will go, it has some serious pumps stacked on to make this level. Yet, the blue line is really just getting started from a pump perspective. The NN HX is performing well here, and I can add more pumps on to improve it further.

So what does the NN HX look like with different pumps...
IATsNNHX.JPG

The orange and light blue lines were on the last graph, and you can see that it drops from there. However, the gains do become more and more difficult as you try push temps down. If you compare the pump combinations you will notice that the lowest IAT's are NOT with the largest pump set-up. There is a reason for that, and it has a lot to do with the reservoir, but I will save that discussion for another post. I will say that the best on that graph, the dark blue line is at 66F. That is 7F better than the lines discussed on the chart above. So consider that when aligning your budget with your options. 73F may be just fine for some of us (22% improvement over stock).

Recommendation:
(More specific 4.0T recommendations here)
The pump should be upgraded. Put in as much pump as you can. Long term this is the best option.

Options:

• If your budget and skill set is limited, than just put in a CWA100. It is a very easy swap on these cars.
• If you have good DIY skills, a good budget friendly option is a CWA50/100 combo. The CWA100 can be installed similar to as shown in the EMP install post.
• A similar, and slightly better option, is to install a CWA50/150 combo. This will pump right around the EMP alone.
• The high end is the EMP pump. This is the biggest pump that fits in well with our system pressures. You can run this alone, or with any of the CWA pumps mentioned above. This is a large pump and a pretty complicated install. This is also a very expensive pump.

I would say pick the one that fits your budget and skill set the best. I also recommend a HX replacement as well, so consider that in your budget.

The use of reservoirs on our cars is questionable. They can work, but they have to be sized to match the hx and the pump(s). If that is not done correctly, the reservoirs will make things worse.

One thing I will call out here, that I have discussed previously, is that while the temperature values above are still a bit high. The temperature gain during the pull (rate of gain) is starting to be pretty comparable to an a2a system at the lower values shown above. While the temperature value itself is not great, the IC response it looking much better; and that is progress. The high value itself is a direct issue with heat soak, and that still needs to be figured out. We do have some clues to work on from the testing completed so far, and that will be discussed more in another post.

There is a lot going on here that I am leaving out of this post, around heat soak temps and recovery time and a few other items. So my recommendations may not align perfectly with this information shown, but I will get into more details later as to not drag this post out forever.

[Dasquade]

What about externally nitrous/water spraying of the HX in case of w2a or intercooler in case of a2a? I know it has been done, altough always tough water only is actually a bad thing but nitrous helps. Might be more effective versus fans as personally i didnt saw much improvement when i ran map 6 on my w2a setup.

[IHave2Turbos]

Im glad this was worthwhile https://www.audizine.com/forum/showt...0#post14531680

[qcrazy]

Im glad this was worthwhile https://www.audizine.com/forum/showt...0#post14531680

Yep! I was thinking of that exchange when I decided to give it a try!! Seemed like a waste to take off the bumper and not try something different. I figured I am running so much pump, if anyone would see an improvement it would be me.

I was surprised by the amount reduced (3F at this stage is a lot). I actually tested it twice on back to back days to confirm the results.

[qcrazy]

What about externally nitrous/water spraying of the HX in case of w2a or intercooler in case of a2a? I know it has been done, altough always tough water only is actually a bad thing but nitrous helps. Might be more effective versus fans as personally i didnt saw much improvement when i ran map 6 on my w2a setup.

I set all my cars up for tracking, even if they never see the track. I don't like using things that can run out (other than gasoline). Just the approach I have always used. If I drag raced, I would approach this completely differently.

Are you collecting data or just glancing at a gauge as you are driving? You have no clarity on the values just looking at a gauge. When driving around I thought my B9 ran near ambient all the time. Then I logged it and it turned out that "near ambient" meant a 25F temp gain on a pull. Big difference! I hope you are planning on actually logging data on your a2a thread.

[qcrazy]

Just real quick....I wanted to look into the fan option, so, why the in-laws were in town, I ran my heat soak route to do some quick testing.

I wanted to verify if this was the right direction to go...so I ran one new test with the hose routing, and one on "max cool" (full fan) map. This is a heat recovery route I have been running for months, almost all of it is me driving at 60mph (about 10 minutes total). Here is the pre-IC chart...
MaxCoolingPreIC.JPG
Not much for differences here pre-IC, just really a lot of noise. Nothing I would call out as material at this point. Again, this is the pre-IC temp. You can see I started the run a little warmer with the "max cool" map (dark blue line). But it still looks like it lines up with with others. I would say the pre-IC temps are the same for all of these runs.

Here is what the Post-IC temps (IAT's) looks like...
MaxCoolingPostIC.JPG
Now we can see some differences. Here you can see the baseline (yellow line) post-IC temps clearly run higher after the little acceleration. Taking longer to recover from the heat jump. This is running a much smaller pump and the smaller HX, so that makes some sense.

The next three runs are all running them same pumps and hx's and they all look about the same. Even the "hose routing" (grey line) fits in the same general range.

But the "Max Cool" (dark blue line), is clearly running colder. This is the same pump and HX as the three above, with only the fans getting cranked up here. It starts higher since the car was a bit warmer when I started, and then drops like a rock, and pretty much stays low, even with the little acceleration in the middle. This is picking up another ~8-10F at the end of the run. This is pretty substantial!!

It looks like the fans are the correct direction to head at this point. I'll be getting more into this in the future. We'll see what we can figure out.

[MattyMarkey]

Just real quick....I wanted to look into the fan option, so, why the in-laws were in town, I ran my heat soak route to do some quick testing.

I wanted to verify if this was the right direction to go...so I ran one new test with the hose routing, and one on "max cool" (full fan) map. This is a heat recovery route I have been running for months, almost all of it is me driving at 60mph (about 10 minutes total). Here is the pre-IC chart...
MaxCoolingPreIC.JPG
Not much for differences here pre-IC, just really a lot of noise. Nothing I would call out as material at this point. Again, this is the pre-IC temp. You can see I started the run a little warmer with the "max cool" map (dark blue line). But it still looks like it lines up with with others. I would say the pre-IC temps are the same for all of these runs.

Here is what the Post-IC temps (IAT's) looks like...
MaxCoolingPostIC.JPG
Now we can see some differences. Here you can see the baseline (yellow line) post-IC temps clearly run higher after the little acceleration. Taking longer to recover from the heat jump. This is running a much smaller pump and the smaller HX, so that makes some sense.

The next three runs are all running them same pumps and hx's and they all look about the same. Even the "hose routing" (grey line) fits in the same general range.

But the "Max Cool" (dark blue line), is clearly running colder. This is the same pump and HX as the three above, with only the fans getting cranked up here. It starts higher since the car was a bit warmer when I started, and then drops like a rock, and pretty much stays low, even with the little acceleration in the middle. This is picking up another ~8-10F at the end of the run. This is pretty substantial!!

It looks like the fans are the correct direction to head at this point. I'll be getting more into this in the future. We'll see what we can figure out.

This sounds very much like what SRM sells and likely why they sell it; upgraded EMP pump and Spal fan w/ reservoir.


Sent from my iPhone using Tapatalk

[sepheroth86]

The use of reservoirs on our cars is questionable. They can work, but they have to be sized to match the hx and the pump(s). If that is not done correctly, the reservoirs will make things worse.

How does one calculate the correct reservoir size? I have always been under the impression the quick and easy answer is, the bigger the better when talking a2w on a street car. That way it takes longer to warm the whole system.