48 replies to this topic

[Mattias]

Hi, 

I've had some cooling issues and also oil smoke from turbo that I suspect is caused by poor cooling of the turbo. I do have a Garrett ball bearing turbo but this probably apply to oem setup also.

 

First off I couldn't find much info on the cooling circuit for the VXT online so I made a layout myself. And now some problems becomes clear:

 

1. Turbo won't see any cooling with thermostat closed unless it self circulates. In my setup with 82C stat and warm engine after track driving I saw no water in the return line from turbo unless I revved engine quite much.

 

2. The electric aux pump that I believed was there to cool turbo after warm shut down actually only circulates the water from cylinder head to the heater matrix...

 

I basically think the circulation is not good enough for serious track driving, from 105C it takes me to cool down laps to get down to 95C with pro alloy cooler and twin spal fans. And it won't cool down if left on idle in pit, I have to hold it on about 2000rpm for the temp to drop.

 

It would be nice if everybody could pitch in with their experiences on this and eg check re-circulation from turbo (small forward facing tube on expansion tank) on idle and after shut down.

 

Also comment in case I made some error in the circuit pictures...

 

Here you have a white paper from Garrett on the importance of turbo cooling: https://www.turbobyg...ter_Cooling.pdf

 

Speedster/VX220 Turbo Cooling circuit PDF: https://1drv.ms/b/s!...TITXQoL5-RbXey8

 

 

 

 

I'm now thinking of adding an additional large electric pump to aid circulation:

 

[Nev]

Excellent pics, thanks, as I never really knew where everything went.

 

 

[Nev]

Looking at the diagram, you could re-route the pipe from the throttle body (which serves no purpose in a hot VX 220 engine bay as your TB will never freeze in reality) and re-direct it into the turbo. This would mean your turbo sees hotish water as opposed to cooled water, though in practice it's probably only 25 degress C. diff, so really not much compared to the turbo temps.

 

 

Edited by Nev, 16 July 2016 - 02:50 PM.

[glitch]

On the Elise for tuned engines it is quite common to add a craig davis electric water pump to help the oe pump at low revs, standing in traffic or on track in the pits. This allows you to circulate water also after shutting engine off. Don't know about cooling the turbo though.

[GeorgeBC]

Ive had cooling issues.

 

I added a ramp in the crash box to push the cold air straight through the radiators. I just used a neoprene sheet and the ramp goes from the back of the crash box to the rads with just the two cut outs for the heaters. Ive added Phenolic spacer between the CC rad and the coolant rad to close the gap forcing all the cool air through both rads. Thats the problem having a CC rad blocking the coolant rad.

 

My coolant pipes to the throttle body are just joined together as Nev said you don't require them to heat the TB.

 

I added an extra oil cooler radiator in the passenger side ear. Ive got a thermostat on the oil filter sandwich plate so it only circulates through the cooler when up to temperature. I also added a louvered vent in the sill and wheel arch liner behind the cooler core to allow the hot air to escape. 

 

This certainly improved the cooling but id still like it to be better myself. Darron Wall ran a bigger electric water pump on his VXT in the same place that you have thought of and that really helped the cooling on his car. This was done at the same time as cutting out the boot and adding vents on the rear clam to extract the heat so it wasn't really conclusive as to what helped the cooling. 

[Mattias]

Looking at the diagram, you could re-route the pipe from the throttle body (which serves no purpose in a hot VX 220 engine bay as your TB will never freeze in reality) and re-direct it into the turbo. This would mean your turbo sees hotish water as opposed to cooled water, though in practice it's probably only 25 degress C. diff, so really not much compared to the turbo temps.

 

 

 

I actually think it's "cold" water running to the TB if the pipe housing isn't devided inside. But I already rerouted those pipes to cool my external waste gate.

[Nev]

 

Looking at the diagram, you could re-route the pipe from the throttle body (which serves no purpose in a hot VX 220 engine bay as your TB will never freeze in reality) and re-direct it into the turbo. This would mean your turbo sees hotish water as opposed to cooled water, though in practice it's probably only 25 degress C. diff, so really not much compared to the turbo temps.

 

 

 

I actually think it's "cold" water running to the TB if the pipe housing isn't devided inside. But I already rerouted those pipes to cool my external waste gate.

 

 

It's funny, I was thinking to do exactly that as soon as I saw your diagram.  

[Mattias]

Ive had cooling issues.

 

I added a ramp in the crash box to push the cold air straight through the radiators. I just used a neoprene sheet and the ramp goes from the back of the crash box to the rads with just the two cut outs for the heaters. Ive added Phenolic spacer between the CC rad and the coolant rad to close the gap forcing all the cool air through both rads. Thats the problem having a CC rad blocking the coolant rad.

 

My coolant pipes to the throttle body are just joined together as Nev said you don't require them to heat the TB.

 

I added an extra oil cooler radiator in the passenger side ear. Ive got a thermostat on the oil filter sandwich plate so it only circulates through the cooler when up to temperature. I also added a louvered vent in the sill and wheel arch liner behind the cooler core to allow the hot air to escape. 

 

This certainly improved the cooling but id still like it to be better myself. Darron Wall ran a bigger electric water pump on his VXT in the same place that you have thought of and that really helped the cooling on his car. This was done at the same time as cutting out the boot and adding vents on the rear clam to extract the heat so it wasn't really conclusive as to what helped the cooling. 

 

I sealed of my rad and pre-rad with nitto (rubber foam) so all air passes both coolers.

An additional oil-cooler is on my list to sort out, however I don't think you need the thermostat because there is one in the LET oil pump assembly already.

Would have been interesting to know how much the oil cooler raise the water temp before it enters the main pump

 

I did a test on track yesterday to se if just adding a cooler would help much by yanking of the heater air hose to the cabin and turned the heat on fully but venting it under the front bonnet. I think it helped a bit but not as much as it should have.

 

I don't think venting the rear helps that much for engine temp, our engines are water cooled not air cooled

 

Nice to hear others have tested adding electrical pump with success, I think that will also solve the turbo cooling in the same solution.

[Nev]

If you still have the main rear floor pan installed on your car, then I think a slit cut in it and bent down, with some added (welded) plate on the inside would nicely direct air onto the sump.

 

Those silly little plastic venturi flaps on the existing floorpan do little IMO, certainly can't combat the 10s of thousands of watts of heat made by the engine.

 

Edited by Nev, 17 July 2016 - 04:41 PM.

[Mattias]

If you still have the main rear floor pan installed on your car, then I think a slit cut in it and bent down, with some added (welded) plate on the inside would nicely direct air onto the sump.

 

Those silly little plastic venturi flaps on the existing floorpan do little IMO, certainly can't combat the 10s of thousands of watts of heat made by the engine.

 

 

I only enlarged the hole around the flex pipe, the naca ducts seem to work good enough to keep engine bay temperatures ok but I have a custom chimney around turbo and manifold and a turbine blanket as well as completely wrapped exhaust to keep the heat where it helps performance. There are som pictures here if you scroll down: http://www.garaget.org/MattiasGbg/blog

 

I went ahead and ordered a davies craig water pump, I plan to run it with pwm based on rpm and temp and I hope it will flow to the turbo also. If not I will probably use the current aux pump to cool the turbo since the new pump will cool entire system after shut down. 

[Mattias]

Got the pump in my mailbox today so I went ahead and fitted it, just wiring and refilling coolant left to do before initial tests can begin. Hope there is some trackday coming up soon so I can test it properly in full attack.

 

I fitted it inside the front under tray, had to drill a new hole and I cut and reused the hose. So more or less a direct fit.

 

[glitch]

Nice mattias! Keep us posted how it works.

[sam220T]

I'm also seeing the same issue as you described. What I did eventually find out is that I still had some air in the rad slowing up the flow. It's worth running the car up to temp and loosening the rad bleed nipple and allowing lots to flow out as air periodically still vents out. Can take some time. Will watch this tread as still stays hot when on track and back in the pits on tickover stays at 105 unless revved.

[Nev]

I'm also seeing the same issue as you described. What I did eventually find out is that I still had some air in the rad slowing up the flow. It's worth running the car up to temp and loosening the rad bleed nipple and allowing lots to flow out as air periodically still vents out. Can take some time. Will watch this tread as still stays hot when on track and back in the pits on tickover stays at 105 unless revved.

 

If it wont cool when idling, it might possibly be your water pump vanes are knackered.  

[Mattias]

I got everything back together today, I have now fitted two fan modules so I can control both fan and pump speed. It's nice to hear the fan following engine temp instead of trying to impersonate a hoover craft

Another good thing about these modules is that they have built in soft start so there is no big current draw at activation.

 

This is the Volvo fan module that drives the pump:

 

So after getting all air out of the system by moving the car around in a slope while having engine running with no cap on for quite a while it was now time to turn on the pump. First try ended in an unsuspected fail. What happened was that the pump increased the pressure to the bottom big hose of the tank enough to revers flow in the small hose that goes to the thermostat, in other words it pushed all air in the tank back in to circulation     

 

After some thinking I started looking for a one-way valve for fluid and managed to find one for fuel in a shop nearby. Went and picked up the valve and a Pizza, the pizza was good but the valve was even better. I fitted it to the hose and checked that it still let coolant flow like normal. 

I then turned the pump on again and it worked perfectly, after the pressures stabilize it even started flowing in the right direction in the pipe.

It was needed quite high pump speed to push the air out of the turbo coolant hose but it did get some circulation going.

 

I then turned fans full on at idle and engine at +90C and warm summer weather, then I increased pump speed until temp started to drop and settled at 85C. It was very obvious that the fan had very little effect but the pump gave instant respons to engine temp.

 

This is the control tables I ended up with for now:

 

The pump can hardly be heard and the fan at 20-40 duty is nice and silent to.

 

I now tested after shut down cooling, before the engine heat soaked after shut down. You can see this by turning on ignition a few minutes after shutdown, it will rise to +105C quite quickly. I have the possibility to monitor this live and can also hear the fan revving up with the heat. When the pump was activated temps dropped quickly instead of rising    

 

So far good results, won't know if it helps with track temps until next trackday, I tried pushing it a little  today but couldn't get the temps over 90C. But is't impossible to continuously keep it in high load on the street...

 

I still might reroute the oem aux pump to cool the turbo since it's useless where it is now. 

[Nev]

LOL, programmable mapping tables for fan duty cycle - whatever next !

 

Nice solution though.

[sam220T]

I'm also seeing the same issue as you described. What I did eventually find out is that I still had some air in the rad slowing up the flow. It's worth running the car up to temp and loosening the rad bleed nipple and allowing lots to flow out as air periodically still vents out. Can take some time. Will watch this tread as still stays hot when on track and back in the pits on tickover stays at 105 unless revved.

  If it wont cool when idling, it might possibly be your water pump vanes are knackered.  

Hi Nev did think this but haven't changed it yet as only done 11k on an oem pump over 4years but that doesn't mean it isn't shot. Will replace it over the winter.

[GeorgeBC]

I'm also seeing the same issue as you described. What I did eventually find out is that I still had some air in the rad slowing up the flow. It's worth running the car up to temp and loosening the rad bleed nipple and allowing lots to flow out as air periodically still vents out. Can take some time. Will watch this tread as still stays hot when on track and back in the pits on tickover stays at 105 unless revved.

That's nothing to worry about! I wouldn't (and don't) worry about seeing 105 degrees whilst moving. When it's static the engine isn't under nearly as much load.

[Nev]

Also, over the years I have tried different temp sender units and they alone can vary by 7 degrees in my experience. I had one that shows 92 degrees, but when I replaced it with another it showed 85 degrees in the same idling situation.

 

 

[Mattias]

Also, over the years I have tried different temp sender units and they alone can vary by 7 degrees in my experience. I had one that shows 92 degrees, but when I replaced it with another it showed 85 degrees in the same idling situation.

 

 

 

Do you mean different sensors or from sensor to sensor with same part number? 

If using oem GM sensor the values should be within a small tolerance but if putting in just a random sensor the calibration needed might not be the same giving large deviation.

I wouldn't put in any sensor not being oem, eg. the temp sensor is used for a lot of correction maps for fuel, ignition etc.