Converting LS1 heads to reverse flow cooling....
#1
Converting LS1 heads to reverse flow cooling....
I researched some postings but there doesn't seem to be as much info on this as I was hoping. I was hoping if someone could corroborate, elaborate, or discredit the following: (sorry it is long and multiple related postings)
"Guys,
If you look at the coolant flow through an LS series block, it doesn't appear to be very difficult at all to convert. I do
know that it has been done and I will be converting Darton sleeved 427 when it goes back into my SS. FWIW, I always drill
and tap the front and rear coolant bleed holes on any max effort deal I do. I run -6 lines from all four corners of the
heads to the top of the radidator in the stock location. This was suggested to me by Kurt Urban a couple of years ago.
Since that time, I have successfully run 12:1 and above with 93 octane pump gas (with good chambers...) I do this mod on
all nitrous and F/I builds and some NA builds if the owner is up for it.
Getting back to the reverse cooling issue, taking a quick look at the passages in the block and head, if we feed the top
hole on the block first and make the lower hole the return, I see no issues with bleeding the system since the fatory
coolant bleed holes are the highest point in the cooling system (assuming an F-Body setup.) This would be a simple setup
using an external WP with Meziere's AN adapters.
I may not be 100% correct in all my assumptions but I am confident enough that I will be doing this on my own build.
Shane
************************************
I posted this on the 944 Hybrids forum where I'm documenting the build....
"I am only looking to do a mod to the LS water pump and remove the thermostat. On the last V8 swap I did to a 2.8l S-10,
all I did was remove the T-stat and did not change to a larger radiator. All my overheating issues went away. In the
winter, I just reinstalled the T-stat and the heater blew nice n hot and engine never overheated. My theory is the stock
Porsche 944 NA radiator with the T-stat removed would be the best cooling set-up and obviously the easiest. The inner port
hole between the heater ports on the LS1 pump can be blocked off. I have read elswhere that a 1/2" hole is drilled in a
freeze plug and installed. I also see that the 160deg T-stats have 4 small holes drilled in that block-off plate. The only
reason I can see why GM put that small rear block-off plate(spring loaded) was in case the radiator (even heater core) were
to be clogged, the coolant would just recirculate within the block. Looking at the Truck Pump, the inner port hole could be
enirely blocked off with an epoxy'ed freeze plug or the stock T-stat could simply have the upper piston drilled/removed and
the heater would function normally and the water velocity would remain constant and increase with RPMs".
"I removed the back cover from the truck WP to double check the flow route of this pump which should be exactly the same as
all the LS pump routings. From what I see, if that inner port hole is blocked off completely and the T-stat removed the
coolant velocity would be higher going to the top radiator and cool much better"
Looking at the truck pump veins, I don't see why the serpentine/pully can't just be reversed and the same mod done to the
port hole and the water would flow backwards for head to block cooling (reverse).
**************************************
I am new to the LS world a short time ago I was mistakenly under the impression that like the LT engines the LS were
"reverse cooled". Several years ago I ran some Evans Cooling products in a Ford engine with very good results. Since at
that time I wasn't working on any GM stuff I got the full story when talking to the Evans guys.
Anyway when we set out to build our first LSX Block Nitrous engine I made a slight mistake. The engine would use a remote
water pump so we welded AN bungs to the motor plate. The mistake that I made on the initial engine is that I reverse cooled
it by hooking the water lines up backwards. That engine (in that configuration) made the most horsepower that we ever had.
We never had a problem with cooling or steam. This was a drag race engine with little time to heat up.
Since my error was pointed out we have run the lines the proper way. One of the things on my list to check on the dyno is
to do back to back pulls with the lines hooked up both ways.
I should be able to get back to the dyno in about a month and will give it a try at that time.
I had some conversations with a thermal engineer regarding the flow through the engine. The gentleman that I discussed this
is very in the know and currently works in the business. The mapping and probing that is done on water flow models is
unreal. He felt that I got lucky because of the short run times. He felt that the gasket restrictions would need to be
studied to provide the correct amount of coolant for enough time to absorb heat from the heads and block.
Hopefully I can try it again and let you guys know what I find.
Robin
********************************
One problem with the small block Chevy, and now I believe the LS*, is that with both heads flowing into a single port, then
into a hose back to the rad, each head does not flow at the same time. This was something Smokey Yunick investigated. What
happens is that as the coolant heats up and begins to flow, one side(head) will heat up faster ie. hotter, with more
pressure. This greater pressure makes only the hotter side flow because of the greater pressure. Then, as the hotter side
flows, it cools down enough, aqs the non flowing side heats up. There is then a transition of flow. In the end what happens
is an alternating flow, resulting in an inefficient system, and hotter heads, and less compression.
What "Smokey" did was drill holes into the ends of the heads edging towards the exhuast side ( hottest ) . Tapped and
installed piped fittings with hoses and a Y juction to a main rad return hose. This allowed the heads to flow more
effiently(from the front of head) and to flow simultaneously, resulting in more stable head temps.
It would seem that the LS1 heads are ready made for this mod, with the casting plugs in the exact optimum area.
"Smokey Yunick" for those that don't know was yester years David Vizard in the racing/performance genre. He has many
patents including the extended tip spark plugs we all enjoy. When I read something of his, I take it as gospel. Though some
new technology and metalurgy has changed from his days. He tried to get GM to use a reverse flow setup for many years.
Your thoughts are appreciated...
Art
"Guys,
If you look at the coolant flow through an LS series block, it doesn't appear to be very difficult at all to convert. I do
know that it has been done and I will be converting Darton sleeved 427 when it goes back into my SS. FWIW, I always drill
and tap the front and rear coolant bleed holes on any max effort deal I do. I run -6 lines from all four corners of the
heads to the top of the radidator in the stock location. This was suggested to me by Kurt Urban a couple of years ago.
Since that time, I have successfully run 12:1 and above with 93 octane pump gas (with good chambers...) I do this mod on
all nitrous and F/I builds and some NA builds if the owner is up for it.
Getting back to the reverse cooling issue, taking a quick look at the passages in the block and head, if we feed the top
hole on the block first and make the lower hole the return, I see no issues with bleeding the system since the fatory
coolant bleed holes are the highest point in the cooling system (assuming an F-Body setup.) This would be a simple setup
using an external WP with Meziere's AN adapters.
I may not be 100% correct in all my assumptions but I am confident enough that I will be doing this on my own build.
Shane
************************************
I posted this on the 944 Hybrids forum where I'm documenting the build....
"I am only looking to do a mod to the LS water pump and remove the thermostat. On the last V8 swap I did to a 2.8l S-10,
all I did was remove the T-stat and did not change to a larger radiator. All my overheating issues went away. In the
winter, I just reinstalled the T-stat and the heater blew nice n hot and engine never overheated. My theory is the stock
Porsche 944 NA radiator with the T-stat removed would be the best cooling set-up and obviously the easiest. The inner port
hole between the heater ports on the LS1 pump can be blocked off. I have read elswhere that a 1/2" hole is drilled in a
freeze plug and installed. I also see that the 160deg T-stats have 4 small holes drilled in that block-off plate. The only
reason I can see why GM put that small rear block-off plate(spring loaded) was in case the radiator (even heater core) were
to be clogged, the coolant would just recirculate within the block. Looking at the Truck Pump, the inner port hole could be
enirely blocked off with an epoxy'ed freeze plug or the stock T-stat could simply have the upper piston drilled/removed and
the heater would function normally and the water velocity would remain constant and increase with RPMs".
"I removed the back cover from the truck WP to double check the flow route of this pump which should be exactly the same as
all the LS pump routings. From what I see, if that inner port hole is blocked off completely and the T-stat removed the
coolant velocity would be higher going to the top radiator and cool much better"
Looking at the truck pump veins, I don't see why the serpentine/pully can't just be reversed and the same mod done to the
port hole and the water would flow backwards for head to block cooling (reverse).
**************************************
I am new to the LS world a short time ago I was mistakenly under the impression that like the LT engines the LS were
"reverse cooled". Several years ago I ran some Evans Cooling products in a Ford engine with very good results. Since at
that time I wasn't working on any GM stuff I got the full story when talking to the Evans guys.
Anyway when we set out to build our first LSX Block Nitrous engine I made a slight mistake. The engine would use a remote
water pump so we welded AN bungs to the motor plate. The mistake that I made on the initial engine is that I reverse cooled
it by hooking the water lines up backwards. That engine (in that configuration) made the most horsepower that we ever had.
We never had a problem with cooling or steam. This was a drag race engine with little time to heat up.
Since my error was pointed out we have run the lines the proper way. One of the things on my list to check on the dyno is
to do back to back pulls with the lines hooked up both ways.
I should be able to get back to the dyno in about a month and will give it a try at that time.
I had some conversations with a thermal engineer regarding the flow through the engine. The gentleman that I discussed this
is very in the know and currently works in the business. The mapping and probing that is done on water flow models is
unreal. He felt that I got lucky because of the short run times. He felt that the gasket restrictions would need to be
studied to provide the correct amount of coolant for enough time to absorb heat from the heads and block.
Hopefully I can try it again and let you guys know what I find.
Robin
********************************
One problem with the small block Chevy, and now I believe the LS*, is that with both heads flowing into a single port, then
into a hose back to the rad, each head does not flow at the same time. This was something Smokey Yunick investigated. What
happens is that as the coolant heats up and begins to flow, one side(head) will heat up faster ie. hotter, with more
pressure. This greater pressure makes only the hotter side flow because of the greater pressure. Then, as the hotter side
flows, it cools down enough, aqs the non flowing side heats up. There is then a transition of flow. In the end what happens
is an alternating flow, resulting in an inefficient system, and hotter heads, and less compression.
What "Smokey" did was drill holes into the ends of the heads edging towards the exhuast side ( hottest ) . Tapped and
installed piped fittings with hoses and a Y juction to a main rad return hose. This allowed the heads to flow more
effiently(from the front of head) and to flow simultaneously, resulting in more stable head temps.
It would seem that the LS1 heads are ready made for this mod, with the casting plugs in the exact optimum area.
"Smokey Yunick" for those that don't know was yester years David Vizard in the racing/performance genre. He has many
patents including the extended tip spark plugs we all enjoy. When I read something of his, I take it as gospel. Though some
new technology and metalurgy has changed from his days. He tried to get GM to use a reverse flow setup for many years.
Your thoughts are appreciated...
Art
#6
TECH Fanatic
iTrader: (14)
Join Date: Mar 2007
Location: Phoenix/Tempe, AZ
Posts: 1,969
Likes: 0
Received 0 Likes
on
0 Posts
wish i could provide some input, but i havent really got any to provide. I seem to remember that being one of the 'big deals' with lt1 vs sbc technology, so i could see the logic behind looking further into it.
Trending Topics
The following users liked this post:
Jay Fisher (08-03-2019)
#9
TECH Veteran
iTrader: (1)
Since they already did it on the LT1 with no patent or legal problems, exactly what makes you think they have those problems with the LS1? I am under the impression that reverse flowing the coolant showed no legitimate gains, but does cause air pocket and hot spot issues.
#10
TECH Enthusiast
iTrader: (9)
Since they already did it on the LT1 with no patent or legal problems, exactly what makes you think they have those problems with the LS1? I am under the impression that reverse flowing the coolant showed no legitimate gains, but does cause air pocket and hot spot issues.
#12
TECH Enthusiast
iTrader: (9)
We know based on the comments of members on this site who have run reverse flow that there are no apparent negative effects and that there are gains to be had so we can assume that GM did not scrap the idea because of performance problems.
So what could have been the reason? Any guesses?
On a side note your valve cover example does not apply because after 20 years patented technology becomes public domain and the valve cover patent (if there ever were such a thing) would have been around for longer than that.
#13
TECH Apprentice
iTrader: (4)
What makes you think you can steal a design and get away with it especially on such a large scale? Don't you think it would have been easier to contract with Evans to use their design, pay them their royalties, and be done?
read this...http://money.cnn.com/magazines/fsb/f...3431/index.htm
The following users liked this post:
Jay Fisher (08-03-2019)
#14
Ceramic coating the pistons, chambers, valves, and exhaust ports helps as well. Water injection really helps, who says its just for forced induction? I am sure 13:1 or even 14:1 could be doable with water injection.
#17
On The Tree
iTrader: (3)
Ex. my company sells a device that uses an internal spring to support the weight of the device as it moves. Our competitors must use an external spring to support theirs due to a patent we have.
#20
TECH Fanatic
iTrader: (4)
Join Date: May 2002
Location: Fairview Heights Illinois
Posts: 1,851
Likes: 0
Received 0 Likes
on
0 Posts
I would not trust simply reversing the direction of pulley rotation to achieve functional reverse-flow.
I have run without a tstat, and with the bypass port blocked, and cooling capacity is indeed increased significantly.
The factory cooling system design of the LS1 is initially at a disadvantage due to the input and output ports on the block being right next to each other. At least some significant percentage of coolant flow will be taking the path of least resistance. The over-sized bypass port only adds to this disadvantage.
I think if I were to reverse-flow an LS1 I would modify the block more-so than the heads. I think the coolant ports are poorly placed for cooling in either direction.
I have run without a tstat, and with the bypass port blocked, and cooling capacity is indeed increased significantly.
The factory cooling system design of the LS1 is initially at a disadvantage due to the input and output ports on the block being right next to each other. At least some significant percentage of coolant flow will be taking the path of least resistance. The over-sized bypass port only adds to this disadvantage.
I think if I were to reverse-flow an LS1 I would modify the block more-so than the heads. I think the coolant ports are poorly placed for cooling in either direction.