12H-T performance Build (1 Viewer)

[troopie]

Looking at the pictures of the head, i sure regret not spending more time on my own. Maybe when i blow another headgasket...

I'm kindoff curious about the Supra compressor wheel in the 12HT ct26. I don't want to hijack a good thread, but if you got more info, i love to hear more about it. PM is good also.

Anyway. I'll be following this thread with great interest. It's good to see someone who knows there stuff working on a 12HT, instead of me, just tinkering and experimenting without proper backround. Although it has really worked out for me.

And i also looked at the 1HD-T conrod. Looks bent to me.

Later,
Mark

 

[gbentink]

The 1HDT rod is bent.........I suggested to a good friend running the same setup I had in my HDJ81V to decoke his vege ol carboned up 1HDT with a water injection system. Problem was he installed a passive system (pre turbo, also not a good idea), didnt put in a one way check valve and mounted the water tank higher than the spray point..........nedless to say, one day he satrted it (you know how these DI engines start righ; full fuel and instant ignition) and proceeded to bend 3 rods. Anyway, I have the block and we still have the cylinder head, crank, which is fine, etc. One thing I have been meaning to do is compare the 1HDT head design to the 12HT, ports and swirl that is. For interest sake.

I have measured up the 12HT. This engine can be bored and stroked significantly. I have done ALL the homework (sourced custom fibre headgasket etc etc) and might use my GT3782VA on that project if it happens, one day.

But.......after getting excited about it and the big numbers, I reran my calcs on my compound setup for the std 4 litre. Guess what, even if my VE falls to 65% by 3600rpm, *in theory* my setup will pump 650cfm. That is about 320hp worth of air based on clean diesel numbers, not the smoke figures you often see in the USA. I am looking for 250, which sits in the centre of the compressor maps for the two turbos peak efficiency islands. The unknown are the turbines..........

The end result is completely dependent on the combined efficiency of the two turbos.....never done this before, so we will see!!!!

 

[gbentink]

Hi Dougal,

A fellow physicist; I also studies thermodynamics at university so it is good to see theory and practice coming together!

Below is an explanation for others not versed in heat engine theory.

When I said lower combustion temps to increase efficiency, this is the avarage and based on there being significantly excess air available during combustion as compared to std engine. Theory being excess air draws latent heat from the combustion flame to:

* increase peak cylinder pressure
* decrease average temperature
* decrease heat transfer into combustion chamber walls

The most efficient heat engine (in the world by design) is the Stirling engine. It is the most efficient because it works on pressure change based on the absolute minimum of temperature gradient bewteen the "heater" and the "heated"

Thus it is pressure that performs the work, not heat; heat is a means to raise the pressure, however the pressure is dependant on joules of heat available and the amount of gaseous matter present to be heated. The more moles, the lower the average temp, but also the less energy wasted into the surroundings ie: combustion chamber wall.

This is the #1 advantage of diesel over gasoline

It is also why if friction and VE loss removed from the equation, a gasoline engines efficiency increases with rpm beacuse in this case there is less TIME for the heat to dissipate into the combustion chamber walls. In practice of course, friction is significant and VE is not usually the limiting factor; its stopping everything from flying apart that is the problem!. It is also true for diesel, however it is flame speed and injector speed that presents additional problems.

On the basis of the above theory, a diesel engine should have peak efficiency at a lower rpm than a gasoline and that is what we find; in say a 2 litre 4 cylinder, around 1700rpm in diesel (and 2400rpm in gasoline is the peak efficiency rpms. There are many factors, but these figures are about right.

Dougal; am happy to hear your comments on the above. I may have not explained sometghing clearly and do not want others whom may not be versed in the subject to take is typical "internet information" garbage in, garbage out.........

 

[Dougal]

But.......after getting excited about it and the big numbers, I reran my calcs on my compound setup for the std 4 litre. Guess what, even if my VE falls to 65% by 3600rpm, *in theory* my setup will pump 650cfm. That is about 320hp worth of air based on clean diesel numbers, not the smoke figures you often see in the USA. I am looking for 250, which sits in the centre of the compressor maps for the two turbos peak efficiency islands. The unknown are the turbines..........

The end result is completely dependent on the combined efficiency of the two turbos.....never done this before, so we will see!!!!

I reckon about 68psi boost necessary for 650CFM through a 4 litre engine at 3600rpm with 0.65 VE.
That's using 60% efficient compressor and 60% effective intercooler at sealevel 20C ambient etc etc.

And 120cc/1000 strokes feeding it gives 20:1 AF ratio.

So yeah, your maths looks good.

I got interrupted typing that, so to update it in response to your post above.
I do agree with your reasoning on running excess air, it just needs to be balanced with the power required to pump that air. Keeping in mind that your turbos are more efficient the higher the intake temp is, reducing the temp you feed them means you get higher backpressure to deliver the same boost.

The stirling engine is an interesting one and those who have hard numbers on modern machines keep their numbers quite secret. Whispergen etc. The best numbers I have heard are high 30's percent which is beaten by production diesels.
One of the mechanical engineers who did his postgrad essentially working for whispergen was interviewed on the radio here the other day (forgot his name, but I have met him). He mentioned the actual behaviour of the stirling cycle follows a thermodynamic cycle which does not yet have a name. He tried to name it the "Raine Cycle" after his supervising professor but John Raine wasn't having it.

The VW 1.9tdi at 197 g/kwh comes in at 40% peak efficiency.

 

[gbentink]

I plan to use a MAF with known voltage vs CFM to check if by increasing the boost I actually increase the cfm... Of course, I dont need actual conversions to check for increase, but would like to know if the changes reflect in flow etc. I would also like to play with the balance between the two turbos to see what gives the most cfm for a given boost pressure (see if backpressure changes). I also will measure boost and ex backpressure between the turbos, so 4 gauges for thsi - 60psi gauges!!

For my calcs, I use 85% efficient intercooler and have not considered turbo efficiency (in this instance). The reason for the intercooler value is that I have found my temps come close to ambient with less than 0.5psi backpressure. As for turbo efficiency, this is an interesting one. I assume you refer to pumping loses to run the turbine rather than applying the 60% on the compressor side of the equation??

Typically the exhaust flows some 2.65 times more mass than the inlet, so turbine eficiency * compressor efficiency * mass compressed should give an idea on the drive requirement.... The higher the requirement, the higher the effective VE loss.....

I am fairly confident I cannot (or is it should not) boost more than 50psi with my system. I expect that I would shear off the GT25 compressor side shaft (5mm....). I expect basicall approx PR of 2 on the smaller compressor and 2.2 on the larger compressor is the maximum. I also have to consider leaks where the plenum seals against the head... might need some thickening there and surely I shouldnt use the stock manifold gasket.

I noted on the 4BT site that you are also doing a compound setup for economy?? This is actually my goal too; I have much bigger turbos in my shed, but am using a smaller turbo. In fact, 680cfm is the maximum the big turbo can pump and since it is a hybrid, I am interested to see if it can even flow that!!

By the way, reagrding the Stirling engine, I should have said "theoretical efficiency". In practice, the engines are very large and friction is a huge factor. I cant remember the maximum efficiency, but it was either 50 or 60% therabouts.

 

[gbentink]

I have the paper for the 1.9Tdi VW, that is actually the engine I usually refer to - very efficient! 40% is sensational.

 

[Tim-HJ61]

Keeping in mind that your turbos are more efficient the higher the intake temp is, reducing the temp you feed them means you get higher backpressure to deliver the same boost.

Hi Dougal, and GB,

So pleased GB has someone that understands what he is talking about

I live close to him and have seen his work on the head. Comparing original to now, it would seem to be a great improvement having all the lumps and bumps removed.

Dougal, just picking up on your comment about intake temps - so how come intercooling is used extensively and we drool over a cold intake manifold as it shows the intercooling is functioning effectively?

I understand the idea of the intercooler is to increase the oxygen content of the air; so how do we then increase the air temp to increase the efficiencies as you talk about. Or is it a matter of intercooler after the turbo so the air charge is cooled into the inlet manifold and hot air to feed the turbo itself to achieve the efficiency you are talking about?

Tim

 

[Dougal]

I plan to use a MAF with known voltage vs CFM to check if by increasing the boost I actually increase the cfm... Of course, I dont need actual conversions to check for increase, but would like to know if the changes reflect in flow etc. I would also like to play with the balance between the two turbos to see what gives the most cfm for a given boost pressure (see if backpressure changes). I also will measure boost and ex backpressure between the turbos, so 4 gauges for thsi - 60psi gauges!!

Hope you've got a nice rack for all those gauges. With a single turbo I find backpressure at worst case is double the boost, with compounds I think you'll have less than double in worst case. But it's just a guess right now.

For my calcs, I use 85% efficient intercooler and have not considered turbo efficiency (in this instance). The reason for the intercooler value is that I have found my temps come close to ambient with less than 0.5psi backpressure. As for turbo efficiency, this is an interesting one. I assume you refer to pumping loses to run the turbine rather than applying the 60% on the compressor side of the equation??

I think 85% is too high for an intercooler. There is a very good thread here by Matt McInnes building a laminova intercooler which has his inlet/outlet temps and effectiveness figures. I think somewhere around 60%.
I once made my own air/water cooler using 2.5m of 3/8" copper pipe. I calculated the efficiency at 10%, but it actually did 15% in testing.

For turbo efficiency I mean compressor efficiency. The less efficient the compressor the more the air is heated as it's compressed. 77% is about the highest I've seen, 65% is pretty common and 60% means you're on the edge of the map somewhere.
I've made temp measurements to check, my T25 does about 65% at 15psi boost, my old turbo with a damaged compressor wheel was running at 50%.

Typically the exhaust flows some 2.65 times more mass than the inlet, so turbine eficiency * compressor efficiency * mass compressed should give an idea on the drive requirement.... The higher the requirement, the higher the effective VE loss.....

Hang on, are you mixing mass with something else?
The exhaust mass flow is the inlet mass flow plus the fuel mass flow.
If you're running 20:1 air:fuel ratio then your exhaust has 21/20 times more mass flow (~5% more).

I am fairly confident I cannot (or is it should not) boost more than 50psi with my system. I expect that I would shear off the GT25 compressor side shaft (5mm....). I expect basicall approx PR of 2 on the smaller compressor and 2.2 on the larger compressor is the maximum. I also have to consider leaks where the plenum seals against the head... might need some thickening there and surely I shouldnt use the stock manifold gasket.

I think with compounds it's not the pressure ratio that kills turbos but the density. A PR of 2 is easy for a modern turbo, but if it's inlet air has twice the density of normal then it's got twice the torque on it's shaft.

I noted on the 4BT site that you are also doing a compound setup for economy?? This is actually my goal too; I have much bigger turbos in my shed, but am using a smaller turbo. In fact, 680cfm is the maximum the big turbo can pump and since it is a hybrid, I am interested to see if it can even flow that!!

That project is going a whole lot slower than I like. It alternates between times when I have no time to spend on it and times when my toy fund is dry so I can't buy the pieces I need.
I still need to do some internal machining so I can clock the variable vane turbo and install it. I bought a mill to do it, then stripped the mill to convert it to CNC.
I've also found a whole lot of other things on the way, for example my flywheel isn't heavy enough for the torque I'm currently producing below 1500rpm, so there's another project before I'm ready.

 

[Dougal]

Dougal, just picking up on your comment about intake temps - so how come intercooling is used extensively and we drool over a cold intake manifold as it shows the intercooling is functioning effectively?

I understand the idea of the intercooler is to increase the oxygen content of the air; so how do we then increase the air temp to increase the efficiencies as you talk about. Or is it a matter of intercooler after the turbo so the air charge is cooled into the inlet manifold and hot air to feed the turbo itself to achieve the efficiency you are talking about?

Tim

Sorry that wasn't very clear.
I was talking about the exhaust turbine, the hotter exhaust your engine spits into it, the more efficiently it runs.

You're correct about the intercoolers on the comrpessor side.

 

[Tim-HJ61]

Sorry that wasn't very clear.
I was talking about the exhaust turbine, the hotter exhaust your engine spits into it, the more efficiently it runs.

You're correct about the intercoolers on the compressor side.

Ta, thanks for clarifying.

Regarding the gauges Graeme could use. I've removed the altimeter and temp gauges in the overhead console and installed 4 VDO gauges instead. Bit of a squeeze but looks okay. As I run vege, one is Vege oil fuel temp, fuel pressure at the IP inlet, turbo pressure tweaked to 12psi, and EGT measured pre turbo. FYI this runs to 500°C pretty easily under full load and I back off at 600°C. With the bigger wheels 100kmh is about 2100rpm and 110kmh - 2500rpm.

Tim

 

[gbentink]

Hi Dougal,

I mixed up terms a bit, my bad. For exhaust flow I meant volumetric flow, not mass flow. The turbine would work better at higher temps due to the temp drop across the turbine, the very reason turbos are a more efficient option than crank driven compressors.

Thanks for clarifying the temps being on the turbine side; I meant to ask you what you meant by that.

I have 4 laminova cores awaiting a housing. I have designed the housing, just need to build it. I have never measured pre/post intercooler temps. I used one of the eBay 600*300*75mm cores in my 80 series and when running 21psi boost for extended periods, my test was that you couldnt touch the turbo outlet but post interccoler was mildly warm. I do NOT want to go water to air, but for my boost pressure, I have little choice but to shorten the boosted air path and reduce the number of joins/clamps. My Air to air leaked over 22psi.

I *really* wanted to reduce my EGT's in the higher revs, and my smoke, but I was unable to accomplish either with the setup I had.

As for turbine back pressure, in my 1HD-T, at 18psi and 1600rpm, it was 1:1 ratio. At 2200rpm and 25psi it was 1:1.3 ratio (25 psi - 32psi). At 3000rpm, it was like 1:1.5. Now I thought these were terrible numbers - my EGTS went high and I was getting ~ 114kw @ wheels compared to std 77kw @ wheels.

The 12H-T head is waiting to be sent to a local head shop, today I need to determine which place to send it.


Tim; Great to heer from you. I hope to use 3.7 diffs and 35" tyres. Final drive in our autos is 0.718

33" tyres, 0.718 and 3.7 @ 100km/h = 1690rpm
35" tyres, 0.718 and 3.7 @ 100km/h = 1580rpm
31" tyres, 0.718 and 4.1 @ 100km/h = 2000rpm
33" tyres, 0.718 and 4.1 @ 100km/h = 1870rpm
35" tyres, 0.718 and 4.1 @ 100km/h = 1750rpm


I just bought an 80th Anniversary 80 series by the way (Petrol) for conversion to diesel. have to catch up some time! By the way, my HJ61 is starting to look pretty

 

[gbentink]

Compressor Maps

OK Dougla,

Time to get specs for some newer compressors

See attached. I hope you can see why I chose to use these.

I would have liked to use the GT2252 since that would work even better for me, but I have what I have and there are so many unknowns.....that I am not in a position to make a drastic change and know that I am making an improvement!

Notice the that the GT25 compressor has an 80% area and that the 20G can do a 2.3 ratio @ 70% and 650cfm......... Clearly the larger trim are more efficienct - I postulate that parasitic losses are less......

Personally, I think these could be a great combination!

What are your thoughts on intercooling between the two turbos? Since I am going to go water to air anyway for the final stage, it wont be too much bother. Initially though, I will probably opt not to do it, but I plan to do it or even run the lower pressure through an ait to air (so much plumbing though, better neat and small)

 

[Dougal]

OK Dougla,

Time to get specs for some newer compressors

See attached. I hope you can see why I chose to use these.

I would have liked to use the GT2252 since that would work even better for me, but I have what I have and there are so many unknowns.....that I am not in a position to make a drastic change and know that I am making an improvement!

Notice the that the GT25 compressor has an 80% area and that the 20G can do a 2.3 ratio @ 70% and 650cfm......... Clearly the larger trim are more efficienct - I postulate that parasitic losses are less......

Personally, I think these could be a great combination!

What are your thoughts on intercooling between the two turbos? Since I am going to go water to air anyway for the final stage, it wont be too much bother. Initially though, I will probably opt not to do it, but I plan to do it or even run the lower pressure through an ait to air (so much plumbing though, better neat and small)

So to push 650 cfm through your motor with a VE of 65% you need a total of 68psi, but increasing the compressor efficiency to 65% (still staying conservative incase the turbo they tested was a better casting than the one they sold you) and the VE to 70% you drop down to 58psi boost.

It's really easy to split the pressure ratio in half with the maths, but in reality I don't know how accurate this will be until the system is actually built.
Your total pressure ratio is 4.94, split evenly it becomes 2.22 for each turbo.
But could the split be 3:1 on one turbo and 1.65 on the other?

I've just wasted an hour setting up a series compressor spreadsheet including intermediate cooling.

Leaving everything else the same, putting a 60% effective intercooler between the stages (and allowing 7kPa pressure drop) increases the density ratio from 3.92 to 4.7.
That's 20% more air all for the price of a plumbing nightmare.


I'll work some more on this later.

 

[circle456]

gbentink, I gotta say - It has not at all occurred to me to soup up my 12H-T. "1HD-T oil squirters" - Was this a direct bolt up? I doubt that it's necessary the way I drive, but it might be some added insurance.

My 12H-T is in a rust free (former) California 1986 FJ 60. The 5 speed went in along with it. At 1500 RPM in 4th the sound is sweet on the Northern Nevada desert. You will be pushing or exceeding the limits of 12H-T it seems.

How is availability of spare parts for the 12H-T where you are? You will keep this thread posted as you progress I would think. Good Luck.

 

[gbentink]

circle456, no worries for parts here - I am in Perth Western Australia and have 2 x 12HT motors, complete. Rebuild kits etc are really inexpensive.

Now if I throw a ord, then it's a different ball game..........

 

[gbentink]

Dougal,

An hour well spent I say!!

I will also do the spreadhseet to allow for efficiency losses reducing air density (or increasing PR, depending on how I set up the table....

I have no doubt it will be beneficial, it is simply that for the initial "up and running" installation, I was planning to intercool just the final stage. There is no other spreadsheets around that I have seen to do this.

You should see my spreadsheet - I just keep adding the tabs...........

Providing the intercooling is excellent, the compressor match is incredible!!

I plan to control the PR of the second turbo. How you say? By back cutting the turbine of the second until I keep it where I want it.

If I may say so, I have developed an awesome way of controling the boost in a completely mechanical fashion that; all going well, will start boost from 25psi and ramp to 45psi through the revs. All I have to now is the simple task of making it work in practice.....hmmm....

 

[Dougal]

I plan to control the PR of the second turbo. How you say? By back cutting the turbine of the second until I keep it where I want it.

If I may say so, I have developed an awesome way of controling the boost in a completely mechanical fashion that; all going well, will start boost from 25psi and ramp to 45psi through the revs. All I have to now is the simple task of making it work in practice.....hmmm....

Does that large turbo come with a choice of different exhaust housings? That or an external wastegate would be far more preferable than backcutting the turbine. I think backcutting will cost you both in turbine efficiency (so more backpressure) and spool will arrive later.

I came across a photo of what appeared to be a caterpiller with factory compound turbos. The wastegate on the small turbo was actuated by the pressure in the large. Such a simple and elegant solution but not necessarily the outcome we're looking for.

Why do you want to ramp the boost? I would rather have the max boost as early as I can use it, of course "as early as I can use it" brings me back to needing a heavier flywheel.

 

[gbentink]

Re: Turbine back cutting and reduction in Turbine efficiency...

My question is "does it?"

I have read posts checking this exact question. I wonder if there is a confusion between "reduction in efficiency" and "narrower band of operation"

I have inspected many of the newer GT series turbines. They all have 2 things in common - large trim and back cut wheels!!!! I have inspected many older turbines, and they have 2 things in common; small trim and non back cut wheels.

I have some spare turbines and I have different housing sizes, so I can try a variety of approaches.

Many comment on the spool up and performance of the GT Ball Bearing turbos and comment that it is due to the bearings solely. The majority of the improvements have come from turbine efficiency, reduction in rotating mass-inertia and compressor efficiency.

So, to revisit my initial question, I am of the opinion that back cutting the turbine will increase the overall compressor efficiency of the whole system by reducing back pressure and restricting full boost from the big turbo until at or after 2000 rpm. If possible, I will not use a wastegate; but it has an internal in case it is required. The compound setup will find its natural balance and I will do airflow, EGT and power tests to find the best setup.

Now when you say big turbo, you are probably referring to the big GT3782VA turbo in the picture which is a VNT - that is for another project...... I will be using a lowly old CT26 with 20G compressor, supra exhaust housing and back cut turbine. So there you have my secret setup.

I tried my largest exhaust housing by itself on the 1HDT, no boost before 2000rpm! So, back cut, I hope to push that out to 2500-3000rpm so that in a compound arrangement, it STARTS to build boost at 1500-1750 ramping to full boost by 2000-2500rpm.

I have theorised (guessed) that the VE will be around 85% at up to 1500rpm and decrease to 65% by 3500rpm. This is why I want the boost to ramp up. I expect 25psi by 1400rpm, 35 by 2000rpm and 45psi by 2800-3500rpm.

If anyone has any literature (such as from garrett) on the matter of back cutting turbines, I would very much like to read it; as I said, all the performance turbos have it nowdays. I am always happy to be worn - except when it costs me money....

 

[gbentink]

Dougal; I started a new spreadsheet, it has been a while since I studied physics and I am trying to work out energy required to compress an ideal gas, apply the efficiency factor of the turbine then put into PV=nRT to determine mass flow...... drawing a bit of a blank Of course, since air is predominantly nitrogen and the behaviour data is available, we don't have to use the ideal gas equation and can be a bit more accurate.

work for compressing a gas is proportional to the integral of the change in pressure/starting pressure.

R = 8.31 J/nK

So we should be able to get the joules of energy change right out of PV=nRT.........

 

[Dougal]

Dougal; I started a new spreadsheet, it has been a while since I studied physics and I am trying to work out energy required to compress an ideal gas, apply the efficiency factor of the turbine then put into PV=nRT to determine mass flow...... drawing a bit of a blank Of course, since air is predominantly nitrogen and the behaviour data is available, we don't have to use the ideal gas equation and can be a bit more accurate.

work for compressing a gas is proportional to the integral of the change in pressure/starting pressure.

R = 8.31 J/nK

So we should be able to get the joules of energy change right out of PV=nRT.........

It's as simple as power = pressure * volume flowrate
Where pressure is Pa (N/m^2) and volume flowrate is m^3/second at the intake.
Divide that by your efficiency and you get the actual pumping power.

Regarding backcutting. IMO there's a large divide between garrett designing and manufacturing a turbine wheel that looks like that and someone having a go themselves.
Keep in mind the second turbo is going to be receiving colder exhaust until the small turbo wastegates.

I designed a simple (simple to manufacture) hydroelectric turbine once. In a group of three we put probably 120+ hours into just the maths to work out the necessary entry and exit angles for the blades. This is in simple incompressible flow.
To put it simply, I have a huge amount of respect for the brain power that has gone into design and development of turbine wheels and I'm very reluctant to screw with them.

When I was saying big and little turbo, I use those terms as they're less confusing than primary and secondary. I didn't know which turbo you intended to use. But did you just say you are going to use a CT26 for 650cfm?