Twin ( parallel) turbos--redux

Michael,

I know exactly what your saying about the $100.00 at a time over time is a lot less painful. If you don't need the truck it would be fine but it would not be unreasonable to expect the truck will be out of service for a year while you build and troubleshoot your new system.

Jim

jauguston said:

Michael,

I know exactly what your saying about the $100.00 at a time over time is a lot less painful. If you don't need the truck it would be fine but it would not be unreasonable to expect the truck will be out of service for a year while you build and troubleshoot your new system.

Jim

Click to expand...

EXACTLY the reason for this thread. IF someone has already taken that year and $$$ to do it, maybe I could benefit from that. Always looking for a little help....

The Ford 2.3L had 2 different turbo's the Garret T3 (with either a .48 or .63 A/R hot side) and an IHI. The IHI are smaller than the T3's and can be had stupid cheap., but are not upgradeable. The T3's can still be found fairly cheap ad have TONS of options for future upgrades.

The turbos I'm thinking of are the .63 a/r TO3 from a 5 speed. The .48 from an auto might just do the trick for a pair of twins for an IDI. Again, I need to re-read Maximum Boost and refresh my memory on how to read turbo maps and such.

Ok, dug out my book...
The Calc for airflow rate is as follows, CIDxRPMx0.5xEV/1728
The .5 is due to the engine being a 4 stroke, and filling the cyls on one half of the revolutions, and EV is volumetric efficiency.
So, for a 6.9, we have 420 CIDx3750rpm(mine turns this)x0.5x.85(probably a bit low, typical of a gasoline engine though) divided by 1728(converts CI/min to CFM.)
So, at 3750 RPM, a N/A 6.9 moves 387.37CFM, this is our basic cfm.
Now, Lets figure out our Pressure ratio for the turbo, so we can figure out our required flow rate under boost.
Pressure ratio is total absolute pressure created by the turbo divided by atmospheric pressure. For the purposes of these calculations, the design goal will be 20 psi @ 3750.
So, 14.7+20(to get absolute pressure, we add atmospheric pressure to boost pressure) 34.7/14.7 = 2.36 pressure ratio, which means we are moving 136% more air through the engine than it could move on it's own.

So, our airflow rate under boost = pressure ratio x basic engine cfm, so
2.36 x 387.37, which = 914 CFM. Some compressor maps are in lbs of air/min. To get this, (lbs/min) we multiply Atmospheric pressure x cfm x 29/(10.73 x Temperature(in Rankine, which is F+460.)
So, 14.7x914x29= 389638.20
10.73 x (75+460) = 5740.55
389638.20/5740.55 = 67.87 lbs/min.

This may just like like a bunch of useless math, but, it is all very important when sizing the compressor, it makes it possible for us to read compressor maps, and have a clue as to what we're looking for.
What we want to see for a non intercooled setup, is a compressor that will flow 914 cfm/68 #m, at a 2.36 pressure ratio, and give us at least a 60% compressor efficency.
Looking at the following charts, we can see by where the lines intersect that a T04 62-1 compressor will just barely do the job, and the efficiency will be below 65%


But, a T66 will do the job nicely, and give us around 70% compressor efficency.


All that math, and we just have a compressor picked out. And, I'm half asleep, so, my math could be out....
For a twin turbo engine, we would cut our required CFM rating in half, but, keep the same boost pressure requirement, so 914 CFM @ 20 psi becomes 457 CFM/34#m. Our pressure ratio hasn't changed, but, our airflow has, so, lets look at some compressor maps.

The V trim compressor looks like a VERY good match


A H3 would also do it, but, the efficency isn't as good, so, the V trim would be a better choice, as it will create less heat.
For the exhaust side, I don't have nearly the information on hand to calculate it, nor the formulas, and of course, all the sizing charts I can find are based on gasoline engines which turn considerably more RPM than what we are dealing with.
Most charts actually have my too large, late spooling non-gated Banks setup down as having a small exhaust housing for the displacement of the engine.
So, just to get a rough idea as to what would be close without calling up a turbo supplier(I doubt any are open at this time) and quizzing up an expert, a turbine sized for an engine half the displacement, that turns twice the rpm, should be roughly about the size we need, so, for a single turbo, a T4, O trim turbine, with a .58 housing should be about the right size, which is about what I had figured before doing any math, and shows just how oversized the 1.0 a/r on my existing Banks unit is, it was clearly designed to have a high boost threshold(lowest rpm at which useable boost is created) and spool well into the engine's rpm range so as to limit peak boost, without use of a wastegate.
For twins, a T3 turbine with a .48 housing.
Again, this bit is only a rough guess.
It looks like those .48 T3's off of a 2.3 may be useable based on the exhaust side, but, I'll need some info on their compressor to figure out what kind of flow and boost they are capable of.
I know on a 2.3, from the factory, they put out 15 psi(85.5/86 SVO) and a 2.3 turns about 6200 rpm. So, back to out inital formulas. 140x6200x.5x.85/1728 = 213cfm
15+14.7/14.7 = 2.02 pressure ratio
so, 2.02x213= 430 CFM/32#/min. Knowing this is a T3, and assuming it's sized so it's reasonably efficent, it's probably a super 60 compressor.
Now, lets see how it would work on our twinned 6.9 at 15 psi, 2.02x387 = 781.74 CFM/2(remember, twins) 390.87 CFM/29#/min per turbo



Looks like for a 15 psi design goal, a T3, super 60 trim compressor would be a very good choice. Like I said, I need info on the compressor side of that TC T3, and, I need to find better information on sizing the exhaust side, which, is really the hard part.

Hopefully I did more to help you guys understand turbo sizing than confuse you....
Boost pressure is a direct product of how much CFM your turbo(s) flow, the more air they move for a given drive pressure, the more boost they will make, provided they flow more than the engine would by itself.

I don't think doing your calculations based upon a 3750 max RPM makes sense. Unless something has been done to the governor a 6.9 should only turn 3300.

Jim

Only 3300? Interesting... I've never driven one that didn't turn at least 3500, if the loading is right, mine will creep to 4000, under normal circumstances, it defuels at 3750. And, I'm pretty sure the rated HP is at 3400.
But, like I said, I did those calculations with my engine, which does turn 3750, and 20 psi in mind. That was not intended to be a one size fits all result.
Using 3300 as the rpm limit, we have a result of 340 basic cfm, and, lets use a design goal of 15 psi for this one, so, a pressure ratio of 2.02x340cfm = 686 cfm/51#/min
Looking at some maps, The TO4B 62-1 compressor will work great


as will a T64:


or a T66:


The higher the efficency the better, and, the larger compressors aren't really necessary, but, they will allow for an increase in pressure later on.
We also have to take into account turbo surge limits, which is relatively simple to do.
First, we assume that the desired pressure ratio is reached at 50% of max rpm, and plot it on the compressor map. So, 686/2= 343 CFM 51/2=25.5#/min @ a 2.02 PR
Now, we draw a line from this point to the point where PR = 1 and CFM = 20% of maximum. so. .20X686= 137.2cfm/10.23/min
The surge limit may not always be marked, but, it is safe to assume that it is the left most line in the turbo map.


The T04B 62-1 is a good choice for this, with the efficency being above 65% the whole way up.


The T64 is not useable, the line we just drew MUST lie to the right of the surge line on the turbo map.


The T66 is useable, and will give us slightly better efficency than the TO4B 62-1, but either would be a decent selection.
Notice these are the same compressors I chose for 20 psi/3750 rpm.
The line in red is for my 20 psi/3570 rpm calcs, in black is for 3300/15 psi. The T34 would be useable for the 20psi/3750 application, but, still would not be ideal.

Thanks for the textbook type of study. I feel like I am in engineering school.

Am I right in assuming that if I want less boost than your example, and the max rpm is a little lower, that the T3 super 60, would still be a good choice?

To be clear, what cars used that unit? I think you refer to the 2.3 liter; I am thinking turbo T bird. What other application used that turbo?

My local U Pull It yards charge $50 for a turbo, regardless of application ( or condition)

Yes, a T3 super 60 with a .48 a/r turbine would be a good choice.
Staying with a 15 psi design goal, lets plot our flow line:


We're a little on the big side, but, definitely still usable.

For 12 psi, well, lets work it out
14.7+12/14.7 = 1.81pr
Our 3300 rpm engine moves 340 CFM basic, so, 170 per turbo.
170x1.81= 308CFM/22.87 #/min.
So, @ 50%, we are flowing 154 CFM/11.435#/min
20 % of our maximum CFM = 61.6 CFM/4.6#/min.


Again, a bit large, but definatley useable.
I still haven't been able to find out for sure which compressor the 2.3T cars do run(Turbo Coupe, Mustangs, Merkurs) It may be smaller and better suited.
I'll also have to do some research on the Mopar 2.2/2.5 turbos, I know they also ran T03's.

Dang 82F100, thats some awsome research. It would be cool to have a twin-turbo'ed IDI...

I don't know what turbos they are, but I have one from an '89 and one from an '85 760 Volvo...

The older of those Volvo units is a TO3, the newer a TD04H.
I used to have a list of turbo specs, and OEM applications, but, for the life of me, I can't find it. What I am finding when I look for that stuff is listings like this:
http://cgi.ebay.com/ebaymotors/USED...1QQihZ013QQcategoryZ33742QQrdZ1QQcmdZViewItem
Which have compressor a/r and wheel size listed, which doesn't really mean much, as compressor a/r is almost inconsequential to the operation of the turbo, except that larger a/r's are generally used on larger flow rates.
A good bit of digging produced this:
http://www.gamesbbs.com/~turbosi/junkyard_turbo_list.html
Which helps lots.
The 2.3T turbo we are talking about here, with the smaller .48a/r exhaust side is a Garrett AirResearch TB0344, and a bit more digging produced this:

Now, for some plotting:


Looks like for a 12 psi application, it will work not too bad, but, we are nearing it's limits. It won't flow enough to make 15 psi.
Now the manual trans turbo has a 60 trim compressor,

Lets do some plotting:

The black is 12 psi, grey is 15, for some reason paint on this machine is grayscale only, and I have no clue why, I'll fix that later

Do a search on turboford.org You should be able to find the compressor maps for the stock T3.

My SVO has a T3/T4 with a .63 hot side an an S-3 compressor for a fun 22lbs of boost

I tossed around the idea of flipping the exhast manifolds while I was brainstorming my twin set up, but the problem is that the drivers side would Probably work but would be tight but the passenger side doesn't aim straight down and putting a drivers side manifold on the passenger side, the flange would get into your air box for your heating system.

I'd think that with the amount of fabbing you'll do, you'd be better off starting with making custom exhaust manifolds first and moving on from there.

Al in all, I think with the right planning and skill, you could build a functional system for less than a packaged kit.

I thought about the flipping the manifolds issue, it might just be easier to cut off the current outlets, weld the ends of the manifolds closed, and weld on a turbo flange where it would fit best, they do seem to be an alloy that would accept welding with proper prep.
But, by the time you do that, you could probably have bought header flanges and built a nice weld el turbo manifold.