Did some calculations for a twinscrew charger

For those other forced induction brothers
Post Reply
DjayS12
Knows Some Stuff About 240's!
Posts: 215
Joined: Thu Mar 12, 2009 7:21 pm
Location: Rimouski, Qc, Canada

Did some calculations for a twinscrew charger

Post by DjayS12 »

I was wondering if I forgot to take anything in consideration, anyway, here are my calculations.

Supercharger is the W100AX from whipple

Spec for supercharger:
-1.6 liter/rev
-18.000 rpm max
-adiabatic efficiency range from 70-85%
All these number came from whipple's website.

spec for ka24de
crank pulley diameter: 149 mm
displacement: 2.4 liters, 4 strokes so 1.2 liter per rev

Let say my ka won't see anything above 7,200 rpm

18,000 / 7,200 = 2.5

This mean the SC pulley must have a circumference no smaller than 2.5 times less than crank pulley

1.6 X .7 = 1.12

According to the least adiabatic efficiency, the SC will flow at least 1.12 liter of air per revs

1.2 / 1.12 = 1.08

To feed the engine with at least the amount of his displacement, the sc must rev at least 1.08 faster than the engine.

so the pulley can have a rev ratio anywhere between 1.08 and 2.5 times the engine revs.

3.14159 X .149 = .468

the crank pulley have a circumference of .468 meters so:

.468 / 2.15 = .436 and .468 / 2.5 = .187

so the SC pulley have to be somewhere between .436 and .187 meter in circumference.

.218 / 3.14159 = .14 and .187 / 3.14159 = .06

so the SC pulley diameter have to be somewhere between 60 and 140mm ( 2.4 and 5.6 inches)


Have I forgot anything? I couldn't find a map for this supercharger so the only info I had was what is given on their website.

The main thing I am unsure about is the need to have the supercharger to outflow the engine capacity, since in real life the engine don't really fully fill his cylinders.. but since the point is to increase the intake manifold air pressure, it kind of make sense to make sure that there is no way the engine outflow the supercharger.
Image
DjayS12
Knows Some Stuff About 240's!
Posts: 215
Joined: Thu Mar 12, 2009 7:21 pm
Location: Rimouski, Qc, Canada

Post by DjayS12 »

Looking back at these, if I wanted the SC to boost around 18psi (it's a twinscrew, not a roots, it will compress that much).

18 / 14.7 = 1.22

1 + 1.22 = 2.22

at 14.7 psi atmospheric pressure, the compression ratio is 2.22

In the post above, I calculated that to flow the engine AT ATMOSPHERIC PRESSURE, the SC need to rev 1.08 time the speed of the engine.

So in order to create a compression ration of 2.22 it need to rev 2.22 time faster than that.

1.08 X 2.22 = 2.4

so it need to rev at least 2.4 time faster than the engine to keep the boost around 18 psi?

.468 / 2.4 = .195

.195 / 3.14159 = .062 or 62 mm or 2.5 inches of diameter

So anything between 2.4 and 2.5 inches of diameter on the SC pulley will give between 18psi and max rev for the SC.

Does it at least make sense? The numbers sound about right I think...
Image
User avatar
superDorifto
Dont Question My Nissan Knowledge
Posts: 508
Joined: Tue Oct 11, 2005 6:58 pm

Post by superDorifto »

Your numbers make sense if you were only designing to a max design point at you max engine speed, but correlating the pressure ratio agains the charger's effeciency curve will let you know if the unit will be a good fit for the engine.

This is taken from another forum, for Porches, but the post is quoting Corky Bell:
Keep in mind that you should be looking at the values where peak torque occurs, not peak horsepower. It should be noted too that a street car will hit it’s torque peak more often than it’s peak horsepower.

The torque peak is the RPM where the engine achieves it's max chamber filling ability, and the trick is to match the compressor’s peak efficiency at this point.

Since we are shooting for peak torque RPM and not peak horsepower RPM, we will end up with a redline airflow point somewhat to the right of the island, exactly what is desired; however, Corky recommends not letting it go below 65%.

For a street car, the rule of thumb is take the torque peak RPM and add 25%. If we use 4,500 RPM as the torque peak, then the RPM we are shooting for is 5,625 RPM.

Using 5,625 RPM as the torque peak and 10 psi (..1.68 PR), then the resulting target is 726.81 CFM.

For 10 psi, find the compressor whose max efficiency occurs at 727 CFM and 1.68 PR.


All info from Langavulin

Link:

<http://forums.rennlist.com/rennforums/9 ... -maps.html>

Did they have a compressor map on their site?

You effectively calculated your ideal airflow if Your VE was 1.0, but you need to factor in your (theoretical)VE as well to get a better idea of the total air volume the motor will be moving at you max flow condition - as long as that point lands you in the sweet spot for the copmressors efficiency you sholuld be good.

This is all assuming that the pulley size you calculated will let you maintain the needed crompressor speed.

It might also be helpful to look at a few other points, like imediately off idle, and clutch engagement to understand what sort of efficiency range you will be in at those conditions. You want to be sure that you are building addequate boost from a stop.
DjayS12
Knows Some Stuff About 240's!
Posts: 215
Joined: Thu Mar 12, 2009 7:21 pm
Location: Rimouski, Qc, Canada

Post by DjayS12 »

Unfortunately no, they do not have a map, I just emailed them to see if they have one.

This is a stock ka24de dyno I found online.
Image

So peak torque seem to be a little bit past 4000 RPM, so having a compressor map I should try to match the best adiabatic efficiency to this RPM if I got the quote right.

Doing this, the max airflow at the redline will be on the right side of the island, meaning that area where the compressor is the most efficient have been used. but if on the redline I fell under 65% of efficiency, I should slow the supercharger a bit by using a bigger pulley.

Since I want my car to be a track/street car, I should aim at 5000RPM for the peak efficiency. (the rule of thumb of +25%)

The part I'm not sure to understand is the off idle and clutch engagement part.

From my understanding a turbo engine will still see vaccuum at idle.. but with a supercharger, the compressor is still running since it follow the speed of the engine... wouldn't that cause the intake to be at a positive pressure on all rpm?

As for clutch engagement, I don't really see what you mean.
Image
User avatar
superDorifto
Dont Question My Nissan Knowledge
Posts: 508
Joined: Tue Oct 11, 2005 6:58 pm

Post by superDorifto »

If you plot the CFM and pressure ratio and you are to the left of the efficiency plots, you are surging the compressor and just cavitating the intake charge.

If you are to the right, Its technically not surge, but its just as bad becasue the compressor is not working optimally

Ideally, you want to size a compressor such that you stay in the "sweet spot" of the compressor map for as much of the engine's operating range as possible.

You want to target that range where the compressor is generating the most delta pressure with the least associated increas in temp. What is hard for a lot of people to picture is that the compressor map is acually a 3D plot where the temperature axis has been omited. You are staring' "top down" at the map which makes it look like a topographical map for a mountain.

Stray too far to the left or right and you fall "off the mountain"...literally

This map is not for your compressor, but it shows how you should read the graph.
http://www.perrinperformance.com/shared ... ompmap.jpg

You map the MIN effective rpm and MAX relative to the CFM flow and PR

Then you conect them, this linear graph represents the engine running through its rpm range, when it gets to your max boost limiting condition (AKA whatever your recirc/BOV is set at, the graph will run directly to the right.

The idea is to size the charger, and the pulley to keep the compressor in its most efficient range for as long as possible without hitting (or getting anywhere CLOSE to surge)

Once youve mapped your line, you want to plot the point for your max tourque and be sure that its right in the meaty, super efficient part of the compressor efficeincy range.


EDIT
I went back and edited my post becasue I am used to reading turbo maps with 75% and 85% efficient regions...went back and looked at an M90, and the best it can do before it starts pumping heat is ~54%, but in that instance, I would be targeting 40-50%.
DjayS12
Knows Some Stuff About 240's!
Posts: 215
Joined: Thu Mar 12, 2009 7:21 pm
Location: Rimouski, Qc, Canada

Post by DjayS12 »

I looked around on lysholm (they make very similar twinscrew) and found their map for the 1.2 , 1.6 and 2.3 L/rev superchargers.

I've plugged in where the car will be at 3000, 5000 and 7000RPM, using this site for the CFM needs of the engine:

http://www.lovehorsepower.com/joomla/in ... &Itemid=49

Looks like the 1.6 L/rev is the one that'll fit the best. I know it's a bit above than the best efficiency at 2.0 of PR, but it stay in a good efficiency zone of 64% from around 4500 to 7000RPM and a discharge temp around 110-120 *C.

Anyway here are my plotting, notice I've drawn the straight line from 0-3000RPM since a SC produce boost at much lower RPM than a turbo:

Image
Image
Image

120*C = 248*F of discharge temperature seems good enough to avoid the need of water-meth or aftercooler. Maybe running a slightly colder thermostat (mine is a 190*F)


***EDIT*** Whipple just sent me their compressor map by email. It seem to be able to build more boost but wild be slightly less efficient.

Image
Image
User avatar
superDorifto
Dont Question My Nissan Knowledge
Posts: 508
Joined: Tue Oct 11, 2005 6:58 pm

Post by superDorifto »

I'm at work, so I cant see the pic you posted....but that link to the cfm calculator is pure gold.

When I was answering you question above I was doing the same calc in excel, but having it autogenerat across the entire rpm range is a fabulous way to understand the engines behaviour throughout th erev range.

As soon as I can log in form a computer that will let me see the pic I'll edit my post if any further coment is needed.
User avatar
superDorifto
Dont Question My Nissan Knowledge
Posts: 508
Joined: Tue Oct 11, 2005 6:58 pm

Post by superDorifto »

Assuming its plotted right, the first charger you posted looks like it will be singing right in the compressor's sweet spot as soon as you hit full boost.

I was a little bored during lunch today so I plotted an M62 and M90, the two most common chargers that seem to pop up for the KA.

The flow numbers are not corrected for temperature, so they are not 100% accurate, also I plotted the initial linear PR ramp as a curve because its pretty...

Assumptions,
Sea level
6500 redline
13 psi max boost (Pulley sized to hit max boost at max NA Tq (4400 rpms)
KA24DE with all necessary mods to make that power.

Image

Image

You can see just how much more breathing room the M90 has. at 6500 rpms, the efficiency is dropping off and starting to dump heat into the charge...

Makes me rethink my own set up.
DjayS12
Knows Some Stuff About 240's!
Posts: 215
Joined: Thu Mar 12, 2009 7:21 pm
Location: Rimouski, Qc, Canada

Post by DjayS12 »

I agree that the lysholm 1200ax seem to fit as much as the 1600ax, but what I like with the whipple is the capacity to go with as much as 18psi of boost and stay efficient... the roots, even if the TVS are great, just have an hard time achieving such boost level without heating up the air.. I'll have to think what I want a bit further... but do you see anything wrong about the 2.2 PR plotting on the Whipple map?
Image
User avatar
superDorifto
Dont Question My Nissan Knowledge
Posts: 508
Joined: Tue Oct 11, 2005 6:58 pm

Post by superDorifto »

Nothing jumps out.

I mean there are the blank sections on the map, but Its not clear if those are an area of surge.
alr11606
Driving Mom's Station Wagon
Posts: 23
Joined: Sun Dec 08, 2013 7:32 am

Re: Did some calculations for a twinscrew charger

Post by alr11606 »

I know this is a dead thread but.... judging by the graphing the M112 is the best supercharger for efficiency all the way to redline but IDK how surge would be in lower rpms.
Post Reply