Topic: Twin Charging: An academic exercise

I've always been fascinated by twin charging. Mostly from a "that seems over-complicated and therefore amazing" point of view. A couple times a year the dumb idea pops into my head that I should try it, not for any particular reason, just to try it. Looking around my garage the other day it dawns on me that I have a whole crap ton of Turbo dodge stuff sitting around, and an electronically clutched supercharger..... Now, will I follow through on any of these thoughts? maybe. I have a good track record for starting projects, and a crap record for finishing them.  Who cares? Lets talk theory.

There seems to be a couple ways to do twin charging.

1. Two separate intake tracks to the throttle body. One has a turbo, one has a supercharger. Find some way to close off one while the other works. Used very infrequently as it is flawed and doesn't work great.

2. Turbo feeding constantly into a supercharger. Supercharger works down low, and then as the turbo spools the supercharger acts as a "boost multiplier" as it compresses already compressed air. This works, just has some issues with making sure the supercharger can handle the compound pressure and additional heat load.

3. Turbo into supercharger with a by-pass valve. Basically the same as option #2, but there is a supercharger bypass that you open up when the turbo spools enough and essentially takes the supercharger out of the equation.

I like option #3. Mostly because I see it as a way to create fairly even boost levels all across the rev range. I'd want to set it up such that the supercharger makes ~8psi, and then the turbo can build all the way to ~12psi. In addition it seems easy to "turn off" and run like a normal turbo engine if the need were to arise. So, how would you set this up? I would do it like this. Turbo setup like normal. The turbo feeds into the supercharger in. The super charger feeds into an intercooler. The intercooler feeds to the throttle body. At the supercharger would be a bypass tube linking the supercharger inlet side to the exit side. When this valve is closed, all air goes through the supercharger. When it opens it allows air to skip the supercharger. Ultimately when the supercharger is fully bypassed it's clutch is switched off and you run purely on turbo. Basically looks like this, but without the intercooler between the turbo and supercharger. I think I'd throw in a water injection setup just before the supercharger to cool the lobes.

http://i14.photobucket.com/albums/a304/boss5705/Car/s4inductionsystem4km.jpg~original

But layout is the easy part. Where I start hitting walls is control. Engine tuning is easy. The turbo dodge stuff is purely MAP based. So it only cares about intake manifold pressure. Adding a supercharger doesn't change anything, it just reads from other parts of it's fueling maps. It's control of the bypass valve and supercharger that scare me. I'm sure you could set it all up to run from basic vacuum/boost and pressure sensors. Something a diaphragm driven valve that starts to open once boost before the supercharger reaches a certain level, and then when the valve opens a certain percentage the supercharger is flicked off. The trouble I see is that you probably want the pypass to stay open on aggressive upshifts, or on light off throttle situations where you mean to be back in the throttle quickly, and I don't really know how to build that in.

So let's complicate more! Electronically control all the things! It can't be that hard to build an arduino controlled system that monitors pre-supercharger boost, intake manifold boost, throttle position, RPM, etc and have all that determine when to open and close a bypass. For the actual bypass you could use an electronically actuated throttle body, or something similar. With the Arduino you can build in those upshift delays and any other advanced control.

Oh right, I suck at programming. Who cares! I'll learn.

Here's my main hangups.
1. Writing the control program for the bypass and supercharger clutch logic

2. Changeover logic. How to you determine when it's time to start opening the bypass. You can't just say "when pressure into supercharger = pressure out" because the supercharger is multiplying the in. I guess you could use logic that says when you start seeing the boost pressure climb you know you're feeding pressure into the supercharger and it's time to start bypassing

3. Do you block off the supercharger as you bypass? If you don't block it everything equalizes as the bypass opens, so as long as the supercharger is turned off it just sits there with equal pressure on either side. But I imagine that you'd start opening that bypass before the turbo is making the same boost as the supercharger, so there would be a brief moment of backward flow as you open the valve. Does it have any benefit to partially block off the supercharger inlet to try and prevent a spike in compounding pressure?

Those are my madman ramblings so far. I need to think a lot of this through, then build up another engine to use as a guinea pig test bed. Who knows what i'll even use this for, other than just a "look what I managed to do" show piece.

20+ Time Loser FutilityMotorsport
Abandoned E36 Build
2008 Saab 9-5Aero Wagon
Retired - 1989 Dodge Daytona Shelby 2011-2015 "Lifetime Award for Lack of Achievement" IOE, 3X I got screwed, Organizer's Choice

Re: Twin Charging: An academic exercise

I'm a fan of option #1 because it can be done simply and mechanically.
https://www.youtube.com/watch?v=7hT_F-KeJDM

Planet Express
"IOE" "C Win" 4834.701 Race Miles and counting
Toyocedes
"Least Southern Pickup Truck" "IOE" "C win" "C win (again?)"

Re: Twin Charging: An academic exercise

What I don't see is when the turbo is "off", how does the supercharger get intake air?  Seems like the turbo needs to be spinning to allow air through but you want to be doing that without providing boost.  As far as control, you could use a bullet type AFM with a super-precision calibrated spring that opens at a preset boost level.  NA RX7's have one.  It's a bullet shaped valve in the intake tube that acts like a plug opens with airflow but seems like you could put a spring in there to control it.  I have no idea how much enjoyment it gives the turbo to dam the output like that though.  I guess you could have a blowoff valve in the pressure tube but then how does your other controls know how much boost it's making?  Maybe just make an educated guess at RPM and have your dump valve open then.
Seems to me like option #1 would be the easiest for a backyard solution.  Transitioning this whole thing is gonna not be fun.  Also not sure how either of these devices like being blown through without operating.

Or why bothering cutting out either device.  Let them run free and then limit boost on the output side of the compressors.  Or limit the intake size.

1990 RX7 "Mazdarita"  1964 Sunbeam Imp (IOE 2013 Sears Pointless) 2002 Jaguar x-type (Winner C-Class 2021 Sears Pointless)
Gone bye-bye
1994 Jaguar XJ12 (Winner C-Class 2013 Sears Pointless)  1980 Rover SD1 (I Got Screwed 2014 Return of Lemonites)

Re: Twin Charging: An academic exercise

Just make the driver manage it.

https://s-media-cache-ak0.pinimg.com/736x/e9/2f/e4/e92fe4f120842c0ef7ff089b9f5e264b.jpg

100% no-fail.

Mistake By The Lake Racing (MBTL)
88 Thunderbird "THUNDERBIRDS ARE GO!", Ex Astris, Rubigo / Semper Fracti
A&D: 2014 Sebrings at Sebring (NSF), 2014 NJMP2 Jurassic Park (SpeedyCop), 2012 Summit Point J30 (PiNuts)
2018 Route Sucky-Suck Rally Miata, 2019 World Tour Of Texas 64 Newport

Re: Twin Charging: An academic exercise

FPRbuzz wrote:

I'm a fan of option #1 because it can be done simply and mechanically.
https://www.youtube.com/watch?v=7hT_F-KeJDM

That doesn't really look like #1. That looks like option #4 which is to have the supercharger feed the turbo, and then let it suck in more air when needed. Interesting approach.


cheseroo wrote:

What I don't see is when the turbo is "off", how does the supercharger get intake air?  Seems like the turbo needs to be spinning to allow air through but you want to be doing that without providing boost.  As far as control, you could use a bullet type AFM with a super-precision calibrated spring that opens at a preset boost level.  NA RX7's have one.  It's a bullet shaped valve in the intake tube that acts like a plug opens with airflow but seems like you could put a spring in there to control it.  I have no idea how much enjoyment it gives the turbo to dam the output like that though.  I guess you could have a blowoff valve in the pressure tube but then how does your other controls know how much boost it's making?  Maybe just make an educated guess at RPM and have your dump valve open then.
Seems to me like option #1 would be the easiest for a backyard solution.  Transitioning this whole thing is gonna not be fun.  Also not sure how either of these devices like being blown through without operating.

Or why bothering cutting out either device.  Let them run free and then limit boost on the output side of the compressors.  Or limit the intake size.

The turbo still flows air when not on boost. The supercharger can pull through it just fine, and in fact that helps with some initial spool up.

if you have two truly separate intakes one with each device, you need to block one at some point because one device will always flow more than the other and you start getting back flow. You could do option #2 and just have the turbo feed into the supercharger all the time. But it creates a lot more heat, and you have to size the turbo to match what's happening. If you push 8PSI to the supercharger it will bump that up further to the engine. Can and has been done, just has to be done correctly.

20+ Time Loser FutilityMotorsport
Abandoned E36 Build
2008 Saab 9-5Aero Wagon
Retired - 1989 Dodge Daytona Shelby 2011-2015 "Lifetime Award for Lack of Achievement" IOE, 3X I got screwed, Organizer's Choice

6 (edited by rmcdaniels 2016-05-06 04:54 AM)

Re: Twin Charging: An academic exercise

I did that a while back with a 1.6L Honda. It was strange, like the car had a 350 WHP V8 under the hood. Torque delivery was incredibly smooth and controllable. The only downside was heat soak, which got bad if you pushed it too hard for too long. It made for an insane autocross car, that thing won every event that it entered. Here's some details and stuff:

http://honda-tech.com/forced-induction- … s-1293497/

I went with option #3, as I wanted the SC to spool the turbo and then get out of the way as it's efficiency dropped off.  Here's a diagram:

http://s115271005.onlinehome.us/TC/TC_files/gif_1.gif

I dumped some pics and video here:

http://s115271005.onlinehome.us/TC/

Here's what it looked like:

http://s115271005.onlinehome.us/TC/TCF1.JPG

http://s115271005.onlinehome.us/TC/TCF2.JPG


Video:

https://youtu.be/1GWAQupFGzc


The Dyno Sheet:

http://s115271005.onlinehome.us/TC/dynoTCF1.jpg

The Car:

http://s115271005.onlinehome.us/Track%20Day%20Pics/thsccapr4277S.JPG

The hood scoop was to direct some air on the SC case.

Everybody grab your brooms, it's shenanigans!

Re: Twin Charging: An academic exercise

I will have to go through and read all of that, thank you. Sounds like you had a slick simple transition system. I may use that as a baseline starting point.

I do like the idea of supercharger feeding a turbo and then getting phased out as the turbo spools too. I may explore both.

20+ Time Loser FutilityMotorsport
Abandoned E36 Build
2008 Saab 9-5Aero Wagon
Retired - 1989 Dodge Daytona Shelby 2011-2015 "Lifetime Award for Lack of Achievement" IOE, 3X I got screwed, Organizer's Choice

Re: Twin Charging: An academic exercise

I could have gone more complicated, but the simple solution worked very well. The car was extremely easy to drive, with no hint of the insanity that was going on under the hood.

Phasing out the SC was based on the compressor map of each device. The SC at any higher engine RPM got to the edge of it's map and lost VE quickly, so I sized the turbo compressor so that it was in it's sweet spot at max engine RPM and cut out the SC.

The other factor that I underestimated is that turbos are essentially spooled by engine load, so the SC loading the engine instantly on any throttle application spooled the turbo incredibly quickly, but also very smoothly. That was probably the best thing about the system; I could run a big turbo on a little 1.6 engine and the torque delivery was perfect. Even rolling on throttle in mid-turn was smooth, where with just the turbo it was dangerous.

Everybody grab your brooms, it's shenanigans!

Re: Twin Charging: An academic exercise

One issue with turbo feeding super is potential oil consumption when turbo boost is low.

Re: Twin Charging: An academic exercise

It was never an issue with my setup. Where would the oil come from?

Everybody grab your brooms, it's shenanigans!

Re: Twin Charging: An academic exercise

TheEngineer wrote:

Mostly from a "that seems over-complicated and therefore amazing" point of view.

In that case be sure to use the output from the supercharger to pressurize the fuel tank, thereby eliminating the need for a fuel pump. This works great on my French car, so what could possibly go wrong?

https://farm6.staticflickr.com/5088/5731724122_d1f23555a3.jpg

1982 MG Metro 1300: IOE 2015 Pacific Northworst GP, Longest Distance 2010 Cd'L Box Wine Country Classic
1980 KV Mini 1: Worst of Show and Fright Pig Supremo 2009 Concours d'Lemons
1978 H Special: Second-Round Elimination 2010 Lemons Pinewood Derby at Sears Pointless
1967 SAAB 96: IOE 2012 Pacific Northworst GP, Organizer's Choice 2022 Hell on Wheels California Rally

Re: Twin Charging: An academic exercise

Really excessive complication here.  Run a big turbo that works well for max power and RPM, then use a modest nitrous system to spool up the turbo.  Works beyond expectations!

Re: Twin Charging: An academic exercise

http://www.roadandtrack.com/car-culture … -charging/

ERM Racing:  NHMS 2016 - Organizer's Choice Award. NHMS 2015 - Judge's Choice Award. NHMS 2014 - "Most work to make an Unrecognizable Movie Car Award" Team Ludicrous Speed:  NHMS Fall 2013 - driver, NHMS Spring 2013 -  support crew, NHMS Fall 2012 - Corner Marshal (How not to Flip Your Car)

Re: Twin Charging: An academic exercise

That works for the street, but no type of motorsports where it would be useful would allow it.

Everybody grab your brooms, it's shenanigans!

Re: Twin Charging: An academic exercise

Who allows it is beyond the scope of discussion.  In reality, nitrous over nitromethane is the hot ticket for power on demand,,,no...really...but I digress. 
If the crank driven compressor will pass more air at low speed, it will increase the pressure drop across the exhaust driven compressor and thus spin it up more quickly...and still get the air to the engine. 
When the exhaust driven compressor becomes dominate it will:
be hindered more by the crank driven compressor if that compressor is of the positive displacement type, or
will add to the output of the crank driven compressor if it is of the centrifugal type.
IMHO, if using the OP"s described system, a simple check valve around the supercharger would do just fine.

Re: Twin Charging: An academic exercise

On the dyno graph that I posted, the red lines are without the supercharger and the blue lines are with the supercharger.

Everybody grab your brooms, it's shenanigans!

Re: Twin Charging: An academic exercise

rmcdaniels wrote:

It was never an issue with my setup. Where would the oil come from?

Turbo seals work by pressure balance. They require higher pressure on the air side compared to the oil drain side. If you are sucking on the turbo, you could create a vacuum and suck the oil past the seal.

Re: Twin Charging: An academic exercise

Would it be the same for ball and journal bearing turbos? I had a ball bearing turbo with a pinhole restrictor in the oil feed line on that setup, so there was not much oil in it. At any rate, I never noticed any oil utilization.

Everybody grab your brooms, it's shenanigans!

Re: Twin Charging: An academic exercise

How about several electric-motor-driven turbochargers (with janky gear-drives to get the turbos spinning fast enough), feeding into a long 8" sewer pipe as a pressurized-air reservoir? That way you can have a massive 60psi blast into the engine when you hit the AIIIIEEEE!!! button!

No, don't. I make joke. The shrapnel potential is too high.

Re: Twin Charging: An academic exercise

rmcdaniels wrote:

Would it be the same for ball and journal bearing turbos? I had a ball bearing turbo with a pinhole restrictor in the oil feed line on that setup, so there was not much oil in it. At any rate, I never noticed any oil utilization.

Yes. Some seal designs are better than others at low pressures, but it is a risk on all.

Re: Twin Charging: An academic exercise

So theoretically I'd want another bypass valve around the compressor until any positive pressure is seen at the compressor outlet.

Everybody grab your brooms, it's shenanigans!

22 (edited by TheEngineer 2016-05-06 12:10 PM)

Re: Twin Charging: An academic exercise

rmcdaniels wrote:

So theoretically I'd want another bypass valve around the compressor until any positive pressure is seen at the compressor outlet.

The guys that are doing the supercharger into turbo seem to use a swing valve to atmosphere between the two so that when the turbo starts asking for more than the supercharger can deliver it pulls the extra air from there. You could do the same with turbo into supercharger. Swing valve between the two to alleviate the negative pressure (mostly) and once the turbo spools up it closes.

http://ulfmapproved.com/wp-content/uploads/2015/01/Disegno-tecnico-check-.gif

20+ Time Loser FutilityMotorsport
Abandoned E36 Build
2008 Saab 9-5Aero Wagon
Retired - 1989 Dodge Daytona Shelby 2011-2015 "Lifetime Award for Lack of Achievement" IOE, 3X I got screwed, Organizer's Choice

Re: Twin Charging: An academic exercise

Why not an rpm actuated solenoid to open some sort of bypass valve?  You could use a solenoid, a vacuum reservoir, and a bosch bypass valve or 2.  You could do pressure and rpm actuation wihtout getting too carried away.

I think using map load sensing you will have tuning issues unless you are using rpm as a switch point.  Turbo back pressure will effect fuel needed.  If you have 10 psi in the intake with no exhaust back pressure (supercharger only) then 10 psi with back pressure (turbo working)  fueling needs will be different but the load reference will be the same.  A mass airflow sensor would make tuning this easier i think.. Or if you could switch between maps, that could do it too.

There are also quick spool valves and litle turbos feeding big turbos to throw more overcomplication? it cant fail! ideas your way.

-Killer B's (as in rally) '84 4000Q 4.2V8. Audis never win?

Re: Twin Charging: An academic exercise

MAF tuning is generally easier, but this can be done properly with speed density.

There will always be back pressure; even off idle the turbo is still in the exhaust path.

Everybody grab your brooms, it's shenanigans!

25 (edited by derekste 2016-05-08 11:00 AM)

Re: Twin Charging: An academic exercise

You'd want the turbo inlet plumbed to the supercharger outlet, not the other way around.

Edit: I see this has been covered since the 1st post

"THE WONDERMENT CONSORTIUM"
Everything dies baby that's a fact,
But maybe everything that dies someday comes back?