Project Thread Supercharged Supra

[scotturnot]

Well one idea would be a top plate with a similar design to this. Forgive the boxy model, this is just a quick and dirty draw up to give you a idea of what I was thinking. It would need lots of filleting ect.
This drawing is a peice of 1/2" thick material but perhaps there is clearance enough for 3/4" or even 1"

 

[cribbj]

How about this: since the top of the vee divider in the manifold is about 1 1/4" from the top, what about making a subpan with a thin, flanged top, that would fit down in the manifold, sealing it off. Then discharge ports could be cut in the subpan? Might take a few iterations of the pan & ports, but it would be better than cutting on the manifold's discharge ports.

 

[mshawari1]

I was planning to do this to my M112:

Calculate the total area of the 8 runners and call it A. bring a soled 5 mm metal top plate, open a rectangle in the middle of it with W:L equals to the W:L of the complete hole with area equal to A. keeping the area A+ over the plate under the SC.

That will keep the pressuer equalized in the haul flow path. Saving the height also.

 

[scotturnot]

Not sure I totally follow you John.

 

[cribbj]

Well I suck at CAD, so just imagine a fabricated pan that looks something like what your wife uses to bake brownies in, only a bit deeper. The flange on that pan would sit on top of the flange on the manifold, and the pan would sit down in the manifold. The supercharger mounting plate would sit on top of the pan flange. Total additional height would be the thickness of the pan flange.

The pan could have one main circular cutout in its bottom, plus perhaps secondary cutouts at the front & rear. The secondary cutouts would be necessary because with only the main circular cutout, the center cylinders (2/3/4/5) would be robbing air from the four corners.

This thing could then be put on a flow bench and the cutouts fine tuned for (hopefully) equal air distribution.

 

[scotturnot]

Gotcha, Actually that might work, My only obstacal would be the intercooler. I need to study this some more.

 

[cribbj]

Or here's another take on it.

Looking at the photo in my earlier post, fabricate two plates out of 1/4" ally, that would sit vertically inside the manifold and run longitudinally, and block air flow to the intake ports. They would fasten to those round bulkheads between the ports.

Then bore different size orifices in these plates at each port to balance out the air flow. Again, the assembly would need to be put on a flow bench for balancing.

I just don't know if a balanced flow setup under low pressure (vacuum) conditions would still be balanced at high boost pressure, and I'm just feeling too lazy right now to sit down and study the orifice flow equation and think about it.

Something like this might still work OK with your intercooler.

 

[cribbj]

Dennis just finished flowing the 2nd motor's set of heads on his Superflow bench and the numbers look really good. They have had a mild street port, which he's going to optimise.

I did not have "stock" head flows of my own to compare against, so I used those from "The Flea", who had his stock heads flowed a few years ago and was kind enough to publish the numbers. I did have stock & ported 2JZ-GTE head flows from this same flow bench, and I've included the ported flows in these graphs for comparison. I've also included the flow figures from RMS's full race head, which has oversize valves and lots of port work.

Don't get too hung up over the absolute numbers here. It's the relative position of these curves to each other that's important. Flow benches are kinda like Dynojets. Each one is going to produce different numbers, and the same one will produce different numbers on different days, depending on ambient temperature, barometric pressure, and humidity. Also, it's not just big flow numbers that matter, but keeping the port velocities high for good responsiveness, scavenging, etc.

What these curves tell me is that doing a big valve modification is very worthwhile for the 1UZ motor, and especially on the exhaust side, not like the Supra, where more than one very high HP Supra owner have said they've seen no measurable difference in their cars' 1/4 mile time & speed with/without big valve heads.

What's also clear is that the exhaust side of the 1UZ head needs a LOT of work, just to compete with a ported 2JZ-GTE head, which still won't win any prizes on its exhaust side.

Once Dennis has finished porting the heads I'll revise these graphs.

 

[scotturnot]

Good stuff John! Looks like our engines need some real help on the exhaust side.
I have some work in process on my intake to try and better the air distribution, I will try to get some pictures up on my thread. I will tell you that my intercooler is going to be my "brownie pan"

 

[JBrady]

RMS exhaust outflows the stock intake!!!

Back to the discussion on air distribution be careful not to over “work” the air… turbulence and friction are not your efficiency friends.

Magnachargers new TVS 2300 intercooled system uses diffuser plates between the intercooler outlets and the intake ports. You can see them in the video below at 50-60 seconds.

http://videos.streetfire.net/video/Magna-Charger-TVSr_198555.htm

For some automotive porn I found the following intake that looks to be created with CNC milling… beautiful! May inspire.

My thoughts include using some A/W IC cores just before the inlets. Those laminova ones like striker is using could be the ticket. The vanes would straighten flow and also add cooling as a bonus.

 

[cribbj]

John, that's a gorgeous intake, please pass the tissues......

On the Richwood intake, I think that putting the same amount of diffusion in front of each intake port won't have the desired effect because it'll just slow down all flow by an equal amount.

Using an electrical analogy where current is flow, and voltage is pressure, if we model the Richwood intake by a parallel circuit of 8 resistors, where the resistors for 1/2/3/4 have a lower value (and a higher current through them) than 5/6/7/8 to simulate the out of balance flow situation.

Then with the Laminova, we add the same amount of resistance to each parallel circuit, we would simply reduce the current in the each circuit, they would still be out of balance.

What we want to do is increase the resistance in the circuits for ports 1/2/3/4 to reduce the current to them, so the currents going to all eight ports will become equal.

What we don't know, and where the simple electrical circuit analogy breaks down is what values these resistors need to be at different power outputs, because we don't know if the resistance to flow to the back ports is linear (and I'm assuming it's not).

So then we need to figure out how to increase/decrease the resistors as the overall power increases or decreases. Hmmm, sounds like some variable resistors (butterflies or slides) are needed ;-)

 

[JBrady]

Actually, I believe the pressure drop would be minimal and the "vanes" of the laminova core would function to balance flow without undue restriction.

But of course opinion with out testing is just an opinion.

The site the pictures came from is a Swedish turbo companies forum. Impressive amount of detail, modelling and testing on the 9 page thread alone. I have not run it through a translator but the pics alone tell a lot.

I recommend some viewing

http://forum.savarturbo.se/viewtopic.php?t=27561&postdays=0&postorder=asc&start=0

 

[cribbj]

Some nice CFD modeling there.

I think additional resistance (pressure drops) are the only way to even out the flow. You can see that in the first pic of that thread where the center tubes of the intercooler are doing the lion's share of the cooling.

In our case, if you can't "encourage" the flow to go to the back cylinders by using diverter channels (decreasing the flow resistance, as Scott has done by machining his manifold), then you'll have to add some resistance to the front cylinders to discourage the flow going to those.

Guess we'll just have to agree to disagree on this one ;-)

 

[cribbj]

I've been fooling around with some harness ideas while waiting for my 2nd motor to be built. Here's some photos of the prototype work I've done so far.

The idea is to build an engine harness using Milspec type connections and SS braided hose. In the pics is an example of an injector lead. Both ends (injector and the wiring rail) of the cable are removable. This same idea could be applied to the COP's and all the other engine sensors as well.

The idea is to harness all the injectors, coils, and other sensors into an attractive looking aluminum conduit or two, then put a large female Milspec connection on the end of that in order to plug the engine into the rest of the harness. Lots of bling, and very functional too. Don't ask how much it costs..... I'm working with some suppliers to get the component parts count reduced. Currently there's two adapters, and a hose fitting required to make up to that plug. I'm trying to get all that reduced to two pieces max.

 

[cribbj]

Remember that MRE (dual Hall Effect) sensor that I picked up a year or so ago? I finally had an opportunity to borrow a scope meter and test it. Below is a pic of its output when connected up to a normal 12v supply.

It was sensing a 36 tooth steel trigger wheel that I had chucked up into my electric drill. As you can see, it has a nice square wave, even if the frequency was only 378Hz (about 630 RPM) I wasn't spinning it very fast, as I didn't want the trigger wheel getting loose and becoming a flying buzz saw.

Cherry, the manufacturer of this sensor reckon it's good for at least 10Khz, which, with a 36 tooth wheel, is somewhere north of 16,000 RPM, and with the little 12 tooth wheel we use in the 1UZ is close to 50,000 RPM.

Even Erol doesn't spin his motors that fast......yet.

Anyway, it puts out a nice tight square wave, that has a fixed amplitude of whatever you feed it with. If your supply voltage is 12v, you get a 12v P/P square wave. If your supply is 5v, you get a 5v square wave. And it works very well with ferrous trigger wheels. None of this magnets-imbedded-precisely-in-the-right-spot-of-an-ally-wheel nonsense.

I'm going to adapt this puppy to the 1UZ for both the crank and the cam pickup and try it out next dyno session.

 

[scotturnot]

Those cables are sweet John! That will be the bee's knee's for sure.

 

[cribbj]

Baby got some new shoes

Haven't been totally neglecting the Supra itself. I recently picked up a used set of HRE 541 wheels & tires from a "seemingly" reputable guy on Supra Forums, named Jim Tripoli (Tripz), then once the wheels & tires arrived, I found out the wheels weren't the width or offset spec he said they were, plus they couldn't be balanced. Sent them off to HRE and they advised that 7 out of the 8 half rims needed to be replaced due to damage or general runout. So my beautiful new wheels were essentially junk, and needed to be overhauled. Not good.

After consultation with some wheel fitment experts in the Supra world, I decided to have HRE replace all 8 half rims, plus go with wider half rims to get 10" widths up front and 11.5" in the rear, so I could run 285/30 tires in front and 315/30 in the rear.

These widths required some body modifications to the Supra, such as removing some fender liner hardware up front, rolling the rear fender lips, and "relieving" the inner fenders at the frame rails by some judicious hammering. Once all this was done, we set 'er down on all fours, and no rubbing! Well, that's not 100% true, at full left steering lock, I have a slight amount of rubbing with one of the front tires, and we're going to sort that out on Monday.

I'm amazed at how well the car rides and handles with this big rubber. The ride is smoother and softer than when I had 235/40's and 285/35's on it recently, and those were new tires also. I think it's the combination of the wider wheels, and/or the Yokohama tire compound that's made the difference. The tramlining, and darting that this car has always done is now gone, and the car feels extremely stable, as well as handling like it's on rails.

Anyway, here's a rear shot of the car on the lift, coming down for its first roll out. I'll get some better shots this weekend after I get it washed and detailed.

 

[MitchP]

Looking good John! My SC is the same way - rolled rear fenders and no inner plastic in the front. When it comes to tires, bigger IS generally better!

Let me know if you need any assistance with the wiring design. I have some experience with mil-spec connectors.

 

[cribbj]

Thanks Mitch, it's been an interesting challenge learning the little that I know now about the world of milspec connectors.

I had settled on using ITT Cannon's Trident line of connectors as they're a metal reinforced CPC (circular plastic connector) type. They're not true military grade stuff, but more harsh environment industrial and/or motorsport, and they only cost about a third of a comparable milspec connector. But they're definitely a major step up from the CPC junk that's coming out of China and being sold by the major electronics distributors under their own name. I sampled some of those before settling on the Tridents.

Once I'd decided on the Trident line, the fun began, because the back threads on this line of connectors aren't the same as on the old MIL-C-5015 connectors. Because the MIL-C-5015 connector has been around as long as I have, there's a huge assortment of adapters available, but that's not the case for these Tridents. Eventually I found some job shops whose main business is building adapters for the milspec stuff, and they were willing to run a few specials for a nutty engineer to try.

These milspec connectors, and the people who make them, and the accessories, definitely have their own little world and their own vocabulary.

 

[cribbj]

Couple more shots of the dawg with its new shoes:



These puppies HOOK! I can still break loose in 1st without any trouble, and now in cold weather, when the boost comes on hard in 2nd, the rear gets loose again, but nothing like before, and 3rd seems to hold just fine, so I think the car is nearly ready for the V8 swap (as soon as I get tired of playing with it on the engine dyno.)

Last night the wife and I went to the Saturday night cruise-in that our local car club hosts at a favorite pizza parlor. There were some cherry '60's muscle cars there, a nice Ferrari, a Lotus, a couple of GT and Cobra Mustangs, several 'Vettes and Vipers, and a couple of beautiful Z cars. Heavy duty company, but the dawg held its own.

When we rolled in, the 25 and under crowd, who had been gawking at a really nice supercharged Viper, and a couple of C6's, recognised my car straight away as a Supra (thank you Fast & Furious....), and flocked over to have a look. The 50 and over muscle car guys weren't sure what it was, but since it obviously wasn't a 60's muscle car, it wasn't their cup of tea. The 'Vette & Viper boys looked like they'd just had their *sses handed to them, while the Lotus and Ferrari owners just ignored us.

The two "Z" owners (husband & wife) were really friendly, so we spent a very enjoyable hour with them swapping lies about our cars. Both were big fans of the Supra and the 2JZ-GTE, as I am as well of the "Z". So in the end, all the birds of a feather, flocked together.