|
|
 |
Eaton supercharger use on the Toyota 3MZ-FE V6 engine
|
Here I will make various observations and comments on how the 3MZ-FE engine, transmission &c. function, and in some cases how to make improvements. New material will be added regularly, and eventually better organized. Click below to jump directly to an individual topic. |
It is possible to adapt a late model GM 3.8 liter V6 (Buick, Oldsmobile, Chevrolet) Eaton M62 or M90 (the latest Generation V· is prefered) to the 3MZ. The nose drive is to the left (passenger’s side) and the throttle body points to the right (driver’s side), whereas the TRD and the 3MZ engine have the TB to the right and angled slightly forward, (as shown below right). |
|
|
The M62 delivers approximately 62” per revolution or 1,000cc. The M90 delivers approximately 90” per revolution or 1,500cc. Either could be used, but although the parasitic loss per revolution is higher with the M90, it appears that spinning the M62 faster requires more power since power requirement rises with the square of the RPM. The M90 will have slightly reduced low to mid-range boost due to normal case-to-rotor leakage.
Click here to read more about supercharger drive choices:
 .
The M90 can produce 15 psi in the 3.8 with the right pulleys etc. and can be had in good shape for about $300 on eBay, etc. It may be possible to adapt the Toyota IAC, MAF and TPS to the GM TB, saves some sizing etc. The 3MZ TB is larger at 70mm than the GM TB at 65mm, so an enlargement of the Eaton inlet port is needed if you wish to use the 70mm body, but in my opinion no substantial power loss will result if you use the GB TB with the Toyota MAF, TPS, etc.
The 3MZ knock sensors can adjust ignition timing to some extent to compensate for knock due to low octane, but cannot cope with strong knock produced by a combination of high compression × high boost pressure. If the original high (10.8:1) compression ratio is retained, maximum boost must be limited; slowing down the supercharger drive ratio will help.
There are many calculators and text on-line (some at reputable sites) that purport to predict the effective compression ratio by simply multiplying the current CR by (boost + atmospheric pressure) ÷ atmospheric pressure: an engine with 9:1 CR and 14.7 psi boost has 18:1 CR, 10:1 CR with 10 psi = 16.8:1, etc. This is also (erroneously) used to predict the power increase: if the boost is equal to atmospheric pressure, the power is doubled, etc. This is completely wrong. Click here for the actual physics:
. Using the correct formula:
ECR = CR × ((B + A) ÷ A).74074
For example, 5 psi boost pressure applied to an engine with 10.8:1 CR, operated at 0 elevation (sea level, 14.696 psi atmospheric pressure), the equivalent compression ratio is 13.42:1. This assumes that the intake temperature remains constant - which will never be true unless an intercooler or water injection is used. If the temperature rises (either the air temperature entering the supercharger, or the charge temperature entering the engine) the effective CR goes up. Here are some equivalent values for common boost pressures at constant temperature. |
10.8:1 compression ratio vs. boost |
Boost in psi | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 |
Effective CR | 11.3 | 11.9 | 12.4 | 12.9 | 13.4 | 13.9 | 14.4 | 14.9 | 15.4 | 15.9 | 16.3 | 16.8 | 17.3 | 17.7 | 18.2 |
Based on some TRD pictures the nose drive length is too long, but can be modified easily to move the drive pulley back for alignment with the existing belts.
Here’s a picture showing a TRD installation by Solaraguy.com member ndeguzma. Click on the picture for a larger view; click here for the thread and more details: | |
|
Here’s the TRD 1MZ supercharger (left) showing the passages to the intake manifold, click on either picture for a larger view. Note that these do not match the 3MZ (right). | |
|
The single biggest problem is adapting the Eaton discharge (exhaust) to the 3MZ upper manifold. It may be possible to cut up a plenum (upper manifold) to get a section with some volume and connections to the lower, and surgically alter it to make an open-topped box with an upper surface matched to the Eaton exhaust port. Only | |
|
the center “triangle” is essential, the small butterfly is a bypass that can enter the manifold with a separate duct or hose if needed. Click on the picture for a larger view. |
The compressor can discharge into the manifold, or duct away to an intercooler. The connecting duct work need not be a regular geometric shape in cross-section. However, it does mean that the TB is subject to boost, and will not behave properly since the density will play havoc with any MAF reading. On the plus side, TB diameter is less | |
|
critical, and the loss from using the stock TB (and the whole plenum/manifold) is probably not bad. Here’s a crude Phtoshop mock-up of what it might look like, click on the picture for a larger view. |
The TB operating in “draw-through” mode (as found on the TRD and Eaton superchargers serving the intake/suction port of the supercharger) definitely requires a bigger TB to reduce pumping loss.
The usual rule of N/A is: D × RPM × VE ÷ 3456 = CFM
For a 3MZ this about 202 × 5,800 × .95 ÷ 3456 = 322 CFM
Now multiply by the pressure ratio: (ATM + boost) ÷ ATM
Example, 10 psi: (14.7 + 10) ÷ 14.7 = 1.68
1.68 × 322 CFM = 541 CFM
However, the 541 CFM is absolute (no pressure loss due to size), so it’s not the same value as used to size carburetors where the rated CFM is at 1.5” Hg (vacuum in inches of mercury) or about .74 psi.
In general terms the diameter of a throttle body determines its flow in CFM. In round numbers the TB area in square inches × 70 will give the approximate CFM at .5 psi pressure loss. Using the above example of 541 CFM, this requires a minimum area of 7.39 square inches, or 3.067” (77.9mm) diameter. If either the engine size, engine speed, or boost were increased to produce 1,000 cfm, the minimum TB area would be about 14.3 square inches, or 4.265” (108mm) diameter.
Cams
Although the VVTi cam mechanism provides nice features for the 3MZ in normally aspirated mode, the exact cam events (opening and closing points of the valves), as well as the timing and amount of re-positioning, are not optimal for supercharged use. I don’t know if the VVTi interval can be re-programmed, which would save the expense of having all 4 cams reground - a very large expense for only a minor improvement.
If you wish, the cams (either the intake, the exhaust or both) can be reground and hardwelded by Web Cam to increase duration and/or lift (click here for the Web Cam web site and call them for a suggestion: ). There make no specific recommendations, but a grind suggested for a DOHC engine with 4 valves per cylinder, similar bore & stroke, and a combustion chamber design with a shallow valve angle of any brand (rather than an engine with a single cam, high valve angle, 2 valves per cylinder) is suitable.
Pistons
I have a suspicion that the only reason a 3MZ needs custom ($$) pistons is because of the dome shape and valve reliefs, which are suited to the exact angle and placement of the valves (not generic).
However, if you keep the stock heads and try to reduce the compression to a number that allows more boost (like 9:1), the volume can be added to the piston but only in the area inside the quench surfaces.
The 3MZ head at 10.8:1 compression has a 56.3cc chamber. To get 9:1 compression you have to add 12.67cc.
If this were a shallow dish in the center of the piston, say 70mm across, it would be about 3.29mm (.130”) deep. This may completely remove the valve reliefs, making the entire dome contour much more generic and making a swap for an existing (commercially available) piston more likely, although the big 22mm pin doesn't help. |
|
