The Top DSM Community on the Web

For 1990-1999 Mitsubishi Eclipse, Eagle Talon, Plymouth Laser, and Galant VR-4 Owners. Log in to remove most ads.

Please Support Morrison Fabrications
Please Support Fuel Injector Clinic

HE351VE Controller Project

This site may earn a commission from merchant
affiliate links, including eBay, Amazon, and others.

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

It will be interesting to see how well this controls the turbo. I was talking with a Cummins rep the other day. One of their guys came to PSU to talk to the FSAE club. Talk switched to VGT turbos (we run a small VGT Aerocharger on our car), and I mentioned something about this project. He was really interested since he used to have a 2g AWD Talon. He had heard that we (DSMers) have become go-to guys for Holset HX35, 40, 52, etc info. He was wondering how long it would take until people started making their own controllers and strapping VGT versions to DSMs. He said he's proud to see we're getting there first. LOL

A name would be great :thumb: ...and do you know which Cummins location he was from?
 
I didn't catch a name. He'll be back later this year. Cummins has a big truck with a bunch of field testing equipment. He said he's going to try bringing it out to PSU later this semester or early next so we can get some data off last years car.

I just found it funny that he knows about DSMers and Holsets right now. He's quite a bit bigger than a nobody in that company.
 
Behold... The NEMA stepper motor bracket machined to replace the Holset motor on the 351VE gearbox. Prototype machining and CMM sure aren't cheap!

This means we will have some video of the controller driving the nozzle ring in the near future.

You must be logged in to view this image or video.


You must be logged in to view this image or video.
 
Programed the interface for the stepper motor. If anyone is interested, I used an ON Semiconductor AMIS30532C5321RG driver IC with SPI interface to my microcontroller. The most difficult parts about interfacing with this IC were setting the registers correctly to match the motor current and time parameters and monitoring the fault registers for error conditions. Can you believe that this humble IC can regulate a stepper coil current to 3.075A?! :hellyeah:

I was able to make a merry video in the nick of time for Christmas. Thank goodness, if I didn't get this video out I would surely be getting coal from everyone following this thread.

Now that the hardware/PCB is verified, the next step is diving into the first generation of usable software for this board (i.e. putting a man on the moon). There are some minor PCB revisions like moving the compressor outlet pressure sensor to the board which I'll post in the near future.

<iframe width="480" height="360" src="http://www.youtube.com/embed/Pym2Wfzpf5c" frameborder="0" allowfullscreen></iframe>
 
If you wouldn't mind, could you see just how much current it's going to take with that NEMA motor? I'm looking at using a AMIS30624 to drive my stepper motor, since the I2C stuff is pretty simple, but it's only rated at 800mA per coil (so 1.6A total). I think you mentioned measuring 1.2Arms before to drive the Holset stepper.

The only reason I'm really looking at it is that I can buy it on a breakout board off eBay. That's more my level of electrical knowledge.

And are you using something similar to Freescale Semiconductor's MPXH6250AC6U for the pressure sensors? They have a few different pressure ranges to choose from and at the moment, that looks to be the route I'm going.

I'm looking to start working on my setup in January. Then use some tax return money to get the manifold and other fabrication done early March..
 
If you wouldn't mind, could you see just how much current it's going to take with that NEMA motor? I'm looking at using a AMIS30624 to drive my stepper motor, since the I2C stuff is pretty simple, but it's only rated at 800mA per coil (so 1.6A total). I think you mentioned measuring 1.2Arms before to drive the Holset stepper.

The only reason I'm really looking at it is that I can buy it on a breakout board off eBay. That's more my level of electrical knowledge.

And are you using something similar to Freescale Semiconductor's MPXH6250AC6U for the pressure sensors? They have a few different pressure ranges to choose from and at the moment, that looks to be the route I'm going.

I'm looking to start working on my setup in January. Then use some tax return money to get the manifold and other fabrication done early March..

800mA is enough to move the vanes, but it depends on how fast you want to go. Since stepping is open loop, you've got to have enough current to drive torque through the gearbox at the stepping interval you choose. The faster you step, the higher the current has to prevent stalling. Better to have more current than required and not worry about stalling or missing steps because that will throw your vane position off.

Right now, I'm using a Honeywell pressure sensor. They're basically differential voltage output from a bridge, so there is some signal conditioning involved. Be careful, that MPX looks like an absolute pressure sensor... which means it will read atmospheric pressure of 14.5 psi at 0 psi boost. So it would saturate at boost pressure of 21.5psi.

Look at MPXH6400AC6T1, also an absolute sensor, but I like it better because it goes up to 58psi... meaning you could sense up to 43.5psi boost.

Second note... mind the temperature error band on those MPX sensors, below 0C and above 85C, they'll need some fierce correction factors to compensate for the gage factor. Also note that will limit you because the error is a positive gain; remember you're limited to 0-5V, so when it's cold outside the MPX sensor has a narrower range than when it's warm.

Anyway, great suggestion... you made me look at different pressure sensors. Might have to try one out!
 
I just pulled that one off my list.

Off Digi-Key:
MPXH6115AC6U-ND - 3-16.7 psia, 1/8" hose, 0-4.8V, 5Vin, $10.24ea
MPXH6250AC6U-ND - 3-36 psia, 1/8" hose, 0-4.9V, 5Vin, $9.90ea
MPXH6300AC6U-ND - 3-42 psia, 1/8" hose, 0-4.9V, 5Vin, $9.90ea
MPXH6400AC6U-ND - 3-58 psia, 1/8" hose, 0-4.8V, 5Vin, $9.90ea

Figured the 115's would work pre turbo, the 300's would probably work post turbine. The 400's would be needed at all the other spots except maybe pre-turbine, where they might not even be enough.

The temperature thing had me a little worried, but I'm pretty sure I'll be mounting the sensors away from 185*F heat. The 32*F is a little bit of a problem in these colder winter months. It wouldn't be hard to throw a little ambient temp sensor on the board to do corrections if it became a problem.
 
I just pulled that one off my list.

Off Digi-Key:
MPXH6115AC6U-ND - 3-16.7 psia, 1/8" hose, 0-4.8V, 5Vin, $10.24ea
MPXH6250AC6U-ND - 3-36 psia, 1/8" hose, 0-4.9V, 5Vin, $9.90ea
MPXH6300AC6U-ND - 3-42 psia, 1/8" hose, 0-4.9V, 5Vin, $9.90ea
MPXH6400AC6U-ND - 3-58 psia, 1/8" hose, 0-4.8V, 5Vin, $9.90ea

Figured the 115's would work pre turbo, the 300's would probably work post turbine. The 400's would be needed at all the other spots except maybe pre-turbine, where they might not even be enough.

The temperature thing had me a little worried, but I'm pretty sure I'll be mounting the sensors away from 185*F heat. The 32*F is a little bit of a problem in these colder winter months. It wouldn't be hard to throw a little ambient temp sensor on the board to do corrections if it became a problem.

Cool, that would work. Just be careful when it's below 32F/0C :)
 
Nice nutcracker action there, Chris. :)
 
Wrote code to set the nozzle ring stroke limits after power up. This is what will happen when you turn your key on to start the car up.

It's easy to control the speed of the motor. It is a stepper and the step interval can be adjusted to achieve whatever travel speed is necessary, so long as the current is available. With a load, the limit for motor speed is the rated current in the motor windings or the max output current from the driver circuit (whichever is lower).

Programmable VGT Turbo Controller Setting Nozzle Ring Limits - YouTube
 
Ok first off congrats Wow :hellyeah:

I bought a couple of these years ago then traded one off to another buddy that had plans for a GN build. At the time had just sold his VR4 but now has a new project car as well and a few more over on GalantVR4.org have these in plans of one day attacking with them. We've all talked sent PM's and phone calls for hours about the best way to control these. Some of the things we've talked about and your already addressed is RPM signal, tps, boost reference etc. But we're all busy and I've been playing Jackstand baller for almost 5 years, when your building everyone else's toys hard to find time and space for your own but Its getting old and I want my car back. Its bad when you have all the parts known to man and no time or space to put them in one pile.

Now my buddy in Florida just went to the PRI show and talked to the Holset guys they're also working on a standalone controller since they've had so many emails but I think your ahead of the power curve.

The wastegate actuator type works but not to 100% efficiency to what this monster can produce. Plus being on a gate actuator during acceleration if the vanes close between shifts its going to create alot of heat and over time I believe will kill the turbo.

Other things talked about is a sweeping effect to keep the assembly clean but all I just watched in the video above you got that covered as well. We discussed having programmed maps one for track, one for road course, interstate cruising, going to get groceries etc. Don't want to be passing grandma on kill all the time and don't need to be slamming shut instant spool either. We really talked about a display like an afc where its a equalizer type control with a row of buttons for different maps, arrows to change every 250 rpm's and a display for turbo rpm, how cool would that be a turbo tach on the dash. But for tuning on the dyno along with a wideband you could achieve some serious power.

But looks like you DSM guys have already touched on all these points and I've showed up late for dinner. But I still might be able to help some after the holidays.



I work out of my basement and do alot of one off custom stuff and when I have time do runs of stuff on the mill for the Galant guys. Gauge panels with logo's, custom brackets, did the design work with Boostx for the 1G/VR4 rear subframe and mustache brace bushings, did the wing mounts and gauge panels for the 1 Lap VR4 a few years ago, cnc cut a tig welding jig to weld the center hub of a VR4/1G steering wheel to the center of an evo wheel so there all centered and perfect when welded and CNC cut about 40 sets of the Evo seat brackets so the Evo seats fit in the VR4. Before you ask the mounts are different in the DSM's than the VR4 but I do still have the jig for the steering wheel. I stay busy but the mill I have could probably knock these stepping motor mounts rather easy , just have to get some billet and write some code after I get it serviced again. I try to tear it down every few years and tighten up the ball screws so the repeatability is dead on. Maybe we can work out a deal for some controllers after your finished for some mount assemblies. That way the Galant guys I know get some controllers and the local cnc shop isn't so deep in your pockets. It could work out good for everyone.



I hardly ever sign in here, I stay busy enough on the galant board, but if you want to talk about it send me a email me through here or just send me a PM on Galantvr-4.org I'm one of the moderators over there and sign on there is curtis.


Again congrats, your doing some serious work over here in DSM land keep it up I want to feel what 40+ by 3K feels like :D:D:D
 
Curtis, thanks for the encouragement. I've put a lot of time and energy into moving this forward and feel that it's something that would make a real difference. I doubt an aftermarket controller will be made by Holset/Cummins because the aftermarket isn't their target market. In fact, it's probably the opposite of what their business model would allow. However, who knows... someone there might take notice of this project and be interested in a better controller for OEM applications (in my wildest dreams ROFL).

I also have a jackstand queen at the moment, mainly because I've sentenced her to radical modifications and have been spending a lot of my free time developing this controller instead of wrenching! I suppose in the end it will be worthwhile. Thankfully, my friend Kevin DD's his car and offered to build a manifold and install this system on his car so there will be no delays in obtaining results.

PM me if you want to talk about machining work with the gearbox housing.
 
We're getting ready to produce a pre-release of the controller electronics. Will have updates soon.

Development is expensive! Since we're just doing this for fun, we've been footing the bill- Thousands and thousands of dollars (don't tell anyone's wife :shhh:).

Well, it was BMW and Merc guys who suggested boosting the project this way anyhow. Too bad.
 
Last edited:
You'll have a big payday when you get these finished as long as you price it reasonably. Just tell her it's an investment in your future. Asking for donations to support development of a product you will receive financial gain from isn't completely legit with the site rules. Didn't you have a supporting vendor account in the past?

BTW, Keep up the good work. What you're doing is awesome.
 
You'll have a big payday when you get these finished as long as you price it reasonably. Just tell her it's an investment in your future. Asking for donations to support development of a product you will receive financial gain from isn't completely legit with the site rules. Didn't you have a supporting vendor account in the past?

BTW, Keep up the good work. What you're doing is awesome.

Forum rules :|

Never had a vendor account on here, could be a possibility for the future.
 
When your getting close to making this work I'm going to be sure to snatch up one of these turbos before the price goes up!
 
Fleece performance just released their standalone Holset VGT controller. The insane part of it is what they are asking for it. It's priced right up there with a Smarty tuner, which is 1000X more complex. As much as I condone it, someone ought to reverse engineer their product. Had they priced it at maybe $400, it might have been liveable. $300 and they would have been selling them like hotcakes for a plug and play product. I'm 99% sure they're not using CAN to operate the OEM controller.

You should try to stick with the OEM motor. The additiional motor/adapter cost is going to drive the cost of the kit up too much. Watching videos of others that have gone the route to control OEM motor, I don't see how that doesn't move quick enough. The engine isn't going to rev any faster with a road load against it, that the full range of motion can't be used. For that matter, I'd think a number of people won't even use the entire range, afterall, the A/R is huge by the time you reach full open.

Anyhow, just a thought. Keep up the good work.
 
Last edited by a moderator:
It's $625, plus a 250 to 300 dollar turbo. So for under a grand you could get the setup. Doesn't sound that bad to me.
 
Support Vendors who Support the DSM Community
Boosted Fabrication ECM Tuning ExtremePSI Fuel Injector Clinic Innovation Products Jacks Transmissions JNZ Tuning Kiggly Racing Morrison Fabrications MyMitsubishiStore.com RixRacing RockAuto RTM Racing STM Tuned

Latest posts

Build Thread Updates

Vendor Updates

Latest Classifieds

Back
Top