Ok find me one person that sets a record, or class races and is under an inducer restriction that says "ok I'll stop here I'm at the end of the map, I'm 200hp short of everyone else, but I'll stop here" Jesus a t25 is probably off the map on a stock dsm.

As for maps are for nerds, look at the HTA86, I'm not an FP nutswinger, but damn if it doesn't work. Look how many cars in the 9's on it, or even a few in the 8's, on a 61mm compressor. The lack of compressor maps clearly has not slowed anyone down with that. Meanwhile the turbo nerds are busy studying maps and can't break 11's. I will admit there is a time and a place for using the maps, but 99% of people don't have the knowledge needed to use them. That includes you and I.

You also can't read well, I never once talked about loosing efficiency with closing the vanes. I never even said I was positive that your turbo would have much effect on turbine efficiency with the nozzle opening. In fact I mentioned that I thought it wouldn't, and that in the case of my turbo i implied that closing the vanes should improve efficiency, to a point.

You just keep missing the point I'll lay some facts out for you.
1. A HX40 in a BEP doesn't spool that much slower than what you have. I'm sure the VGT can do better, but it's not like 2 fold
2. An HX40 in a bep doesn't start having back pressure issues until you get to the 30psi+ area
3. At 30psi and 8000 rpm a decently setup 4g is going to be in the 60+lb/min range. That's getting near the choke region of an HX40 compressor. The compressor efficiency is heading downward fast at that point.
4. The compressor power requirements, and the compressor rpm is increasing rapidly as well.
5. A decent turbine map will show you that as you get past peak efficiency increasing rpm lowers turbine efficiency.
6. Declining efficiency on each side means a rapidly increasing enthalpy requirement.
7. You are not going to open the nozzle your way out of increasing backpressure as you get close the the limit of the compressor

I've said it several times, the turbine side operates in choke nearly always, so I don't know how you keep getting confused. There's no reason to distinguish between compressor and turbine choke because the turbine is always choked. You also need to remember the turbo is a SYSTEM, when you operate the compressor outside of it's peak efficiency, you are also operating the turbine outside of that area too. Unless the turbo is poorly designed.

I told you to calculate this. You havn't tried, either you know I'm right, or you don't know how to. And in that case you really " have a lot to learn about radial inflow turbines"


As for my car, I hope you realize there are tons and tons of OEM wheels that garret doesn't publish the map for right? Mine was a gtx style with a small hub and all major blades. - A design that clearly trades flow capacity for efficiency(moving the surgeline to the left ), and is nothing like the 57/88 wheel out of a powerstroke, but you knew that right? Because you can calculate the performance of a setup you have no clue about?

You know why I don't publish a tech article? Because I don't know enough to do that, and if I did, I'd just a have little dickweeds trying to refute everything.

The 2nd 6-bolt is getting planned out as we speak as my controllers firmware is pretty much done and the data proves I can drop spool up by 2000 RPM so its time to move to the next step. You won't have that string to pull at much longer so enjoy it while you can.

Yeah here's more SALT. I'm sure there will be a third string to pull soon.

Keep going! I'm learning stuffz

I wish I had more, but there really isn't, likely anybody that builds turbos out side of oems pretty much cut and tries. There hand calculations but they are "back of the napkin" at best. These days it's not that bad to have a seat at a Cfd software. The problem is the guys that can actually run it and do a good job are few and far between. Here's another jab at an me education, but the one class an undergrad get in Cfd is no where near enough to do a good job, so that locks out a whole group of candidates, even a fresh Ph.D. Really isn't ready to do it unless the dissertation was in this area. That basically leaves us with senior level Cfd analysts that command $200k/yr to go along with that 50k/yr Ansys seat. Doubtful a place like fp or pte has that kind of Capitol. Far cheaper to cut and try.

Most of the stuff I deal with is big diesel shit, we get 1700bhp out of a 76mm wheel. It all cut and try. The turbo guy has a Dyno mile and 7axis. If your lucky you get the good wheel. Lol.

This spring I was able to able to goto the SAE world congress as a presenter (paper 2017-01-0633) and I was fortunate enough to see several other presenters. Talk about a humbling experience. anyway one of the papers was by a senior researcher at fomoco, on the topic of vane end clearance on the Garrett Vgt, turns out it's worth a few percentage points of efficiency. I was also able to see. A presentation on hcf failures of turbine wheels due to the shock wave created by the Vgt vanes, and how to eliminate it with vane shape.

So, what's the update on this. Some cool stuff. Wish I could just find the time. I've just been knee deep in life and floundering around, I haven't had the time like I should for my dsm's. You buy or build the adapter flange?

Morphius said:

So, what's the update on this. Some cool stuff. Wish I could just find the time. I've just been knee deep in life and floundering around, I haven't had the time like I should for my dsm's. You buy or build the adapter flange?

Click to expand...

As requested

Update:
- I ran out of summer, so I was insulating/heating my garage, then the holidays, then flu season....blah blah. Life got busy as it sometimes does, but I am still on this.
- I've refined the controller to V2.0. Spool is better than V1 but there is still a little more left on the table. I want to do a little more to improve it in 1st gear.
- I built a motor (see my build profile for deets) and got about 800 km's into the break in and ran a bad tank of gas which gummed everything up pretty bad. I have a separate thread on this as well... what a nightmare. I am going to pull the head the minute the garage has heat (it dips to -30*C for weeks at a time here regularly soooooo)
- The test session on the last tank of "good" gas I was able to push the Holset to 33 psi on 94 Octane with just a hair (0.3*) of knock....I have the logs to prove it nasayers. hahahah. I have a fast response/exposed tip EGT probes installed and at 33 psi its was just just starting to see 1000*C exhaust temps at 33 psi which is I believe is getting to hot so I am keeping it at 28-30 psi for now. Clearly I had to pull a lot of timing to keep knock away on 94oct so torque numbers (via link) were meh....but how accurate is Link anyways.
- From the same session the spool is incredible. No surge at all....it's totally dynamic now and just hugs that surge line. Before you had to tweak the wastegate to avoid surging but now the controller does it which means you can do whatever in Link and the controller will just max the spool to whatever boost you set in Link (or on your MBC). I totally changed my algorithm and it makes that compressor scream everywhere now. Spool to +25 psi is 4000-4500 RPM in 2nd and above. 1st gear spool is still at 5000-5500 rpm but without the VGT it used to not build boost until redline in 2nd gear so its still a huge improvement. I have some final changes coming that shooooould pull that down to the 4500-5K RPM range. To be honest though anything over +25 psi in 1st gear makes it pull HARD.
- The pre-turbine-backpressure to boost ratio is at or below 1:1 after the turbo is spooled....even while its spooling its near 1:1. At 25-30 psi of boost I was seeing around 22-25 psi of exhaust back pressure at ~7K RPM. The really cool thing is that the back pressure drops as the RPM's climb....just breathes better and better as you climb in RPM. 1.85 AR will do that I suppose.
- In between shifts (7.5K redline) I'm holding 15-18 psi of boost ( I do not have NLTS yet...) and within 0.4 seconds (or less) I'm well over 25 psi and climbing. As long as I shift and the next gear has me at an RPM that is over 4500 RPM the boost feels unrelenting.
- I've tested and tested and the VE table does not change when the VGT moves at all. The AFR's are all bang on target the whole way no matter what you do with the VGT. Solid setup that is tons of fun.


To move forward I need to check the health of my motor and seal up an exhaust leak the developed. I'm suuuuper close to having the V3.0 beta controller ready to go. Super simple standalone setup will be cheap to make.

@Morphius is there interest in the adapter flanges? I feel like mines kinda jenky but I could refine it and get a batch fab'd up in stainless. TBH tho I thought peeps would fab their own as it's a bit of work to relocate the radiator and there is very likely a better solution for the install to what I have done.

I will for sure offer a small batch of controllers....this thing is a too much fun not to share.

I wish they made a tranny that could take 600 ftlbs that didn't cost $25K hahahaha.


EDIT: I do have all the logs/data and will share all very quickly, just as soon I have a dialed in motor....to many peeps rag on a half tuned log no matter what the story. hahaha. I was just getting there and then the bad gas....

Something's wrong there, your ve does change with vane opening. I'll enjoy seeing how you try and show that. That's like saying switching from a .63 to a .82 on a 35r won't pick up power....

As for your back pressure, that decent but my stock he351cw was just about that good at 25psi. Get that thing up to 40 and let's see what happens.

bastarddsm said:

Something's wrong there, your ve does change with vane opening. I'll enjoy seeing how you try and show that. That's like saying switching from a .63 to a .82 on a 35r won't pick up power....

As for your back pressure, that decent but my stock he351cw was just about that good at 25psi. Get that thing up to 40 and let's see what happens.

Click to expand...

Hahhaha. OHHH... will you enjoy it? Do you enjoy anything? You are generally a Negative Nancy on the forums so I doubt you will enjoy anything I say.

And I was just waiting for the comment.....I knew as soon as I said "VE" or "backpressure" anything you'd chime in with your usual naysayer-ie-ness. Classic @bastarddsm comment.

On paper the VE does change for sure.....but how fast....and how much of that change will Link be able to pick up? It's resolution isn't that great....in fact it's shiiit.

In practice the VE table in DSMLink does not change any appreciable amount. I've ran different boost levels (upto 33 psi) across all RPM ranges (upto 7500) and throttles and my AFR's are on target. They do not vary any more than if I was running with the VGT fixed at full open. I don't know what more you want....the vanes move and the AFR's are on target. Seriously, the vanes move all over the place and my AFR's are no more than 1-3% on WBFactor when WOT so I really don't know what more to say.

To prove me wrong get a lab with equipment that could measure down to a resolution that mattered. Say every 10 RPM and Load Cells that were in the 1% or even 0.1%.....so a VE table that was 1000 x 1000 (instead of link's 20 x 20 or whatever). Is innovate's WB sensitive enough? Maybe maybe not I'm to lazy to look it up plus you get my point. Then get a datalog that has a resolution of 1 microsecond (instead of Link's 33,000 micosecond) so that we can actually see if there is a variance in the AFR's that will tell us if there is an issue worth addressing.

Why the VE changes are to small (or maybe too fast) to worry about At 4000 RPM there is a cylinder firing event every 15 milliseconds....Link can only refresh the data every 33ms (at its max rate). So Link can only refresh the data AFTER 2 cylinders have fired. That means it's "missing" 1 entire cycle in its data log....which of those 2 events do you thing Link captured? AND when during that 1 cycle did Link take its sample? That's at 4000 RPM too....at 7500RPM its 8 ms between events and ~4.125 events are happening between data refreshments....which of those 4 events do you thing Link captured? AND when during that 1 cycle did Link take its sample? My point is that even if there was appreciable VE changes (which performance wise I am not feeling) Link does not have the resolution to SEE them let alone TUNE for them.



I get where you are coming from (SAE papers?) and I understand that there is clearly going to be a VE change, but the big question you are not asking is to what degree? You are trying to get me to admit that there is all the minutia level detail in the data that needs to be addressed. When I address it by saying "I am ignoring that b/c its not big picture" it doesn't satisfy you and you keep hammering on different little details that, in all honesty, do not matter at this stage of development.

I'm not trying to squeeze every % point out of efficiency out of a setup that has been around for years for some recognition from some giant corporation that I have worked at for years. I am simply trying to get a monster turbo to spool as fast as I can way that is different yet reliable and repeatable. I have done that now and I have all the data in the world to prove it works and what someone else can expect on the same setup.

Also Your comparison was a classic strawman tactic. It's not like saying "switching from a .63 to a .82 on a 35r won't pick up power...." classic strawman tactic. Lets change the subject slightly and make it about how I'm right about that. Its more like saying "switching from a .63 to a .82 hotside on a 35r won't change your overall VE any appreciable amount as long as the intake tract is un changed."

You want to prove me wrong? Buy one of my controllers and tell me it sucks and how I am all wrong....then I can wipe my tears with your money. Hahahah.

TL;DR The setup is a beast and works as stated with data to back it up. VE unchanged during VGT movement with AFR's on target. Small turbo spool with big turbo potential. Loads of fun.

kmetiuk said:

@Morphius is there interest in the adapter flanges? I feel like mines kinda jenky but I could refine it and get a batch fab'd up in stainless. TBH tho I thought peeps would fab their own as it's a bit of work to relocate the radiator and there is very likely a better solution for the install to what I have done.

Click to expand...

Interest? Hard to say. You'd have to make a few and see if they sell. I don't see many Mitsu to Holset type adapters online. A range of T3, T25, T4 types.

What type of flange is the VE? T3 or T4 or is it some type of hybrid?

Morphius said:

Interest? Hard to say. You'd have to make a few and see if they sell. I don't see many Mitsu to Holset type adapters online. A range of T3, T25, T4 types.

What type of flange is the VE? T3 or T4 or is it some type of hybrid?

Click to expand...

It's called a T4i. It's slightly different than a T4. Not sure if it's only on the VGT Holset or not as I only have the one turbo. Not sure if I wanna get into adapters and building jigs and stuff. The electronics are easy tho as there are so many contract assembly shops these days prices are low.