Boost control with recirculated external waste gates questions
 

So long story short, I'm running my single turbo FD with a pair of 40mm external WG's (TurboSmart Gen IV Comp 40's) that recirculate back into the 3" downpipe near where the DP bends to enter the mid-pipe section. Turbo is a BW S300SX-E series, ~87mm compressor, & ~76mm turbine wheels, 1.0 A/R turbine housing, Welded stainless Turblown engineering manifold with the WG placement & orientation optimized for WG function.

When I still had the open dump tubes, and with brand new 7psi WG springs, controlling boost on WG springs alone (boost control disabled), I'd see 7~8psi boost pretty consistently, with no boost spiking or creep issues. While road tuning (Link G4+ ECU), I had set a hard boost limit of 11.5psi boost, and a "soft" boost limit of 10.2psi that is enforced by my DBW throttle, and with the open dumps, I'd never have either boost limit intervene. But I hated the noise and unwanted attention of the open dumps, so I had a fabricator recirculate them into the DP. The fabricator had to do 2 iterations of the job to get it right, story behind that & pictures of the end "v2" result as it stands now are in my build thread, post #52 here: https://www.rx7club.com/build-thread.../#post12496200

I figured this would impact boost control to some degree, given the additional back pressure behind the WG dumps that the recirculation plumbing surely imposes. But I wasn't expecting it to be as big of a difference as it is. Now with the same 7psi springs, no other changes, if I do a 3rd gear pull on a warm day, I'll see it boost up to 8~10psi consistently. But if I ramp up on the throttle a bit faster than what I typically do on a 3rd gear road tuning pull, or on a colder day, I'll see the boost quickly spike beyond 10.3psi, where the DBW soft boost limit intervenes. Same thing happens if I do a longer sustained pull in 4th or 5th gear, where instead of spiking, the boost will eventually creep beyond the soft boost limit.

Ideally I want to set up a "daily driver mode" boost profile where my max boost stays below 10psi. I'm guessing to get there with the hardware I have now, I'd need to go to a pair of lighter WG springs? The remaining options for lighter springs are 3psi or 5psi springs. Any advice on which ones to select? I'm thinking the 3psi springs would work, and if that results in boost that's a bit too low on springs alone, I can increase that with my boost control (Link G4+ has both open & closed loop boost control options, and I can do multiple boost/DBW profiles).

@Pete_89T2 Any updates?
I’m thinking it’s cooler weather contributing to the creep.
Did you change any timing, AFRs or boost control.
Can you check logs from prior to wastegate rerouting that you could compare?
ie.. air intake temps and such…

Steve

Yeah Steve, I do have logs comparing the boost response with the new 7psi springs before & after the recirculation plumbing job. Colder weather is definitely making the boost spiking & creep worse than it was before. No changes to timing or target AFRs, and boost control is off - this is just running on springs to characterize boost response on springs alone.

Before the recirculation job with the 7psi springs, I would see boost repeatably level off at no more than 8~9psi MGP under all conditions with the open dump tubes. Didn't matter what gear I'm pulling in, or how quick/hard I ramped up the throttle. Unfortunately I don't have much logging with that configuration in colder weather to see how much temperature is affecting boost response.

After the recirculation job, in warmer weather, it will boost up to 9~10psi on my typical 3rd gear road tuning pulls. Same air temps, if I do a sustained pull in 4th or 5th gear, boost will eventually creep past my 10.3psi MGP set point where the DBW soft boost limit pulls it back

In colder weather, after the recirculation job, if I hit the throttle harder on a 3rd gear pull (i.e., steeper delta TPS curve), it will boost spike past the 10.3psi MGP soft boost limit. Similarly, if I do a sustained pull in 4th/5th gear, the boost creep past the 10.3psi mark will happen sooner than it would when compared to warmer IAT conditions.

If I switch to the 3psi springs, I figure that will solve the boost spike/creep issue and net me a "low boost/daily driver" mode where I can limit the boost under all situations to <10psi. But I'm concerned that when I want to set up for higher boost, say 12~15psi MGP, the 3psi springs will want to crack themselves open, and I'd need to run very high duty cycles from the MAC valve boost solenoid to get & keep the boost up that high, which could make things hard to control/manage. OTOH, I'm not sure how much better the 5psi springs would be compared to the current 7psi springs for solving boost creep/spiking at the low end.

Have you considered a 4-port boost control solenoid? You can use a lower psi spring and still get higher boost if needed.

I might have rushed and missed it, but going back through both your and the previous owners build threads I didn’t see any pics that actually detail the wastegate bias and positioning on the manifold. However, if the photo below of the downstream wastegate piping merging into the main exhaust pipe completely perpendicular to the main exhaust flow is what’s on there now, then that’s not helping.

https://cimg7.ibsrv.net/gimg/www.rx7...e4eead8bf.jpeg


My suspicion is that the entire thing likely needs to be repiped to strongly bias flow to the wastegates and then merged in smoothly downstream into the main pipe as well. That assumes the WGs are functioning properly both mechanically & controls, also the space and clearance to pull it off as needed.

Having 90° connections on any of those exit-entrance points is all going to work against what you’re trying to achieve. Because expecting flow to divert near or at perpendicular into a hole is the equivalent of it being a much smaller size. The exhaust stream can’t just immediately divert and turn sharply into a hole on the side. It’s trying to, but can only turn so much before slamming into the back side of the WG entrance. So all the chaotic turbulence that results there then effectively chokes itself off.

I’m guessing at that without pics though, but it seems to be a constant theme on many manifolds here. So then a similar situation exists where those WG discharge pipes merge at 90° to the main flow. Those need a long angle merging smoothly all in the same flow direction together. You can somewhat get away with sharper flow angles more readily running high boost, but not low boost. Without pressure bias it then needs flow bias.
.

https://imagizer.imageshack.com/v2/1...922/BXrNak.jpg

Congrats on choosing a Link! i will bet you love it..

WG engineering is not to be approached casually, especially if you are wanting to run a larger turbo at low boost. your turbo can make 75 pounds of air at one bar. assuming you are not wanting to make 500 rwhp at 15 psi your WGs will be VERY busy. unfortunately they will be busy not flowing the proper amount of bypass as both WG runners are situated so they come to a stop at the DP. you need to visit the welder a third time. it will pay you big dividends.

while my WG is purposely disadvantaged, so my turbo gets a better driving flow, the blend into my DP is such that i can run modest boost levels if i wish. as you know the "flow system" starts at the air filter and continues all the way to the tailpipe. anything restrictive after the combustion adds to egts, adds to higher IAT, adds to increased chance of pre-ignition etc. so there are numerous reasons to go the extra mile as to WG blend.

this is the DP i have run since 2013. i recently switched to a G40-1150. the back end is 4 inches so i had to do a bit of modding which was in process at the time of the pic..

another example from here https://www.rx7club.com/build-thread...hread-1130035/


https://cimg2.ibsrv.net/gimg/www.rx7...0369f9f5a8.jpg

Not yet, but that's a possibility assuming a set of lower pressure springs on their own can get low boost modes managed well enough.

I suspect and I'm afraid you're correct - those 2x 90* bends into the DP are certainly not helping my boost control situation. The design & flow out of the waste gates looks pretty good to me up to that point, at least as good as it can be done with dual WG's going into a 3" DP given the available space to work with on the FD. Here's a picture of my engine sitting on the stand, oil pan up, where you can kind of see how the manifold/WG installation looks on the manifold end - manifold has the gates oriented & angled for optimized WG flow, and they did work great when I was still running open dumps through the short 90* bend that you see there.


https://cimg2.ibsrv.net/gimg/www.rx7...d3b645d155.jpg


Your WG setup looks great Howard, and one big single WG in lieu of dual gates certainly would make the recirc plumbing easier. And I am loving my Link G4+ ECU, when used with a DBW throttle, it's been easy to prevent over boosting from blowing my motor. Pretty easy to setup a soft boost limit with the DBW to intervene if your WGs are not bleeding off boost as well as they should.

So I guess the question now is, short of spending lots of cash on a better recirculation job, that will probably also require me to get a larger diameter DP/resonated mid pipe to accomplish, would swapping to a set of new 3psi springs be worth a try? As long as they end up getting my boost to some level of CONSISTENT control (on springs alone), say anything between 9~11psi, I figure I can avoid the expense of another big fabrication job. Higher boost levels can be managed by the Link's boost control, using open or closed loop modes, and if necessary I can switch to a 4-port MAC valve boost solenoid. Sound like a plan?

plan B, just run more boost. :P

This :nod:
Im running 15-16 psi all day every day... well that was before I tore apart a perfectly good running FD :o:

Steve

Actually running more boost isn't such a bad idea - as long as the setup I have now can manage boost consistently, I can keep using the DBW imposed soft boost limit to keep my keg happy for a long time and treat that as my "daily driver" mode. That has been working very consistently & reliably - when it hits my threshold, the throttle immediately pulls back to 1/2 of whatever my right foot is demanding; then as soon as the boost drops about 1psi or so less than the threshold, the throttle goes right back to 100% of whatever my right foot was demanding. It's so quick, that I just notice the hiccup in power delivery from behind the wheel before I even have a chance to react to it. In the logs the whole process elapses in less than 1 second typically, and the accel pedal position doesn't change a bit, indicating it happens too fast for me to react to it :rofl:

what fuel are you running. AI?

i don't have any experience w low spring presssure WG springs. FWIW, Evans is not a fan of 4 port solenoids. at some point i would tune up your blend. you can have a really good setup flow-wise but the net is it is only as strong as the weakest link. restrictions cause higher egts, IATs and pre-ignition.

🤣

If the WGs are angled right off the port exit elbows (pic was a bit off-angle) then that’s as good as it gets and congrats on that part being one less thing to mess with. You may need a larger or more freer flowing exhaust system overall as well with all the flow directed through it now, but a better WG discharge pipe merge angle into the main exhaust tube and some more boost to balance the pressure-flow bias ratio might get ‘er done for you.

I'm running 93 octane pump gas. E85 is pretty much unobtainable locally so that's not an option, though I do have a ethanol sensor built into my fuel system to support it. No AI yet, though I'm planning on getting a basic AEM V3 water/meth injection kit that I'll use primarily to keep my IATs in check under boost and keep my engine internals clean. Basically use it as an add-on reliability enhancer, and will just integrate some fail-safes from it to my Link G4+.

I recall how Evans hates 4-port solenoids. I kind of breezed thru another video of his where he goes into using a pair of 3-port solenoids to essentially get you the same level of control that a 4 port valve can offer, but also gives you a wider range and more control resolution. Down side IIRC is you'll need to set up a 2nd boost control output on your ECU to drive the 2nd 3-port solenoid make that work - will need to revisit that video.

Yeah, the manifold has the WG placement about as good as it can be - right off the ports and with very little angle, almost a straight shot out. That's why it worked so good when I was running open dump tubes - wish I could tolerate all the boy racer noises!

Since a pair of new 3psi springs is only about a $60 investment and a little wrench time, I think I'll give that a whirl and see how it turns out. But you're right - the proper fix would be a less restrictive exhaust & dump path plumbing.

Was your WG design driven from a packaging standpoint? How would your turbo get any worse driving flow of it had the WG pipe come off the merge at a better angle to allow for a smoother transition of flow into the WG? Right now it seems your WG will struggle with control since exhaust glasses have to back up and make almost a 180* turn to the gate.

​​​

he runs higher boost for standing 1/2 and 1 mile top end speed events, but even though the WG exit angle seems extreme, the angle cut itself provides a wider opening with it’s placement on the upward bend, plus with it being on the backside of the upward bend the flow slams into it creating bias. It’s a pretty clever design really, though I can see how it might be deceiving in the regard you mentioned. So ultimately the situation has to viewed and assessed on it’s own merits rather than rely on generalities that may or may not apply.
.

https://imagizer.imageshack.com/v2/1...923/szgwzf.png

i am happy with my WG. no boost creep w a large turbo around 15/16.

as you know there are many factors that effect WG flow. i have a section on my site entitled "System Design."

SYSTEM DESIGN

my 60 mm WG has an area of 4.38 sq inches. OP has 2 40 mm which have a total of 3.89 sq inches as well as more wall area/drag.

as @TeamRX8 said, looks deceiving but flow to the WG is ok. It would be fun to see CFD of various manifold designs!

https://cimg5.ibsrv.net/gimg/www.rx7...4349b82529.png

A lower spring rate won't solve boost creep. What it will do is delay boost response, but you'll still get creep @ approx the same rpm you currently do, maybe a little later.

Rerouting the recirc tubes is your best chance at getting what you want. Or choke the intake side of the engine.

IWG is a tad different, but I imagine the principal is still the same. I wanted 14 psi boost and the only way to fix it was porting my WG larger. Before doing that, I wired wastegate fully open and it still hit the same boost threshold that it was creeping to.

Routing recirculated dual gates has been kicking my ass for a while. This is my latest iteration. No problem controlling low boost (9psi spring up to 15 with EBC) but the couple tiny exhaust leaks drive me nuts. I'll probably redo this again over winter with 321 stainless bellows instead of the spiral stuff.
I'm running a 0.91AR S360 with pretty aggressive early opening exhaust porting to promote spool. My porting might actually be helping boost control. I have trouble keeping my gates from blowing open as boost builds and the the initial blow down is going to be at a higher pressure than stock porting.
https://cimg1.ibsrv.net/gimg/www.rx7...d6df931026.jpg
https://cimg8.ibsrv.net/gimg/www.rx7...2edf19b210.jpg

WG flow is limited by the valve flow area. For a poppet valve that's measured in curtain flow area. 2x 40mm wins by a large margin.
That's not to say the single 60mm isn't better though. Each 40mm is only flowing for 1 rotor, the merged (smei-divided) 60mm get's to flow for each rotors alternating pulsed flow. If I was starting from scratch again, and I might, I'd go this route.

Slightly off topic but so thrilled to see that you have a website, just started reading...this will be very enjoyable!! Thank you!

off topic but...
is it advisable to start tuning with a low psi spring and increase boost with a controller and then switch to a higher base spring? I know it's a lot of work... But say for peace of mind and safety you start with a 5 psi spring then end up with a low boost spring of 14 and a high boost controller setting of 20+. I'm nervous boosting a basemap tune.

^ I like how you/your fabricator has the WG pipes entering the down pipe at about a 45* instead of the 90* entry that my fabricator guy did. You're doing exactly what I'm trying to achieve -- 9ish psi on springs alone, up to about 15ish psi with open/closed loop boost control enabled. My motor is all OEM Mazda though, no intake/exhaust porting done, BW S300SX-E series turbo, ~87mm compressor, & ~76mm turbine wheels, 1.0 A/R. Surprisingly this setup spools up easily, and I'll see positive boost as low as 2000~2500 RPM in the higher gears.

^That is precisely what I'm doing with my DBW throttle at the moment to "control" the lowest boost possible with my setup. I'd rather have lowest possible boost level controlled by my WG springs alone, hence why I made the original post. So the plan now is to take advantage of some black Friday sales and pick up some 3psi and maybe 5psi springs, swap 'em in and do some road testing to characterize boost control response on springs alone.

If none of these springs (existing 7psi, new 3psi or 5psi) can net me a consistent & repeatable boost response that keeps my boost at <11psi or so on spring control alone with the current exhaust/WG plumbing, then Plan B for a low boost setting might be to limit boost on the intake side via the DBW throttle.

Run dual 44s here, 7lb springs, dubious of comparing flows across brands, but very similar size turbo. It's been tuned to be capable of running at that minimum pressure for wet events and is consistent going by dyno print.

If that existing pipework you have is 1 1/2" off the WG, that DP(?) doesn't look particularly large, and with both joining at near the same point normal to flow probably less than ideal when combined area is not much less. I also see that Canadian car mentioned in the earlier link now has a 90 degree entry on it's latest iteration too. Plumb back here is in the 35 and 75 degree range from memory.

One consideration, the tighter you get on entry the more difficult it is to weld the inside corners, hail marys required sometimes and possibly part of the reason the mig was wheeled out in that first abomination.