Has anyone gone from 60 mm TB BACK to stock?

[Tony D]

It's not a Pop Off Valve, as it's not something that opens to relieve pressure. It's not an unloading valve perse, because we are not unloading it. Really BOV is describing it's function, and in both cases it does exactly the same thing: keep the turbo from surging by relieving (blowing off) excess air to prevent the turbo compressor wheel from going into a low-flow surge event. It stabilizes minium flow rates through the compressor as it's primary function (blowing off excess flow) and at the same time allows shaping of the intake manifold pressure curve dependent on speed, throttle position, or whatever other variables you would choose to corporate. Traction control is a great example. If you had a small turbine housing you could keep the wastegate closed, and run some pretty high boost pressures at very low engine speeds. You blow off the excess to prevent a surge at that low a flow, but the engine doesn't care that you have 12psi at 1000 rpms, and are blowing off 75% of the air produced by the turbo---it only knows it's got all the air it needs to have 12psi in the inlet manifold and it produces power accordingly.

 

That is the hardest thing for people to grasp---you will launch harder (if you have traction available) being able to launch at 12psi and run a midrange psi of 20 rather than the conventional thinking of getting 30psi but only after 3400 rpms or whatever. The 'area under the curve' is huge compared to conventional wastegate only control.

 

And it's great I checked in this morning, that means I got to get the manuals form Bill today as I leave tomorrow from here!

[Gollum]

Thanks for the clarification Tony. That's along the lines that I was thinking as to why it would still be referred to as a BOV.

 

Next question, or ponder. I'd assume there wouldn't be anything wrong with venting the air back into the inlet of the turbo, right? If anything I'd assume that'd actually be a GOOD thing helping the turbo to continue to spool easily and prevent compressor surge.

 

So knowing what you know, do you foresee any issues getting megasquirt to control a system like this? I personally don't see any issues, just concerned I'll have to write a new version with the code for the loop.

 

I must say I think I grasp that area under the curve is much better and important than peak numbers and making power where your gears can't utilize it. Obviously you still need some raw HP figures otherwise you have a torque only big rig that's mostly just useful for towing things and needs tons of gears to make use of that power. But we should all know and understand this. But that being said I'd rather have a 225hp V8 that makes peak torque more than 3k rpm bellow peak HP, than a 240hp NA 4 cylinder that makes peak torque about 2k before peak HP.

[Tony D]

In this case, venting to atmosphere would be all that is required. Remember the wastegate is now controlling to peak efficiency speed and will only open when the overspeed of the wheel is looking like it would occur.

 

Generally, venting to the lowest pressure place would be the best, and that would be the plenum under the air cleaner to blow backwards through the filter to dislodge debris. Blowing into the turbine won't really keep the speed up, the wastegate isn't controlling the boost any longer under this control scheme. The wastegate controls turbine speed, and the BOV is now the effective pressure-control valve!

 

 

On an aside, there was a device called the "Miller Woods Turbo Group Fueler" which had a fuzzy logic control of the wastegate to severely overboost the engine at lower engine speeds, using excess capacity of the fuel system to supply fuel at lower engine speeds. As the engine speeded up, the fueler tapered boost back so the stock injectors could make the horsepower requirements. Their philosophy was that if you can supply fuel for 200HP at 6000 rpms, then you can DOUBLE your torque at 3000 rpms by boosting higher at 3000 than you can at 6000. Really amplifying the low-speeed response of the vehicle. They basically douobled the boost and the injector pulsewith was moved to 'full load/full rpm' type of mapping at the lower engine speeds, porportionally adjusted depending on what kind of response you had in mind.

 

Great profit for them...Not a lot inside, but the chips.....ooooh, the chips!

[Gollum]

So let's see if I understand this... at least kind of, as it comes to application.

 

The more I've looked at compressor maps over the years the more I think I don't quite understand them completely. I think I grasp what it's displaying, but it seems like in application your load lines your draw won't be followed. They seem more like just a guess to where you'll end up.

 

So according to this map/load line here:

Compressor Picture

 

 

Based on this config here (which might not be accurate volumetric efficiency for a relatively stock L28ET

Config

 

 

Now, according to the map, if the wastegate is fully closed I'll make 17psi by 2k. Sounds reasonable. But because these load lines are figured for wastegate control on the hot side, it says i'll be going way past the efficiency peak of the turbo by redline. BUT, if I'm controlling excess then I can make the turbo follow the "island" in the middle of the map, increasing RPM and pressure ratio as I reach redline, then vent the excess on the cold side, right?

 

All that being said, that's WAY more turbo than you'd need for a meek 350hp on a stock L28ET. People have made 300 wheel on stock T3's, showing that a GT35 should be more than capable to blow up the stock bottom end.

[Tony D]

Well, you would not want to overspeed the wheel. You would keep the turbine controlled via wastegate to an optimum speed, and vent the excess flow to keep the compressor side in the stable flow area of the map (right of the surge curve).

 

So yes, normally you couldn't apply this and get those results because you would overspeed the turbo and would need humongo wastegates to bypass at redline.

 

You still control to wheel speed and then move the load point over right by venting the BOV and that would keep you in the 76% efficiency island. Keeping wheel speed higher gives you access to the higher pressure ratios in the longest island, but you have to have the flow to do it...the bov makes your 'consumption' off the turbine side similar to the larger engine the compressor wheel would normally be be applied.

 

ACTUALLY, the instigator of this whole thing, was looking over my shouldere and said "why only 6500rpms" but he assures me his PID BOV controller manuals are with him on his flash drive ready to transfer to me for digestion. I just reminded him in no kind terms I've been hunting that information for some time now... GIVE IT UP DUDE!

Edited October 29, 2010 by Tony D

[Gollum]

What I'm still not quite getting though tony, is that isn't the optimal speed for a turbo variable based upon the pressure ratio it's up against and also the air that's actually flowing? I thought that was kinda the point of a compressor map, so you can see if you engine will land the turbo in an efficient range.

 

It does seem though that in theory by running high boost at the bottom and then tapering it back down at the top you actually STAY in the peak area of efficiency of the turbo easier.

[Tony D]

Yes, but that's a different control loop than simple Tachometer Based wastegate control.

That would use multiple variables in addition to engine speed.

 

The map presented can be modeled in a 3D controller to vary a two variable output controller to move the turbine speed, and then blow off excess in minimum flow situations.

 

But it's terribly more complex than simply maximizing speed on the turbine and moving the flow point right based on pressure.

 

A compressor wheel could be cut to optimize this kind of control scheme. Right now they are generally compromized knowing they poor overall control the wastegate provides, and that the flow demands of the engine will rise as speed rises so the maps look like bananas to mangoes. If it was mangoes and bigger mangoes, control would be easier!

[Gollum]

Yea I was thinking that a compressor with a map that looked more like a squat pumpkin would be much easier to use in a control scheme like this.

 

From what I'm understanding, it wouldn't be hard at all to program in a map to do this on megasquirt. It's just a basic table, same as spark or fuel. You have to big variables, speed and speed (opposed to speed and load for fuel) and a map that tells the ECU where to aim the wastegate. The complicated part is that you need TWO new tables. One for Wastegate control of the turbo and one for BOV control. I'm not exactly sure MS has enough room for that, I'll look into it though. As stated before, might be possible with some custom scripting done to the source code.

 

Oh, and I'd been thinking about this... the BOV that's venting for a target PSI, being control by the ECU, can it also act as the BOV for when the throttle plate gets closed during shifts and coast down? I'd think you'd need a separate BOV just controlled by the manifold vacuum.

[Tony D]

Actually it's venting the BOV portion based on minimum flow per psi (programmed in 'surge line slope' in industrial installations this is normally expressed in amps per psi...here it would be based on likely engine rpm and psi with some sort of modifier from the TPS signal as an 'overlord' which dumps pressure when rate-of-change indicates it is necessary.)

 

 

The thing is to have ONLY ONE BOV. The throttle drop BOV function is not necessary when the control being discussed is enabled. The reason for a 'drop throttle bov' is to stop minimum flow surge which stalls the compressor wheel and attempts to minimize turbine shaft speed drop on shifts. With this control of the wastegate, the drop will not be an issue, and by venting excess air based on surge point the surge issue ceases to be an issue.

[Gollum]

I'm having ah ha moments, while creating more questions/problems for myself. I feel like I don't understand turbos at all anymore. It's one thing to understand the concept of how they work, it's another entirely to KNOW what's going to happens in a certain situation.

 

 

Load created Compressor Surge = When the engine is requesting more air than the compressor is actually able to give, correct? This is why if your compressor is too large for your application at lower RPM if you full throttle it your compressor isn't up to speed yet and you hit the surge limit. The engine is suddenly drawing more air than the turbo has to give, now you have flow separation.

 

Throttle change Compressor Surge = When the throttle is rapidly shut under boost conditions, causing air to reverse and force air back to the compressor, causing the compressor to be pushing against an "unmovable object" if you're thinking about it in fluid dynamics (I think)

 

 

Now, I'm understanding (I think) why you don't need a second BOV. I believe this is because your BOV won't suddenly "close" or "open" just because you closed the throttle. But if you're at a peak PSI for a given RPM and you slam the throttle shut, the PSI will increase as your RPM's are dropping, and this will throw your BOV into an area of the map that it'll open anyways... correct?

 

And for surge line following.... I'm thinking that in order to prevent surge, accord to how it's programmed, you'd be OPENING the BOV to let air IN if you're in surge terretory. IE: You're at 1000rpm in 5th gear and you opened the throttle to WOT. This would allow air into the engine bypassing the turbo acting as a restriction at this point, and then using that exhausted air to spool the turbo. Once the turbo is into non-surge area it will be creating PSI and the BOV will close to follow your requested PSI/RPM data.

 

 

Looking back at the dyno you posted in the other thread, I believe they could have had full PSI much sooner, but they've tapered it the way they have in order to make the power more usable and come on smoother.

[Tony D]

I will take this in the order presented, and suggest 'What is Surge' sticky elsewhere... or maybe I should assume you read that already and is where you drew your conclusions?

Anyway, here goes:

 

Load created Compressor Surge = When the engine is requesting more air than the compressor is actually able to give, correct?

No, the situation you describe is 'Stonewall' and is to the lower right portion of the curve moving vertically paralell to the bottom horizontal axis, away from the surge line. The compressor will not surge in this situation, but will merely not produce boost. Air demand is not 'load', load on a compressor is ALWAYS flow+pressure. The more pressure, the lower the flow tolerated, the lower the pressure, the more flow will be generated till you hit Stonewall.

This is why if your compressor is too large for your application at lower RPM if you full throttle it your compressor isn't up to speed yet and you hit the surge limit. The engine is suddenly drawing more air than the turbo has to give, now you have flow separation.No, what is happening in that situation is the engine is INCAPABLE OF ACCEPTING the flow generated from the huge compressor, and as a result the pressure is too high for the stable minimum flow requirements, and the compressor surges. This would occur in the lower left of the compressor map rising vertically paralell to the vertical axis. In this instance, keeping the wastegate closed to speed up the compressor, and opening the BOV to induce more flow through the compressor would stabilize the flow and move the load point to the right away from surge. Pressure is vertical, flow is horizontal. Low flow moves you left, high flow right. Low pressure moves you down, high pressure moves you up. The speed intersecting curves throw these lines to a curve, but if you speed it up the compressor is capable of more pressure and more flow, so if you then VENT through the BOV overboard at that speed you don't move VERTICALLY on the axis and risk hitting the surge line, you move HORIZONTALLY to the right AWAY from the surge line. All commercial compressors will always VENT AIR when close to surge---inducing flow will ALWAYS move you out of a surge condition as it usually instantly moves the load point on the graph to the right, and usually down.

 

Throttle change Compressor Surge = When the throttle is rapidly shut under boost conditions, causing air to reverse and force air back to the compressor, causing the compressor to be pushing against an "unmovable object" if you're thinking about it in fluid dynamics (I think)Again, the mechanics of what happens when the throttle is suddenly closed is that 1) pressure rises -- movement vertically UP on the graph, & 2) flow drops or stops -- movement horizontally on the graph to the LEFT. You see that the combination of UP/LEFT puts you into the surge line quite quickly. In EVERY surge situation it's ALWAYS a movement UP/LEFT that will put you into surge. Some cases it's simply that the pressure the compressor is operating at is simply too high for the flow it generates (lower left corner of the map). When you rapidly shut the throttles your BOV should open to induce flow and keep the pressure from rising (move it straight to the right away from surge line) With the controlled BOV you would only open the BOV enough to add flow, but KEEP PRESSURE---a true horizontal right movement. Most BOV's now when they lift over-vent and as a result move to the right, as well as down vertically as the pressure drops. In either case, you move away from surge.

 

 

Now, I'm understanding (I think) why you don't need a second BOV. I believe this is because your BOV won't suddenly "close" or "open" just because you closed the throttle.

On a conventional BOV It WILL. A proper BOV would open at the SLIGHTEST change of throttle position to allow the load point to move along a 'flat line' to the right, keeping a constant pressure (vertical orientation) and keeping the turbo away from surge. For a PID controlled BOV when the throttle was quickly closed, the BOV will OPEN and move the load point right. When the throttle is suddenly OPENED you would see the load point move vertically down, but the BOV at that point would either already be closed, or SLOSE in phase to move the load point up (maintaining pressure) and depending on where you are on the map, may stay open to keep the load point away from surge by venting air and inducing flow (movement right along the horizontal axis).

But if you're at a peak PSI for a given RPM and you slam the throttle shut, the PSI will increase as your RPM's are dropping, and this will throw your BOV into an area of the map that it'll open anyways... correct?

This is only possible with the PID controlled BOV---and is why I mentioned it. A conventional BOV will open under those conditions because it's ported to the manifold and when the manifold goes under vacuum it will overide the spring and open as it's signalling the BOV that the throttle is closed. During normal movements of the throttle the BOV if it is a simple spring type, and not ported to the manifold would do nothing and is totally pressure based. That is why it would surge in the first example you gave---it only knows 25psi, and won't lift until then.

 

And for surge line following.... I'm thinking that in order to prevent surge, accord to how it's programmed, you'd be OPENING the BOV to let air IN if you're in surge terretory. IE: You're at 1000rpm in 5th gear and you opened the throttle to WOT. This would allow air into the engine bypassing the turbo acting as a restriction at this point, and then using that exhausted air to spool the turbo.

Nope, now you are talking 'compressor bypass valve' and a different function altogether. That would be for N/A operation. A BOV is PURELY a venting device, and not a bypass around the compressor. A PROPER conventional BOV would act like this if ported to the manifold---and ducted to the intake tract between the air filter and turbo inlet. Most aren't like that. In that case, before boost threshold, the compressor bypass valve is open to remove load on the turbine wheel to let it speed up quicker. Once any boost is present the Compressor Bypass Valve (which is really what they should have...) will CLOSE, and if ANY change in throttle position is affected, will open to either vent pressure and bypass the turbo. The old Cartech Systems have VERY GOOD BOV/Compressor Bypass Valves. If there was one thing Corky Bell did right was make a damned good BOV/Bypass Valve. They DO NOT sound like BOV's today---they are 'sigh' valves. If you hear the old original Wangan Midnight S30, you can hear the Bypass Valve sighing all the time. Lifting to keep ANY pressure rise from happening in the intake tract on even the slightest lift-throttle situation.

 

Once the turbo is into non-surge area it will be creating PSI and the BOV will close to follow your requested PSI/RPM data.As explained above that is proper Compressor Bypass/BOV operation. That is NOT how 99% of them on the market today operate. They just dump pressure on drop throttle. They do not perform the bypass function. And that was the direction I was going, you have to give up the bypass function with a PID controlled BOV/Wastegate secenario. But this is not bad, since the paradigm for control is totally different. The bypass valve is predicated on parasitic loss reduction and bypassing the turbo to let it spool. That is because the wastegate is open in this situation and the turbine speed is slowing. When the wastegate is controlled to STAY CLOSED and ONLY open upon reaching an optimum speed, then ALL control can be done off venting! You would use a smaller hot-side A/R than on a conventional wastegate setup as you look to generate speed at the LOWEST possible rpm, and then control air on it's own. You can always bypass more exhaust gas if your A/R is too small, but it's impossible to generate full boost at off-idle conditions when you conventioanlly size the hot side A/R to handle peak RPM flow through the exhaust turbine. Look at Big Phils GT35R, he went with a .82 hotside to stop a low rpm surge issue when he had the .63 A/R hotside. The also lost boost response down low. With the PID controller, you don't loose it down low! You keep insanely low boost threshold (say 1500 rpms) and still can use a turbo that flows big air for top end feeding of the engine. If you need more exhaust area, open the wastegate(s)!

 

You did read 'what is surge' right? A lot of this sounds familiar to me...

 

 

Looking back at the dyno you posted in the other thread, I believe they could have had full PSI much sooner, but they've tapered it the way they have in order to make the power more usable and come on smoother.

Yes, it's for drivability, absolutely! And you can then see this is not really a 'new' idea, just one not widely applied because of the tuning complexity. I ran a T3/4Hybrid on my car with a 0.48 A/R. I could generate 20psi at 1700rpms. I also had 265's out back... And to keep this compressor working when stabbing and lifting at 2000-3000rpms I had to have VERY responsive BOV. This would be an 'abberant' sizing for the A/R, but my compressor was not THAT oversized. It would natural surge at 25psi, so 20 was my limit. But now with this kind of control, I could run that SAME 0.48 A/R on the hot side, coupled to a GT35R wheel to produce 25psi at 1500rpms (ball bearings baby, gotta love em!) and simply vent all that extra flow overboard to keep from low-flow surging below the point where the engine demands could take stable flow off the turbine wheel. This was Phils issue, surge in the midrange, but not on the top end. By having a BOV vent during this period, you would MAINTAIN the 25psi for power production (the turbo can make the air, the reason it's surging is because you are too far LEFT on the curve for your vertical point) and not do what the 'fuzzy logic' controllers do (boost per rpm)---they would drop your psi at that point to move you vertically away from the surge line, then raise it afterwards. This gives a power dip, whereas by simply venting excess air and stabilizing flow you maintain the power curve from 25psi! Turbochargers are odd in that you have variable speed wheels and various rise to surge points. By limiting wheel speed to an optimum point, control becomes easier as the points you have to worry about decrease. Natural surge point is the pressure at which the compressor won't produce any more pressure---that is for a given speed. Same as low-flow surge, for a given speed there is only so little flow the wheel will tolerate, it can be looked at as 'natural surge' in that respect, because at that point it can't make any more pressure. What the control system must do is always make sure the compressor can flow enough air across it to keep stable flow. I digress...

Edited November 3, 2010 by Tony D

[Gollum]

You know Tony, I'm VERY SORRY, but I hadn't read that sticky... Before that last post I'd just read 3 articles on surge, and you explained it better than all three COMBINED!!! Serious lack of UNDERSTANDING out there... One article was even from garrett's website and was just plain poorly written and not expansive enough.

 

 

I see now that I was thinking of it reversed. Thank you.

 

 

I can understand how people (and OEM'S) have stuck with a pure wastegate control for simplicity, but I'm seeing that now as a very sad reality. We have companies like nissan and BMW with throttles that are only used in warm up and then total valve timing control for "throttle change". This is only about a thousand times more complex than the BOV turbo pressure control we're talking about here. I'm also kinda lost as to why the OEM's don't do this since it would allow them to use a SMALLER hot side allowing some cost saving on materials, and also a smaller package. Seems like everything they're normally after.

[Gollum]

Back to maps...

 

So, in theory, with a system that's controlling flow through BOV control, can we program the BOV to not just blow off enough to prevent surge, but allow enough flow to let the compressor follow it's most effient path? (as long as pressure ratio request are met according the the PSI/RPM targets.

 

Because we can set a target load line like this:

 

http://www.squirrelpf.com/turbocalc/graph.php?version=4&pr0=1&pr1=1.07&pr2=1.78&pr3=2.5&pr4=2.5&pr5=2.5&pr6=2.5&pr7=2.5&airflow0=3.1&airflow1=3.1&airflow2=8.2&airflow3=15.7&airflow4=29.7&airflow5=42.4&airflow6=44.6&airflow7=46.2&product_id=42

 

So for this plot, it looks like 98,000rpm would be a good limit for the turbine. Then the BOV will be opening enough to make up the difference between the load line and where we want the compressor to end up. So say at the 1.5 pressure ratio, the engine would be at an estimated 6 lb/min of flow (for theory sake, i don't actually know), and then the BOV would be flowing around another 6lb/m to get past the surge line. What I'm still not quite understand is where the PSI the engine will see is actually going to end up. The engine will only "breathe" so much, hence the surge issues, but I'm assuming that if we vent JUST ENOUGH to bring the turbo out of surge range, then that will give us the MOST available PSI for the engine. If we were to vent MORE than needed to get into operable range for the compressor, then we'd be sacrificing more air than needed, thus also available usable PSI for the engine, right?

[Tony D]

Generally yes.

 

You are now confusing flow with PSI tho... They are interrelated, but not directly as you would think. You don't pick a pressure ratio, IT PICKS YOU! Follow me here:

 

What you need to know is power needed/desired at that point---if you want 200HP you need the airflow in pounds per hour to support that horsepower.

 

You need to know the VE of the engine at that point, so you can calculate the perfect capability of the engine breathing.

 

Then, the disparity between what the turbo has to flow, and what the engine CAN flow will result in the PSI present in the manifold and therefore pressure-ratio you will need to run to accomplish your power goals.

 

You are looking at it from the back end of the equation.

 

First you need to determine horsepower, then determine fueling and air requirements, then actual engine efficiency, and where you would be on whatever map you choose.

 

The engine will only flow so much, the disparity between what it will flow and the flow required to make the power you want determines the PSI needed to reach that goal. THEN that pressure ratio will be charted on the map, and speed/flow balances can be determined. Personally I would shoot for the lower speeds at first if possible to move to the centermost portion of the map, and then where the map has the 'dogleg' on the left side, UP the turbine speed to the 110,721 to get the advantages of the higher pressure ratios. Really though the charted line overshoots the 'sweet spot' of the centermost island of efficiency. That is really where you want to operate the compressor.

Edited November 4, 2010 by Tony D

[Gollum]

I'm pretty sure I understood all that, so I'm wondering if I'm just not phrasing things correctly.

 

I realize that pressure ratio/psi AND flow will be direct variables depending upon engine displacement, volumetric efficiency and power goals. If you have a large efficient engine it won't need as much PSI as a smaller less efficient engine to reach the same power.

 

IE: Even a POOR efficiency ford 5.0 stock only needs a quarter the PSI to reach 400hp as a high efficiency honda 1.6 liter. But by the same token the 4.6 ford DOHC needs LESS PSI than the 5.0 to reach the same power, due to being a more efficient engine. Three engines, same power, very different pressure ratios/PSI requirements.

 

I guess my point is that it seems as though YOUR point is that this system allows you to run a much larger compressor/smaller turbine (much more offset than the typical T3/T4). And I'm saying that by what I'm seeing on load calculations that you'll be well into surge territory under normal wastegate control (duh, I think we've covered that, thanks for your patience with me ). So in this system you dump the excess air the compressor is making at lower RPM (or any for that matter) to get it back into stable region of the map, but what I'm asking is "won't that be a low efficiency area of the map?" and "would you want to vent MORE air to create MORE flow through the compressor to FORCE it into the better region of the map?"

 

I think that make sense. And honestly, based on the dyno in that other thread, I think by ramping up the boost gradually like they (electromotive) did, they're staying in a good area of the map anyways. They're obviously not shooting for JUST past the surge point at 4,000RPM. Though we don't have a compressor map for the turbo they used, so it's hard to guess that kind of stuff.

 

PS On a side note I was looking through pictures of the F1 turbo systems back from the 80's and I saw on several of them obvious speed sensor devices. What I didn't expect was the pre-turbo throttle bodies. From some reading it looks like they were used in tandem with a regular throttle body usually to keep idle speed of the turbo higher. Interesting alternative to the modern conventional BOV.

Edited November 5, 2010 by Gollum

[Tony D]

"ARRRGH!"

 

 

"PSI" "PSI" "PSI"

 

The TURBO CURSE!

 

NO NO NO NO NO!

 

A larger, more efficient engine WOULD NOT "use less PSI"!

 

Listen, and FOLLOW CLOSELY:

 

An L28 stock will take X POUNDS PER HOUR of fuel and air to make 300HP.

 

An L28 with an Isky cam and Ported Heads will still take X POUNDS PER HOUR of fuel and air to make 300HP.

 

"PSI" is NOTHING but an indication of RESISTANCE TO FLOW.

 

"PSI" does not give you HORSEPOWER!

 

"FLOW" gives you horsepower. It's Pounds per HOUR.

 

"PSI" is a static measurement, nothing more. You can have 8psi coming out of a T28 and 8psi coming out of a T56... which do you think will support more horsepower? They are both "PSI"

 

The larger, more efficient engine will SHOW LESS "PSI" as it's restriction to flow is less, and therefore will work in a different pressure ratio zone than a smaller engine which may be less efficient, or simply have a reduced runer diameter restricting flow.

 

In EVERY example you gave above, the SAME turbo on EACH engine would HAVE to flow the pounds-per-hour necessary to get 400HP from the engine. The airflow will be exactly the same (for these purposes). Because of differing flows through the engine though, the pressure ratio it operates at will be dictated by the resistance each individual engine takes to FLOW that air into the combustion chamber. If it's restrictive, then yes the PSI may raise. Likely you will see more PSI on the 1.6 than the 5.0...

 

A prime example of this is JeffP's engine, making 380ft-lbs of torque at 4000-4500rpms at 8.39psi of boost.

Take a look at what a stock engine takes to get there (with a near stock turbo) 16-18psi? But what if that is a GT35R on there same as Jeffs? Then you have a big-phil style surge issue in the midrange because the engine simply doesn't FLOW enough air to support the minimum flow requirements for the turbo when it's spinning at whatever speed it is---it supplied more air than possible for the engine to digest. So boost builds. Problem is boost builds realllly high, and due to the flow restriction of the engine at 3800-4800 rpms, the turbo experiences midrange surge at 25psi. It does not at 17psi.

 

A ported head and cam would/could solve this---letting more flow in and keeping stable flow at that pressure...but because it FLOWS MORE the PSI SEEN will be LOWER. It's a double-edged swort, you flow more and lower your pressure ratio because the turbo puts out the same flow, but now you are using more. Pressure drops, flow rises, you make more power, and you don't surge...

 

A larger exhaust housing will help, only because the turbo slows down, pushing less air, and maybe matching the engine's needs better. Problem is at the top end you may have lost something...and for SURE on the bottom end you lost boost threshold and made the engien 'more peaky'.

 

Had a BOV system like we've been discussing been implemented, the engine in it's stockish form could then TAKE 25psi without surging (excess venting to stabilize flow at that pressure ratio and turbine speed) and therefore make more power than previously because you had to turn it down to 17 to prevent it form surging. In fact, you could downsize the wheel on the turbine to up the wheel speed allowing HIGHER pressure ratios, and more flow consequently (normally this would mean more boost pressure, and it does, but in the former iteration you couldnt' do that, you would SURGE SURGE SURGE. Now you can dump excess air, and run a higher psi and flow into the engine than formerly possible.

 

Confusing, no? And I'm swamped at work and can't concientiously log on tomorrow because I HAVE to get more paperwork done! It may be monday before I get back on here, realistically...

[Gollum]

Just because you say you're busy (which I believe) I won't ask any more questions for now, since I think I'm starting to get it.

 

Again, everything you've said I've TOLD and tried to TEACH to others, to explain why PSI is poorly understood among people, and how little it really matters, but again, my language just isn't there to communicate it obviously. Here, I'll edit my last post to better reflect what I was getting at, which you in turn answered well enough for me.

I'm pretty sure I understood all that, so I'm wondering if I'm just not phrasing things correctly.

 

I realize that pressure ratio/psi AND flow will be direct variables depending upon engine displacement, volumetric efficiency and power goals. If you have a large efficient engine it won't need as much PSI as a smaller less efficient engine to reach the same power.

 

IE: Even a POOR efficiency ford 5.0 stock will only end up with a quarter the PSI when reaching for 400hp as compared to a high efficiency honda 1.6 liter which will require much more increase in flow with less displacement to move it. But by the same token the 4.6 ford DOHC though smaller than the 5.0 will see LESS PSI than the 5.0 to reach the same power, due to being a more efficient engine with better breathing heads and higher stock HP levels. Three engines, same power, very different pressure ratios/PSI requirements. But each engine would require very different turbos to reach that HP efficiently since they'll all need a very different proportion of increase of power over NA application of the same given engine, and varying starting HP levels.

 

I guess my point is that it seems as though YOUR point is that this system allows you to run a much larger compressor/smaller turbine (much more offset than the typical T3/T4). And I'm saying that by what I'm seeing on load calculations that you'll be well into surge territory under normal wastegate control (duh, I think we've covered that, thanks for your patience with me ). So in this system you dump the excess air the compressor is making at lower RPM (or any for that matter) to get it back into stable region of the map, but what I'm asking is "won't that be a low efficiency area of the map?" and "would you want to vent MORE air to create MORE flow through the compressor to FORCE it into the better region of the map?"

 

I think that make sense. And honestly, based on the dyno in that other thread, I think by ramping up the boost gradually like they (electromotive) did, they're staying in a good area of the map anyways. They're obviously not shooting for JUST past the surge point at 4,000RPM. Though we don't have a compressor map for the turbo they used, so it's hard to guess that kind of stuff.

 

PS On a side note I was looking through pictures of the F1 turbo systems back from the 80's and I saw on several of them obvious speed sensor devices. What I didn't expect was the pre-turbo throttle bodies. From some reading it looks like they were used in tandem with a regular throttle body usually to keep idle speed of the turbo higher. Interesting alternative to the modern conventional BOV.

 

I think the irony in the three engines I gave examples of, is that though they'd all require very different setups, they could all in theory use around the same compressor, since the flow requirements are the same in the end. The exception would be that the Honda 1.6 liter will end up in a very different part of the map due to it's higher pressure ratio required in order to reach 400hp. If finding a compressor to meet the needs of all these engines, you'd probably end up with a compressor that's never really perfect for any of the engines. I'll calculate some load maps and see what I come up with, for theory sake.

[Gollum]

As said, there really isn't such a thing as "perfect compressor" that would fit those three engines, despite their similar HP goals, and thus similar CFM/Min of air and fuel requirements to meet that HP goal.

 

listed first is the 5.0, then 4.6 and then the 1.6. Vol. Eff. is based on dyno software emulating true factory torque curves. Should be close enough for theory's sake.

 

http://www.squirrelpf.com/turbocalc/graph.php?version=4&pr0=1&pr1=1.07&pr2=1.66&pr3=2.26&pr4=2.26&pr5=2.26&pr6=2.26&pr7=2.26&airflow0=4.5&airflow1=7.3&airflow2=14.5&airflow3=24&airflow4=34.2&airflow5=44&airflow6=47.8&airflow7=49.3&product_id=109

 

http://www.squirrelpf.com/turbocalc/graph.php?version=4&pr0=1&pr1=1.07&pr2=1.41&pr3=1.75&pr4=1.75&pr5=1.75&pr6=1.75&pr7=1.75&airflow0=3.8&airflow1=6.1&airflow2=10.8&airflow3=17&airflow4=28.2&airflow5=40&airflow6=42.1&airflow7=42.7&product_id=109

 

http://www.squirrelpf.com/turbocalc/graph.php?version=4&pr0=1&pr1=1.07&pr2=1.86&pr3=2.66&pr4=2.66&pr5=2.66&pr6=2.66&pr7=2.66&airflow0=1.5&airflow1=5.6&airflow2=13.8&airflow3=26&airflow4=32&airflow5=37.6&airflow6=37.9&airflow7=37.8&product_id=109

 

Compressor surge might be an issue still for the 1.6 there, but it's in the decent range for the V8's. Though it's amazing how two V8's of similar displacement, with stock power levels being semi-close can read so differently on a map.

 

 

 

The turbo used is the GT3076R. One of the larger compressor's for the GT30 turbine frame, not sure how it compares to a T3/T4 typical hybrid.

 

If we were seeking IDEAL turbos I might choose something more like this:

 

The 5.0 would get a GT3782R, leaving some room to shoot for 500hp and in a good efficiency area for street use

http://www.squirrelpf.com/turbocalc/graph.php?version=4&pr0=1&pr1=1.07&pr2=1.66&pr3=2.26&pr4=2.26&pr5=2.26&pr6=2.26&pr7=2.26&airflow0=4.5&airflow1=7.3&airflow2=14.5&airflow3=24&airflow4=34.2&airflow5=44&airflow6=47.8&airflow7=49.3&product_id=44

 

The 4.6 would get a GT4294R letting it get into a really nice range for upper RPM use when you stretch out the motor and use those rev's it's got. Might be sacrificing some street performance but giving good race use and quite a bit of flow left on the table if more HP was desired.

http://www.squirrelpf.com/turbocalc/graph.php?version=4&pr0=1&pr1=1.07&pr2=1.41&pr3=1.75&pr4=1.75&pr5=1.75&pr6=1.75&pr7=1.75&airflow0=3.8&airflow1=6.1&airflow2=10.8&airflow3=17&airflow4=28.2&airflow5=40&airflow6=42.1&airflow7=42.7&product_id=49

 

 

The 1.6 would get a GT3571

http://www.squirrelpf.com/turbocalc/graph.php?version=4&pr0=1&pr1=1.07&pr2=1.86&pr3=2.66&pr4=2.66&pr5=2.66&pr6=2.66&pr7=2.66&airflow0=1.5&airflow1=5.6&airflow2=13.8&airflow3=26&airflow4=32&airflow5=37.6&airflow6=37.9&airflow7=37.8&product_id=41

 

This is probably the most interesting of the choices. It's "within" use of this application, but it's hard to find a turbo for this application that doesn't say "danger danger" all over it, because you need the larger compressor to have enough air flow for the HP goals, but it's low displacement makes it easy to put into surge... So you end up doing what big phil did basically, go with a larger AR on the hot side, or a larger hot side all together. So this GT3571 has a huge turbine/hot side frame, and a smaller compressor than the other engines. But this prevents borderline surge constantly, and puts you in a decent efficiency range for street driving, but racing on this engine a lot might lead to early turbo failure... Hmmm... wonder why those turbo honda guys seem to blow up turbos a lot when they're pushing 500hp.... hmm....

[Tony D]

 

Your edited post is correct, I think you got it. In theory each of the engines could have the same turbo but in reality each would require quite a bit different turbo to match it for the PR eventually required. The application of the compressor becomes less critical when you have a control system that will allow you to 'fudge' some of the variables.

 

The advent of the 'hybrid' turbo (and I mean a late 180 HP compressor wheel on an early 63-64 Corvair hot side) to get more boost lower in the RPM range has made simple wastegate control a compromise at best.

 

The advent of super efficient compressor profiles has allowed a hybrid to crank serious flow numbers at lower and lower boost thresholds---unfortunately the surge line limits their suitability in lower-flow applications.

 

The active venting control method effectively allows you to run much higher pressures at those lower rpm ranges for better acceleration. It also would allow you to size the compressor for peak efficiency and flow at MAXIMUM horsepower and rpm levels (Big Phils Setup) without the midrange surge issues inherent in the old paradigm of 'bigger hot side to slow the flow down low'...

 

Traditionally big compressors ALWAYS had a higher boost threshold becuase using wastegate only control that is what they HAD to use! You make a peaky engine.

 

By actively venting flow down low, you recover driveability because you can then use what would be considered a 'normal' hotside A/R (0.63 as Phil originally had, as opposed to the 0.82 he had to resort to eventually).

 

The paradigm of 'backpressure in the manifold' has to also be rethought as even at over 500HP, the stock manifold and 0.63 A/R hot side on JeffP's engine was still equal to the intake manifold boost pressure at peak HP and peak RPM (7000). The old saw that 'you need the bigger A/R to breathe on the top end' seems to be less an issue now that turbine wheels have been redesigned as well.

 

The big news is that now it may be possible for a guy with a 'stockish' engine to bolt on a GT35R with a hot side A/R of 0.48, and get full boost as it would in a stock turbo application (2000 to 2500rpms as opposed to 3400 rpms) and then use that to his advantage.

 

For guys like JeffP with a ported head, he could run that 0.48 to drop his boost threshold even lower, and get even more power under the curve for low speed drivability!

 

Back to work, just checking in!

Edited November 5, 2010 by Tony D

[Tony D]

BTW, in the first example, the 4.6 looks 'ideal.'

The 5.0 looks to need a 'smaller' setup.

The 1.6, on the other hand, would benefit from the boosting of wheel speed and active venting to move the curve to the right, and as such would have plenty of reserve. This control scheme would allow what normally would be a 'poor or marginal' selection to operate pretty well in fact, and with a smaller A/R hot side still retain deecent boost threshold.