Tutorial:how to read a turbo compressor map
08-12-2010, 11:55 PM
#22
So much misinformation...
Garrett's published turbine maps are of little use for matching, as JasonC mentioned. His attachment is a "real" map.
I was going to post a painfully in-depth explanation of how to use a proper turbine map for matching purposes. However, without actually doing the exercise of working through the numbers it would only totally lose everyone. And without proper maps, you can't do a match anyway.
Suffice it to say that matching a turbine to a compressor requires a flow, speed and power balance. You first determine the compressor operating conditions, and then use a turbine map to determine whether it is suited to drive the compressor under those conditions. It is a highly iterative process. You have to make assumptions. You have to know the limitations of the map (steady flow conditions, measurement uncertainty, etc)
A higher flowing turbine requires lower expansion ratio, which is why an engine thusly equipped makes more power at a given boost level (less backpressure; higher VE).
Blah blah blah
Garrett's published turbine maps are of little use for matching, as JasonC mentioned. His attachment is a "real" map.
I was going to post a painfully in-depth explanation of how to use a proper turbine map for matching purposes. However, without actually doing the exercise of working through the numbers it would only totally lose everyone. And without proper maps, you can't do a match anyway.
Suffice it to say that matching a turbine to a compressor requires a flow, speed and power balance. You first determine the compressor operating conditions, and then use a turbine map to determine whether it is suited to drive the compressor under those conditions. It is a highly iterative process. You have to make assumptions. You have to know the limitations of the map (steady flow conditions, measurement uncertainty, etc)
A higher flowing turbine requires lower expansion ratio, which is why an engine thusly equipped makes more power at a given boost level (less backpressure; higher VE).
Blah blah blah
Reply
0
0
08-13-2010, 12:48 PM
#23
Elite Member
Join Date: Jul 2005
Posts: 6,420
Total Cats: 84
Nagase and Joe:
(Nagase the passage you quoted about maxing out the turbine is incorrect)
The turbine side CAN flow more than the turbine "map" shows because that is only the flow through the turbine, the rest of the exhaust gases can flow through the wastegate. Even if the turbine is in the "maxed out" or level portion of the line, it's not really maxed out - if you move to the right along the line, even though flow is flat, pressure can rise, and the turbine *will* develop more shaft power for the compressor, and more exhaust flows through the wastegate.
The actual shaft power that the turbine produces is proportional to the efficiency times the exhaust mass flow rate times the pressure drop. To figure out the exact power developed, IIRC, requires one to look up the enthalpy of the exhaust gas (I'm an EE and my thermodynamics is a fuzzzy memory now).
The shaft power required by the compressor is likewise proportional to the air mass flow rate times the pressure gain, divided by the compressor efficiency.
---------
With a given turbine/compressor/engine combo, there is a point where, as boost is turned up, the engine starts to show a torque curve that drops off more sharply at the peak power area (i.e.the % gains are smaller per add'l psi of boost near peak power vs lower in the RPM range). If the compressor isn't approaching the choke region, it is the *engine* that actually "chokes" at the top-end as the turbine requires much more backpressure to generate boost, because it reduces engine volumetric efficiency. You will see this as sharply increasing turbine backpressure at the topend (I saw this with my GT2554 at >10 psi). At this point I think anything that improves topend breathing will "uncork" some hp, and it will be good for another few psi. This can be done with a more hi-rpm tuned intake mani, a larger a/r, or possibly retarded cam timing or duration.
(see discussion between me and Savington about intake manifolds and turbines. https://www.miataturbo.net/showpost....&postcount=127
The question is, which mod doesn't give up so much low-end)
----
Joe, the "corrected flow" is just the flow referred to standard atmo conditions. It's a convenient way to quote flow numbers, or to plot flow vs pressure, because otherwise the plot will change according to temperature (and you'd have to plot a family of curves).
(Nagase the passage you quoted about maxing out the turbine is incorrect)
The turbine side CAN flow more than the turbine "map" shows because that is only the flow through the turbine, the rest of the exhaust gases can flow through the wastegate. Even if the turbine is in the "maxed out" or level portion of the line, it's not really maxed out - if you move to the right along the line, even though flow is flat, pressure can rise, and the turbine *will* develop more shaft power for the compressor, and more exhaust flows through the wastegate.
The actual shaft power that the turbine produces is proportional to the efficiency times the exhaust mass flow rate times the pressure drop. To figure out the exact power developed, IIRC, requires one to look up the enthalpy of the exhaust gas (I'm an EE and my thermodynamics is a fuzzzy memory now).
The shaft power required by the compressor is likewise proportional to the air mass flow rate times the pressure gain, divided by the compressor efficiency.
---------
With a given turbine/compressor/engine combo, there is a point where, as boost is turned up, the engine starts to show a torque curve that drops off more sharply at the peak power area (i.e.the % gains are smaller per add'l psi of boost near peak power vs lower in the RPM range). If the compressor isn't approaching the choke region, it is the *engine* that actually "chokes" at the top-end as the turbine requires much more backpressure to generate boost, because it reduces engine volumetric efficiency. You will see this as sharply increasing turbine backpressure at the topend (I saw this with my GT2554 at >10 psi). At this point I think anything that improves topend breathing will "uncork" some hp, and it will be good for another few psi. This can be done with a more hi-rpm tuned intake mani, a larger a/r, or possibly retarded cam timing or duration.
(see discussion between me and Savington about intake manifolds and turbines. https://www.miataturbo.net/showpost....&postcount=127
The question is, which mod doesn't give up so much low-end)
----
Joe, the "corrected flow" is just the flow referred to standard atmo conditions. It's a convenient way to quote flow numbers, or to plot flow vs pressure, because otherwise the plot will change according to temperature (and you'd have to plot a family of curves).
Reply
0
0
08-13-2010, 01:27 PM
#24
^yup.
The escalating turbine inlet pressure you see with small turbos running a lot of boost makes VVT veeery nice to have. You minimize the reversion of the high pressure exhaust gas into the combustion chambers by dialing out some valve overlap (retarding the intake cam). You still have the pumping loss associated with pushing that high pressure exhaust gas out of the chamber, but at least some of the VE implications of the high exhaust pressure can be ameliorated.
The escalating turbine inlet pressure you see with small turbos running a lot of boost makes VVT veeery nice to have. You minimize the reversion of the high pressure exhaust gas into the combustion chambers by dialing out some valve overlap (retarding the intake cam). You still have the pumping loss associated with pushing that high pressure exhaust gas out of the chamber, but at least some of the VE implications of the high exhaust pressure can be ameliorated.
Reply
0
0
08-13-2010, 06:56 PM
#25
Elite Member
Join Date: Jul 2005
Posts: 6,420
Total Cats: 84
My VVT tuning at 10 psi on my GT2560 wants it to go to full retard at about 6000 RPM IIRC. Are you suggesting that if I turn up the boost, it's gonna want full retard at lower and lower RPM?
Also, I'm playing with the idea of purposely retarding my intake cam 1 tooth, for further max retard. The VVT range is 47* (crank degrees), and in the midrange my motor doesn't like anything more than about 22* advance from max retard, so retarding it another tooth will not prevent me from reaching optimum advance. I'm thinking the engine may like more retard than the standard maximum at higher boost... and perhaps during cruise, extra retard may reduce pumping losses... are there any major downsides to miniscule overlap or even underlap at cruise, idle, and maybe high boost/high RPM?
Also, I'm playing with the idea of purposely retarding my intake cam 1 tooth, for further max retard. The VVT range is 47* (crank degrees), and in the midrange my motor doesn't like anything more than about 22* advance from max retard, so retarding it another tooth will not prevent me from reaching optimum advance. I'm thinking the engine may like more retard than the standard maximum at higher boost... and perhaps during cruise, extra retard may reduce pumping losses... are there any major downsides to miniscule overlap or even underlap at cruise, idle, and maybe high boost/high RPM?
Reply
0
0
08-13-2010, 07:55 PM
#26
Never go full retard.
Seriously, you'll just have to play with the vvt on a dyno. At full load/revs there will be a point of diminishing/negative returns with more and more intake retard, since you're moving the IVC point to later in the cycle. How much retard is best will depend on intake & exh pressure, and how aggressive your cams are. Idle usually works well with minimal overlap. OEMs use a good bit of overlap at cruise for EGR.
I'd move the intake cam one tooth and then experiment.
Seriously, you'll just have to play with the vvt on a dyno. At full load/revs there will be a point of diminishing/negative returns with more and more intake retard, since you're moving the IVC point to later in the cycle. How much retard is best will depend on intake & exh pressure, and how aggressive your cams are. Idle usually works well with minimal overlap. OEMs use a good bit of overlap at cruise for EGR.
I'd move the intake cam one tooth and then experiment.
Reply
0
0
Thread
Thread Starter
Forum
Replies
Last Post
Mikel
MEGAsquirt
4
09-28-2015 04:46 PM
