My harrowing journey through the murky swamps of compressor map interpretation. - page 1 - IDI Engine - VWDiesel.net The IDI, TDI, and mTDI source.
-
My harrowing journey through the murky swamps of compressor map interpretation. by JoeCanuck on 31 Mar, 2014 04:01
-
Well tonight I spent a bunch of google time trying to figure out the voodoo of compressor mapping. I stumbled upon a few sites which helped. Using the information linked to below, I calculated that a K14 on a 1.9l engine will be 70% efficient pushing 20psi at 5000rpm...and according to the hondaswap tutorial, anything over 70% is good. The K14 in this application can produce 20psi boost starting at around 2200rpm. According to my calculations, even 25 psi below about 4800rpm is still above 60% efficiency.
In contrast, the K24 can't produce 20psi until around 2500rpm....but it can stay over 70% efficiency at 30psi until 5200rpm.
Now one should take all this with a grain of salt....this is my first crack at playing with compressor maps and I haven't done anything resembling math for decades. Ideally, someone smarter with more time on their hands would work out compressor efficiencies for the more popular turbos....perhaps supplying some real data to one of the most frequently asked questions.
There are also some interesting youtube videos on the subject.
This site gives a really good tutorial to understanding compressor maps
http://hondaswap.com/threads/compressor-map-reading-for-dummies.129194/
Here is a big collection of compressor maps
http://www.not2fast.com/turbo/maps/all.html#kkk
This one is a compressor map flow calculator...(It seems the VE, (efficiency), is typically 85% for 2 valve engines and 90% for 4 valve engines)
http://lovehorsepower.com/joomla/index.php?option=com_content&view=article&id=35:compressor-flow-map-calculator&catid=8:mr2-helpful-stuff&Itemid=49
This is a handy unit converter, because some of the maps are in kg/sec and/or m3/sec.
http://www.convertunits.com/from/lb/min/to/.04+kg/s
EDIT- I did a whole bunch of searching today to double check if the maps in the link above are labelled correctly....and it seems they are....or everybody has been fooled.
Here's another really good turbo site with maps and info.
http://www.rbracing-rsr.com/turbotech.html
Here's a link to the Borg-Warner(KKK) site with a good turbo application search engine
http://www.3k-warner.de/en/aftermarket/productSearch.aspx
If you just can't get enough cold, hard theory and math, this site really gets into the nuts and bolts of turbocharging.
http://www.enginelogics.com/read-a-turbo-compressor-map/
-
#1 by RabbitJockey on 31 Mar, 2014 08:59
-
i've never seen compressor maps for the k24 that is actually found on a vw diesel, there are a bunch of different k24's and even k14's, which 14 map are you looking at?
-
#2 by JoeCanuck on 31 Mar, 2014 11:40
-
The maps were found in the link at not2fast.com. I have no idea what particular model of K24 or K14 the map is from, but I would be very surprised if they differ from each other to such a degree that the maps would be significantly different...but I could be wrong. These were used on a number of cars so perhaps the differences were more external and appication based, rather than performance based.
It definitely would be interesting to be able to physicaly compare different models. -
#3 by vanbcguy on 31 Mar, 2014 12:02
-
The compressor maps do actually vary a fair bit from one version of these turbos to the next. IE one map might be perfect while another will have the turbo in the surge zone through most of its normal operating range. The various trims of the Garret T3 are a perfect example of this. Or for a more modern version, check out the 3 different GT2052 maps on the Garret site. The differences are definitely more than just external features.
-
#4 by JoeCanuck on 31 Mar, 2014 14:12
-
The compressor maps do actually vary a fair bit from one version of these turbos to the next. IE one map might be perfect while another will have the turbo in the surge zone through most of its normal operating range. The various trims of the Garret T3 are a perfect example of this. Or for a more modern version, check out the 3 different GT2052 maps on the Garret site. The differences are definitely more than just external features.
Interesting...I'll check the garret stuff out. So far, I'm just trying to grasp how to interpret maps. Whether or not these maps are actually for my particular turbos may never be known....hopefully someone can nail this stuff down since it would be very useful information.
At least the K14 and K24 based on the maps are making sense, in that the K14 is shown to spool quicker and top out earlier than the K24. I do find it interesting that the K24 doesn't spool up significantly slower than the K14, yet can handly significantly more boost. -
#5 by Alcaid on 01 Apr, 2014 01:45
-
Remember that all factors are different on these IDI engines than what they are on petrol and DI. And all factors vary over the RPM range to complicate things even further
To predict how fast a turbo will spool one needs to interpret the turbine map as well
-
#6 by RabbitJockey on 01 Apr, 2014 04:40
-
Remember that all factors are different on these IDI engines than what they are on petrol and DI. And all factors vary over the RPM range to complicate things even further
To predict how fast a turbo will spool one needs to interpret the turbine map as well
haha yes! then remembering that temperature and atmospheric pressure changes everything even more, makes the whole process even more fun -
#7 by JoeCanuck on 01 Apr, 2014 14:15
-
I''m basically just looking for ballpark figures....comparing one against the other, not absolute numbers. In that, respect, the maps can add some data to the discussion rather than just relying on subjective observation.
We do get some information on how quickly a turbo will spool and on the turbine since, I believe, the map is made based on lab data with the turbine spinning the compressor.
As for how the mapping might differ on different engines, it looks to me that they treat the engines basically like air pumps. The amount of hp/torque you get from a different type of engine will differ, but the air flow should be in the ballpark.
The compressor map flow calculator does give a good indication of flow over different rpms.
As I said, I look at it as a general comparison tool, rather than trying to calculate how much power you'll get from an engine/turbo setup. That's a different kettle of fish altogether.
What we really need is some objective, real world data....basically dyno runs where both boost and drive pressures are measured. That will give the best information on a turbo's efficiency range and how it affects a particular application..
Failing that, I find the maps interesting...and they are starting to make sense. -
#8 by Gizmoman on 01 Apr, 2014 17:20
-
I''m basically just looking for ballpark figures....comparing one against the other, not absolute numbers. In that, respect, the maps can add some data to the discussion rather than just relying on subjective observation.
We do get some information on how quickly a turbo will spool and on the turbine since, I believe, the map is made based on lab data with the turbine spinning the compressor.
As for how the mapping might differ on different engines, it looks to me that they treat the engines basically like air pumps. The amount of hp/torque you get from a different type of engine will differ, but the air flow should be in the ballpark.
The compressor map flow calculator does give a good indication of flow over different rpms.
As I said, I look at it as a general comparison tool, rather than trying to calculate how much power you'll get from an engine/turbo setup. That's a different kettle of fish altogether.
What we really need is some objective, real world data....basically dyno runs where both boost and drive pressures are measured. That will give the best information on a turbo's efficiency range and how it affects a particular application..
Failing that, I find the maps interesting...and they are starting to make sense.
Your last link in the previous post is most interesting. I would imagine that if you knew what's going into the engine (fuel and air @ X psi) and you knew what the ideal ratio of fuel to air should be, then you'd have a close estimate of output at a given RPM. -
#9 by JoeCanuck on 01 Apr, 2014 18:47
-
Your last link in the previous post is most interesting. I would imagine that if you knew what's going into the engine (fuel and air @ X psi) and you knew what the ideal ratio of fuel to air should be, then you'd have a close estimate of output at a given RPM.
The problems with calculating power for a particular vehicle/engine combination is that you'd have to know the max flow rate through the entire system, starting at the air filter and ending at the tail pipe. But, one can get close by comparing their setup to another....or just dyno the darned thing.
As for the maps, the guys in the links actually make it a bit harder to figure out than it has to be. Here are my steps, (yours may differ). Feel free to play along at home.
1. Fill out the compressor map calculator. This will give airflow through your engine. You will use this data to plot points on a compressor map.
2. Dig out your handy compressor map. (Since there are a number of difference units for air mass/volume flow, you may need to change some units to something you're more comfortable with, lbs/min, CFM, kg/sec, m3/min....whatever tickles your fancy, really, The compressor map calculator only uses cfm and lbs/min and the kkk maps I was using used kg/sec and m3/sec. It may be easiest if you convert the units from the map into either cfm or lbs/min right off the bat so there's less playing around later).
You're interested in the oval islands with numbers like 70, 74, 60, etc. Those islands represent the pressure ratio, airflow areas of efficiency in the percent labelled. The higher the percentage, the more efficient your turbo at those pressure ratio/flow combinations. From what I've read, ideally you're looking for over 70% for continuous operations but I'm guessing as long as your pyro isn't gving you fits, down to 60% will probably be ok for limited periods.
3 Take the pressure ratio reading from your compressor map calculator at the boost you are interested in, find that PR on the compressor map, and draw a horizontal line through the map. Where that line crossed the surge line, draw another line straight down. That will give you the lowest flow at which your turbo can give you the requested boost. (If you take that number back to the compressor map calculations, you can find the rpms where that boost will occur. The lower the rpms, the faster the spoolup, in other words, less turbo lag).
4. To find out the max boost your turbo can put out efficiently, (before it starts making as much heat as boost), move along the horizontal pressure ratio line you drew, until you come to the right side of the 70% island, (or whatever is close). Draw a line straight down from that point. Take the flow number back to your compressor map calculator results. Wherever on the plot you find the number, is the maximum boost/flow that your turbo can efficiently produce.
So you'll get the two most important numbers when trying to fit a turbo to your engine;; turbo lag and maximum efficient boost. This will not give you power readings unless you know the factors involving your fuel, airfuel ratio, and flow effiency of your entire engine system. These are also not absolutely correct numbers since we're just guessing at the volumetric efficiency of your engine.
I'm not sure if I've made things any clearer but it works for me.
-
#10 by Alcaid on 01 Apr, 2014 23:04
-
We do get some information on how quickly a turbo will spool and on the turbine since, I believe, the map is made based on lab data with the turbine spinning the compressor.
Not correct at all, compressor is spun with an electrical motor to produce the compressor maps, compressor efficiency maps have nothing to do with the turbine performance -
#11 by RabbitJockey on 02 Apr, 2014 04:28
-
i support this messageWe do get some information on how quickly a turbo will spool and on the turbine since, I believe, the map is made based on lab data with the turbine spinning the compressor.
Not correct at all, compressor is spun with an electrical motor to produce the compressor maps, compressor efficiency maps have nothing to do with the turbine performance
also, my favorite turbo calculator is at squirrelpf.com -
#12 by Gizmoman on 02 Apr, 2014 04:40
-
I still applaud your efforts Joe, I never even thought about trying to understand them till you took a stab at it.
-
#13 by JoeCanuck on 03 Apr, 2014 08:06
-
We do get some information on how quickly a turbo will spool and on the turbine since, I believe, the map is made based on lab data with the turbine spinning the compressor.
Not correct at all, compressor is spun with an electrical motor to produce the compressor maps, compressor efficiency maps have nothing to do with the turbine performance
I stand corrected. Regardless, since we don't seem to have any turbine data, the interpretations from the compressor maps will have to do....at least for me.
At least it's something a bit better than guessing. For instance, according to my calculations, a K14 compressor, is better than 70% efficient at 20psi boost up to 5000rpm....which is the max I'm aiming for....at least at first, and since it can reach 20psi before the K24, I'll probably try the smaller one first. On the other hand, I plan on making sure the K24 will fit on the mounted engine if I decide to switch....which I probably will, if for no other reason than curiosity.
Now if the damned snow would finally get the hell away so I can work on the beastie without dying of exposure.
Gizmoman; I've been reading your turbo transplant thread...you are doing some good work there....you have way more patience for fabbing than I do. I'm hoping that pushing the K14 to 20psi will add enough air to let me make some long pulls through the hill without making molten aluminum. Time will tell. -
#14 by RabbitJockey on 03 Apr, 2014 09:53
-
those k14 and k24 maps are definitely not for the vw turbos, i would not make your decision based on what you see from those 2 maps.