VWDiesel.net The IDI, TDI, and mTDI source.
Engine Specific Info and Questions => IDI Engine => Topic started by: fspGTD on February 02, 2006, 01:23:23 pm
-
I'm taking a trip to a diesel injection specialist to have my GTD's fuel injection pump "dyno tested" on a calibration test bench. I'm going to record fuel output quantities and timing advance at various pump speeds so I can plot them versus RPM and see the characteristic curves. I plan on posting the results to this thread as soon as I get them.
This information could come in handy to us for better understanding what's possible with simple injection pump adjustments, and governor and LDA adjustments/modifications. The pump is from a 1984 Jetta 1.6lTD (non prestroke, 9mm diameter plunger and 2.2mm lift camplate), and has its intermediate governor spring removed, an extended throttle lever geometry to raise the revv-limiter, a full-travel LDA, and a fuel screw adjustment that has been increased to the point where the RPMs just barely return. Basically I feel this pump is already if not close to "maxxed out" for what is achieveable with the stock 1.6lTD pump internals.
(http://www.imagestation.com/picture/sraid83/p9096be55ac139d34ec6bca060e467f95/fad2d912.jpg)
Based on performance increases resulting from these modifications (and a look at what comes out the exhaust pipe :lol: ) I predict it is kicking out a lot more fuel than a stock pump, but I am really looking forward to quantifying how much. Stay tuned...
PS - a more sophisticated version of the "governor mod" is in the works! It will probably not be tested in this first round however.
-
How much does it cost to have a pump bench tested? I have an evil pump that may need to be tested. :twisted:
-
This sounds really interesting!!!
I can't wait to hear the results (ad adjustments made to get to that point!) would be nice to see what a stock pump can do!
I'll be the first to say, thanks for sharing!
8)
-
I injection pump dyno tested my GTD Autocrosser's VE injection pump successfully yesterday! Although it was more time/effort than I had previously expected, it was a lot of fun, and I feel that I came home with some really great data on my pump.
Cutting straight to the chase: the following graph I put together shows the results:
(http://home.comcast.net/~vwgtd/GTD_VEPumpDyno_2-06-2006.png)
An analysis of the results will be coming soon, but as a quick teaser: peak injection quantity measured on my pump is 89.4mm^3/stroke, which is 208% of the stock 43.0 specification.
PS - VWMike, shop rate is $75/hr, and figure 2-3 hrs for a VE pump test. There is a considerable amount of setup work to configure both the test pump and the test bench for each other. Also the testing process itself takes time, as each data point is individually tested and recorded manually (IE: paper and clipboard!)
Test equipment used:
(http://www.imagestation.com/picture/sraid200/pbefaf204e5c46a777d5d8bdc7b6ee7dd/f04032ad.jpg)
(http://www.imagestation.com/picture/sraid200/p9398ecb154eff5b1ba62bf4d90bec68b/f0403254.jpg)
My friend Jay at the diesel injection shop took a funny video of me operating the test bench for my first time. Anyone want to host?
-
An analysis of the results will be coming soon, but as a quick teaser: peak injection quantity measured on my pump is 89.4mm^3/stroke, which is 208% of the stock 43.0 specification.
Very interesting thread/post, fspGTD.
... and in case someone else didn't notice, fspGTD's 89.4cc is a bit more than the claimed 86cc from the pump used in andy2's twin turbo 1.9TD! It's an interesting thought...
Now Andys pump from Bowmanville/Ontario:
1.9L TD pump he's setting up with twin turbo
Full boost fuel 86 cc
edit: His 86cc was probably measured when the fuel screw was in medium setting, possibly with more room to turn it up.
I also find your drop in fueling (yellow line in graph) interesting, particularly the 3,000 to 4,000 RPM range. Your new sophisticated governor mod sounds nice from what you've told me earlier, I'm looking forward to it :)
-
this seems to be very informative, and i dunno how comparable it is, but most people can run 20psi with out sending the pump for a nice performance build, and you just showed ur pump giving the double the fueling of a stock pump, and 20psi is about double the boost of a stock pump... just my thoughts, not really sure how accurate they are...
-
Fine numbers with 9mm plunger. Where is that 2,2mm camplate from? Just for comparison, my 10mm pump has 100cc max with stock camplate.
-
Audi80 - very interesting... thanks for the info! I assume that this injection quantity was measured also on a pump test bench?
This pump being tested has a stock, untouched "bottom end" (everything below the governor) from a US-spec 1984 Jetta 1.6lTD. Engine code "CY", Bosch pump number 0 460 494 135, VW part number 068 130 109 Q. :)
-
My friend Jay at the diesel injection shop took a funny video of me operating the test bench for my first time. Anyone want to host?
www.putfile.com will host movies up to 10 MB for free.
-
I'm taking a trip to a diesel injection specialist to have my GTD's fuel injection pump "dyno tested" on a calibration test bench. I'm going to record fuel output quantities and timing advance at various pump speeds so I can plot them versus RPM and see the characteristic curves. I plan on posting the results to this thread as soon as I get them.
This information could come in handy to us for better understanding what's possible with simple injection pump adjustments, and governor and LDA adjustments/modifications. The pump is from a 1984 Jetta 1.6lTD (non prestroke, 9mm diameter plunger and 2.2mm lift camplate), and has its intermediate governor spring removed, an extended throttle lever geometry to raise the revv-limiter, a full-travel LDA, and a fuel screw adjustment that has been increased to the point where the RPMs just barely return. Basically I feel this pump is already if not close to "maxxed out" for what is achieveable with the stock 1.6lTD pump internals.
(http://www.imagestation.com/picture/sraid83/p9096be55ac139d34ec6bca060e467f95/fad2d912.jpg)
Based on performance increases resulting from these modifications (and a look at what comes out the exhaust pipe :lol: ) I predict it is kicking out a lot more fuel than a stock pump, but I am really looking forward to quantifying how much. Stay tuned...
PS - a more sophisticated version of the "governor mod" is in the works! It will probably not be tested in this first round however.
Now look what I've just found... :shock:
Obviously you've got the smoking under control these days; now this is what I call a spliff :D
Try two turns anti clockwise on the max fuel screw :wink:
http://img212.imageshack.us/img212/6227/jakesbigsmoke2zv.jpg
-
Mark-The-Miser: nice work uncovering that long lost pic! I'm gonna say that shot is a mid-90's "bug in" dragrace / VW swap meet/show at Portland International Raceways. Those were the 14 psi boost pressure glory dragracing/daily driver days. :) Wyatt Markus put that picture on his popular (at the time) web site about VW Rabbits - unfortunately now gone.
But back on topic: that car was running the exact same injection pump (just different governor and LDA tuning) as the one I just tested. :wink:
Jackbombay: thanks for the link to putfile... I may give that a try!
------------------------
Pics!
The pump looks like this when it's ready for the test bench:
(http://www.imagestation.com/picture/sraid201/p23a8bad5d243a1f275e39682ac4454db/f03f47ca.jpg)
It gets bolted up to a special mounting plate and also gets a special input shaft drive adapter bolted up to it:
(http://www.imagestation.com/picture/sraid200/pd9c3ff4b135da76998f7b24b4381553c/f040334e.jpg)
...and it get this neat timing advance measuring gizmo bolted up to the timing advance cover:
(http://www.imagestation.com/picture/sraid201/p163f3c77df9f178371651304c307c3fe/f03f468e.jpg)
Here is Bill of Precision Pump and Injector getting the pump ready to go on the test bench:
(http://www.imagestation.com/picture/sraid200/p3c5089633b1e888432a7984ed74e3049/f0403351.jpg)
Here is a close-up of some of the controls we used: testing oil heater (upper left) precisely regulates temperature of the "simulated fuel" used during the test. A super-precise tachometer is at the upper right. Guages below were used to monitor pump vital pressures.
(http://www.imagestation.com/picture/sraid201/pe22df7685eba72dcb4c71adddcf9cd25/f03f4179.jpg)
The blue lucas box shown earlier measured injection quantity very precisely for each cylinder and also the average quantity and "scatter" (the largest difference in cylinder-to-cylinder quantity.)
-
Mark-The-Miser: nice work uncovering that long lost pic! I'm gonna say that shot is a mid-90's "bug in" dragrace / VW swap meet/show at Portland International Raceways.
Those were the 14 psi boost pressure glory dragracing/daily driver days. :) Wyatt Markus put that picture on his popular (at the time) web site about VW Rabbits - unfortunately now gone.
But back on topic: that car was running the exact same injection pump (just different governor and LDA tuning) as the one I just tested. :wink:
Jackbombay: thanks for the link to putfile... I may give that a try.
This was the foto that Loren sent me and then blew up his computer a few weeks ago after a long time trying to revive it. I had thought it lost forever. Would make a nice poster in the workshop! (Just another couple of feet of plume next time :)
-
what kind of 1/4 times were you running, any time slips as i am curious about the 1/8 and 0-60 and 60' thanks!
Mark-The-Miser: nice work uncovering that long lost pic! I'm gonna say that shot is a mid-90's "bug in" dragrace / VW swap meet/show at Portland International Raceways. Those were the 14 psi boost pressure glory dragracing/daily driver days. :) Wyatt Markus put that picture on his popular (at the time) web site about VW Rabbits - unfortunately now gone.
But back on topic: that car was running the exact same injection pump (just different governor and LDA tuning) as the one I just tested. :wink:
Jackbombay: thanks for the link to putfile... I may give that a try!
------------------------
Pics!
The pump looks like this when it's ready for the test bench:
(http://www.imagestation.com/picture/sraid201/p23a8bad5d243a1f275e39682ac4454db/f03f47ca.jpg)
It gets bolted up to a special mounting plate and also gets a special input shaft drive adapter bolted up to it:
(http://www.imagestation.com/picture/sraid200/pd9c3ff4b135da76998f7b24b4381553c/f040334e.jpg)
...and it get this neat timing advance measuring gizmo bolted up to the timing advance cover:
(http://www.imagestation.com/picture/sraid201/p163f3c77df9f178371651304c307c3fe/f03f468e.jpg)
Here is Bill of Precision Pump and Injector getting the pump ready to go on the test bench:
(http://www.imagestation.com/picture/sraid200/p3c5089633b1e888432a7984ed74e3049/f0403351.jpg)
Here is a close-up of some of the controls we used: testing oil heater (upper left) precisely regulates temperature of the "simulated fuel" used during the test. A super-precise tachometer is at the upper right. Guages below were used to monitor pump vital pressures.
(http://www.imagestation.com/picture/sraid201/pe22df7685eba72dcb4c71adddcf9cd25/f03f4179.jpg)
The blue lucas box shown earlier measured injection quantity very precisely for each cylinder and also the average quantity and "scatter" (the largest difference in cylinder-to-cylinder quantity.)
-
Trevorbr: On your question about quarter mile times. The most recent quarter mile test I was able to find with this car was a 16.3 @ 82MPH, which I reported in 8/2003 here:
http://www.hostboard.com/cgi-bin/ultimatebb.cgi?ubb=get_topic&f=4037&t=233
My car has undergone considerable development since then however, and is doubtlessly faster in more recent configurations (perhaps into the 15's).
Ever since I bought an RPM-sensing G-tech that logs data and can generate graphs of RPM, Horsepower, and Torque in a specific gear, I have switched to using that method of testing as it give much more information than 1/4 mi testing provides.
PS - quoting a long reply especially when pictures are involved can make the thread a lot more unweildy, and difficult to read through. I would appreciate it you only used quoting when necessary, and only quoted the portions that are relevant to make your point clear. Thanks in advance!
---------------------------
Moving on to an analysis of the results...
(http://home.comcast.net/~vwgtd/GTD_VEPumpDyno_2-06-2006.png)
I was shocked to see in the VE pump dyno, the fueling being pulled back at an earlier than expected 4000RPM. Based on looking at the most recent and applicable hp and torque versus RPM curves I've got, and taking other factors into consideration, I thought I'd have seen the main governor spring kicking in more like around 4750 or somewhere around there. This pump should have a revv limiter significantly higher than stock! Even looking at the stock fueling with the early-acting intermdiate spring, fueling drops only modestly at 4500RPM from it's peak. I don't think the main governor spring of even a stock pump is kicking in below that point!
So... the mystery is: what is pulling back fueling on my pump as early as 4000RPM? Could it really be the main governor spring? Could there be some other unexplained phenomena at work here?
The "ski jump" like characteristic of the fueling curve approaching 4500RPM is also very intriguing. I'm not quite sure as of yet about what to make of this.
-
so its not what you wanted that the fuel got it's climax at 4000? am i right that this would cause the engine to run something like "lean"?
-
Yes, it is very surprising! Reduced fueling causes reduced torque in a diesel. If fueling were maintained higher, it may very well produce significantly more power and torque! :shock:
There are a number of factors possibly contributing to a difference in pump dyno versus real world performance, but I'm not sure if they're the main culprit behind the 4000RPM fuel dropoff or not:
* fuel temperature precisely regulated (to a little over 40 deg C in my case) on the pump dyno, versus varying in the real world (I have not measured mine, but would like to!)
* fuel supply line pressurized slightly on the pump dyno to the given specifications (about 2psi in my case), versus being under suction, pulling fuel from the tank and through a filter when on the car.
* fuel injectors are different, "test orifice injectors" with different characteristics and set to their own breaking pressures versus the real stock injectors (or in my case, non-stock injectors.)
* There is potential for the RPM of a chassis dyno machine or G-tech to be miscalibrated, however with the pump dyno giving an RPM readout resolution to the nearest single RPM, it would seem to be very trustworthy.
-
Yes, it is very surprising! Reduced fueling causes reduced torque in a diesel. If fueling were maintained higher, it may very well produce significantly more power and torque! :shock:
There are a number of factors possibly contributing to a difference in pump dyno versus real world performance, but I'm not sure if they're the main culprit behind the 4000RPM fuel dropoff or not:
* fuel temperature precisely regulated (to a little over 40 deg C in my case) on the pump dyno, versus varying in the real world (I have not measured mine, but would like to!)
* fuel supply line pressurized slightly on the pump dyno to the given specifications (about 2psi in my case), versus being under suction, pulling fuel from the tank and through a filter when on the car.
* fuel injectors are different, "test orifice injectors" with different characteristics and set to their own breaking pressures versus the real stock injectors (or in my case, non-stock injectors.)
* There is potential for the RPM of a chassis dyno machine or G-tech to be miscalibrated, however with the pump dyno giving an RPM readout resolution to the nearest single RPM, it would seem to be very trustworthy.
Jake
What account is this test giving of real life loading and the positive feedback that would come through the spaceship [oops aneroid]? Could this graph effectively be only showing a series of 'no-load' cruising speeds 'pencilled' together to form an artificial graph...
-
So... the mystery is: what is pulling back fueling on my pump as early as 4000RPM? Could it really be the main governor spring? Could there be some other unexplained phenomena at work here?
The "ski jump" like characteristic of the fueling curve approaching 4500RPM is also very intriguing. I'm not quite sure as of yet about what to make of this.
The only way I know of to solve that mystery, is to pull the pump apart and put somthing other then a spring in where the main spring is. I have used a piece of fuel line in the past, it has a little give like a 1/32nd - 1/6th of an inch. This would take the spring out of play and let you know if it is being pulled. My suspetion is that it is using the main spring, They had to have designed it that long for a reason. In my NON expert look at the graph, I would say that your shimming it of a 1/4" as i recall only gained you about 500rpm's. If you compare the slope of the stock graph where it begins to fall off, and the slope of the modified, they are approximatley the same from 3500-4000, at which time I would guess that the govnor completely overcomes the preload and pulls it down like stock. But like I said I'm no expert.
As far as the ski jump it might just be a dyno annomoly that happened that round? or it seems to be really close to where your timing adavace runs out, though I'm quite certian that fuel delivery and delivery timing are independent in the pump.
Finally what did the fuel injection shop have to say for the dyno? did they give any indication as to why the ski jump and why the fuel curve, or how to change the fuel curve?
You also need to try your modified timming cover on a dyno pull and see how much extra it gives you as of right now even if you had "full" fuel to burn to 5500 or 6000 rpm you would be running out of time to burn it.
...One more thought to this already long post... Mabey it drops at 4000rpm because the pump has hit the delivery wall?
Anyway lots to think about and discuss. :wink:
-
...One more thought to this already long post... Mabey it drops at 4000rpm because the pump has hit the delivery wall?
Anyway lots to think about and discuss. :wink:
If it was a max fuel available limit; wouldn't the slope drop off slower say 8cc @4000rpm and 4 cc @8000 rpm? Just a brain storm (all 10 cells working for that one!) :shock:
-
Hmm, maybe the plunger is skipping on the camplate at around 4500 rpm? I thought (dangerous thing) that I read once that was a problem with the TDI's when they approached 4500 rpms or so. I guess the only real way to know is to completely take the govenor springs out of the equation and see if fueling starts dropping the same way. I'd be willing to bet that it would.
Wonder if "pumpmaster Giles" could shed some light on this situation? ;)
Brendan
84 Scirocco 8v
00 Camaro L36 M49
-
All the data points shown for my "Fueling" plot as well as the "Fueling, Stock Spec" plot are at full throttle position and with 10.8 psi boost pressure applied to the LDA.
I talked about the question of why the fueling is dropping off earlier than the governor was expected to kick in with the guys at the diesel injection shop that helped with the test. The latest theory cooking down there is that there is a maximum flow limit being reached in the inlet port of the high pressure head.
Certainly "disabling" the main governor spring and retesting would help confirm or deny my belief that it's something other than the governor spring acting at 4000RPM, however it may not be practical for me to run the complete injection pump dyno testing process through again.
At first, I dismissed the possibility of camplate float, because the stock pump didn't seem to have any dropoff at 4000RPM nor did it seem logical for the camplate to float prior to the rated speed (which is above 5000RPM.) However, I am now re-thinking this possiblity. Pehaps at higher flow rates, a greater pressure drop is caused through the inlet port. This "increased suction" on the fill cycle would increase the tendency for the camplate to float now, wouldn't it!
The test and possible fix would be to shim the plunger return springs to a higher pressure and see if it makes any difference.
Now that I think more about it, I do recall noticing a pronounced noise coming from the pump happening at about 2000RPM (4000 crank rpm)... right where the dropoff in fueling starts!
-
Here is a corny video of me running the pump test bench for my first time while Jay is running my digital camera for his first time. I take it up to a little over 2000 pump RPM, then drop it down too far and it runs backwards a little. :oops:
http://media.putfile.com/VE_Pump_Test
-
hi guys
during my R & D with my pumps i have found that there is def camplate
float at higher rpms.
i've tried diferent spring rates and it helps allot.
Giles
-
Jake, I too was wondering about the fuel dropping after 4000 RPMs. Assuming that it is NOT the main gov'nor spring (which it shouldn't be, right?), then could it be that the pump is actually pumping at full capacity, and that volume per stroke is dropping? The line looks too steep for that, but it's a thought.
-
I injection pump dyno tested my GTD Autocrosser's VE injection pump successfully yesterday! Although it was more time/effort than I had previously expected, it was a lot of fun, and I feel that I came home with some really great data on my pump
looking at your graph you show that u have aprox 78cc at 1200rpm
i gather that is with full boost ?
the more important fueling is the no boost fuel off idle.
did u record that?
let me know
and yes the 86cc of andy's pump was not maxed out by any means
but he still had stock no boost fueling.
Giles
-
Stan: regarding fueling rates... We're probably over-analyzing this, but I added some calculations of the fuel curve plots, now showing the following:
Fueling (quantity per stroke, which would correlate with crank torque)
Fueling x RPM (or the fuel rate, which would correlate with horsepower)
Fueling x RPM x RPM (may not represent anything real, but nonetheless kind of interesting to look at)
(http://home.comcast.net/~vwgtd/GTD_VEPumpDyno_2-06-2006_fuelrate.png)
It is interesting to note that the fuel rate indeed peaks at 4000 crank RPM then starts a decline. Fueling*RPM^2 however is fairly constant from 4000-4800RPM, and then abruptly makes a downward turn.
-
It is interesting to note that the fuel rate indeed peaks at 4000 crank RPM then starts a decline. Fueling*RPM^2 however is fairly constant from 4000-4800RPM, and then abruptly makes a downward turn.
Thats interesting as my Giles powered AAZ pulls hard to exactly 4500 rpm and Still works ok to about 4700 rpm.I guess these 1.9er's arn't all that bad of revers afterall 8)
-
So... the mystery is: what is pulling back fueling on my pump as early as 4000RPM? Could it really be the main governor spring? Could there be some other unexplained phenomena at work here?
You will need to find some stiffer spring for your injection pump head. Maybe there is no cam floating with stock pump at 4000RPM... but I don't think your spring can still be call "stock" spring....they have worked a lot.
Maybe some shim would also help.