VWDiesel.net The IDI, TDI, and mTDI source.
Engine Specific Info and Questions => IDI Engine => Topic started by: crazycaddy on September 05, 2007, 03:19:48 am
-
Well, I guess I'm just to dumb or something but I can't figue quite what is the best way to control the vane actuator on my vnt15 turbo that I have on my 1.6L. Do I use a different actuator or use a manual valve or what. Can anyone help me out on this. This engine I am biulding is an experiment. :?
vnt15 turbo
custom IP
Merc injectors
3 inch exhaust
ect....
-
Here's a thread over at Jack's forum that has some thoughts:
http://vwdiesel.net/phpBB/viewtopic.php?t=8065
-
I think that setup looks more complicated than it has to be. I would think with a little trial and error you could make a normal wastegate actuator work. Otherwise I'd be tempted to run a line to the original VNT actuator with a solenoid controlling the bleed off (much like how the stock system worked). A small PWM circuit would probably do the trick.......just some random thoughts.
-
Thanx guys, this at least gets me leaning in one direction. :lol:
-
Little offtopic, but:
What merc injectors? 617? 617 turbo? 603? 603 turbo? 606? 606 turbo?
I´d say vnt15 is WAY TOO SMALL for 1,6 :roll: At least with modded pump etc.
-
I think the easiest way is to get an old wastegate assembly, take it all apart, cut off the wastegate valve off its stem, use a die and make threads on the shaft, put the can back together, screw on a turnbuckle type thing onto the shaft and screw that onto the vane actuator arm.
Then use a boost controller to set your boost. You may need to make some kind of cam system so the movement of the shaft moves the vane actuator the appropriate amount.
With that said I think VNT turbos really need to be controlled by an ECU due to the complexities of the vanes and rpms, engine load, throttle position, etc.
You can't use the actuator that's on the VNT because it works on vacuum and you need a "regular" wastegate actuator that works on positive pressure.
-
With that said I think VNT turbos really don't need to be controlled by an ECU. Due to the complexities of the vanes and rpms, engine load, throttle position, etc, a controller like Named Tin Tin's, Libbybapa's or Jimfoo's works just fine.
You can use the actuator that's on the VNT if you take it apart and convert it to work off boost instead of vacuum..
Fixed it for ya. :roll: :wink:
-
I use a VNT with a regular pressure actuator (wastegate actuator). Click on my "GT2256V" link in my signature.
That method works fine as a "get it running right now" option but it is not acceptable as a long-term solution, especially if you want high boost pressures. From everything between idle and your set high boost pressure, the vanes will be closed all the time. This is because the actuator will try to make the turbo get up to your set maximum boost regardless of engine load, engine speed or throttle position.
All that does is restrict the exhaust during low load situations (Idle, cruise, light acceleration, etc) which kills fuel efficiency. It makes the engine work hard to push the exhaust out of the cylinder and through the narrow restrictions of the closed vanes.
IMHO, a VNT turbo needs to be electronically controlled to do it right. The "complex" mechanical linkages that some have used are a good compromise but definitely not ideal for every situation.
-
Certainly less complex than having to put in a TPS, speed sensor and all the other associated electronics. I got an IDI to get away from all that stuff. Need to change a setting, just change a spring, or preload, or the length of an arm. Don't knock it till you've tried it. :twisted:
-
Hey Andrew, I might have missed it but do your vanes open at idle?
-
DEAR LORD WHATS WRONG WITH YOUR PIC ANDREW!!!!!!!!! :shock: :shock: :shock: :shock: :shock: :shock: :shock: :shock: :shock: :shock: :shock:
/hijack
-
It's as simple as I could come up with without any drivability issues. Because the setting of the vanes does not correspond to specific boost pressures (like a wastegate), but rather the rate of boost increase, if just a wastegate actuator is used to push the vanes open, you will have a very jumpy car at higher rpm transitions from cruise to acceleration.
Another criteria I was opting for was to have the vanes open at idle and open as much as possible at cruise in order for better engine efficiency. With simply a wastegate actuator on the vanes, the vanes will be in a closed position at idle and cruise with resultant increased backpressure, etc.
The third criteria was for a system that would fail to a safe mode. A wastegate actuator attached to the vanes will close the vanes (regardless of rpm) if you end up with a boost leak. That can overclock the turbo.
It would be significantly more complicated IMO to work out a circuit to control the vanes like the stock unit. You would need an accelerator sensor, rpm sensor and boost sensor and create a map to correspond to the various situations. If it is simple for you to do, then by all means please come up with a simpler control that is as effective, so I can use it on my next one. :wink: If a control system was created that was as effective, easier to install and overall simpler, I would be a repeat customer.
Andrew
I think we have the knowledge and the resources here to build a stand alone vnt controller. I would start with all my inputs needed,
1)a potentiometer for the tps
2) A 3 BAR MAP or some type of pressure sensor
3) temp sensor to monitor engine temp
4) engine RPM's from the alternator w
5) ?
Then choose a CPUmaybe a motorla 68hc11 or a PIC any 8- 16 bit micro controller will do. Nows the tricky part, finding someone who can write the program . The map and be stolen from vag-com'ing any tdi for base line maps to extrapolating the algorithms need for the program. Then make some graphs with a chiped tdi :wink: Granted thats a lot of data logging and math but again all possible with the right help.
Last would be a PWM signal output from the CPU to the N75 valve to contol the VNT actuator. The vacume could be provided from the vacume pump for the brake booster.
There is definatly a demand for a standalone VNT controller so why not try to make one :)
-
An op-amp would be wicked simple, but I have a feeling it would require calibration in most cases. The nice thing about a programmed microchip is, provided it had feedback in the form of boost, it would pretty much be plug and chug once someone wrote the code w/ the appropriate inputs, and I imagine we could even have different driving modes.
-
FWIW, just a simple wire to the vanes that corresponds to throttle lever position is a more drivable setup than just a wastegate can IMO, but you don't end up with any control over the max boost.
Andrew
maybe a simple wire in conjunction with an external wastegate plumbed to the egr port to control max boost?...(maybe someone already said this?)
-
Very very cool Andrew.... there's definitely a controller of some kind there.. might just be for the stepper motor but perhaps it's even smarter. Might be tough to reverse engineer the position curves it's mapped to, but you never know.
5 leads sounds like power, ground, vac sensor, throttle sensor, ???. Did a bit of googling but didn't find any schematics... you probably already have an engineer lined up but if not happy to help wherever I can.
-
I'd hate to bump an old thread.
I have an Arduino setting here infront of me looks to be perfect for VNT control. I do believe it's more than enough I/0 for this and pretty easy to program. PIC programming and setup is a bit more complex.
http://arduino.cc/
The best part: the boards are $20-30.
-
...and someones already in the process:
http://forum.arduino.cc/index.php?PHPSESSID=ldlp4u4sv27kjq81g3p6pocqg3&topic=204748.0
-
Because the VNT builds boost so high and fast I've found it's best to have the vanes closed at idle and drop off to wide open flow by the time it gets to highway speeds.
It's what it does in the middle that is up for grabs, when I had the manual controller on the dash; I adjusted according to EGT temps.
The aircraft spruce EGT probe reacts so fast, it makes a great load sensor.
Somewhat like a vacuum gauge on a gasser car.
What EGTs/Boost levels are most people running on the highway?
-
Because the VNT builds boost so high and fast I've found it's best to have the vanes closed at idle and drop off to wide open flow by the time it gets to highway speeds.
It's what it does in the middle that is up for grabs, when I had the manual controller on the dash; I adjusted according to EGT temps.
The aircraft spruce EGT probe reacts so fast, it makes a great load sensor.
Somewhat like a vacuum gauge on a gasser car.
What EGTs/Boost levels are most people running on the highway?
And you could indeed include the egt sensor in the program. When I figure out how this thing works, and to program for it, I'd like to do alot. An LCD display or two could give you info, a few buttons to program what boost, how fast of spool and all that one the fly....It's pretty powerful
It's going to take awhile for me to build something, as the most programming I've done was today (besides the normal "hello world" deal) and it was just a digital led controller.
-
http://dmn.kuulalaakeri.org/vnt-lda/
not really interested until the tach-from-alternator thing works, myself
plus i think it would be really interesting to see what the possibilities are with a stepper linear actuator controller rather than using a vacuum valve
-
http://dmn.kuulalaakeri.org/vnt-lda/
not really interested until the tach-from-alternator thing works, myself
plus i think it would be really interesting to see what the possibilities are with a stepper linear actuator controller rather than using a vacuum valve
I'm thinking servo control, doable with arduino
-
Someones already doing it: http://forum.arduino.cc/index.php?PHPSESSID=ldlp4u4sv27kjq81g3p6pocqg3&topic=204748.0
I'd like a GUI bit to adjust the variables and how they affect the vanes position.
There are plenty of projects with a tach. I'll have to look into it.
(I thought I linked this here yesterday....)
-
Thats all been done aswell: http://dmn.kuulalaakeri.org/vnt-lda/
looks like I need to find me a VNT!
-
You mean something like this?
(https://scontent-a-sea.xx.fbcdn.net/hphotos-prn2/1520670_10152115944941427_583343293_n.jpg)
My "end result" is based on the DMN project linked above. I've already set it up to use a serial LCD (so I can save some pins) and I'm going to be running my EGT probe through it too. I'm also including a datalogger shield so I can record actual vs. requested boost along with EGT and TPS for tuning purposes. I just ordered a second pressure sensor too so I can also measure EMP. :D
-
It'd be great to see this sort of thing integrated into the MultiDisplay project, which already has a wastegate boost controller w/ full PID.
http://mdforum.designer2k2.at/
'course, I think it would be great if MultiDisplay board sets and/or kits were readily available and cheap.
-
You mean something like this?
(https://scontent-a-sea.xx.fbcdn.net/hphotos-prn2/1520670_10152115944941427_583343293_n.jpg)
My "end result" is based on the DMT project linked above. I've already set it up to use a serial LCD (so I can save some pins) and I'm going to be running my EGT probe through it too. I'm also including a datalogger shield so I can record actual vs. requested boost along with EGT and TPS for tuning purposes. I just ordered a second pressure sensor too so I can also measure EMP. :D
Mind making a build thread?
That's amazing.
-
It's (sort of) in my build thread over in the TDI section. Got a bunch more done on it today - EGT sensor is hooked up along with the control transistors for the "N75" valves.
Once I have everything working perfectly I'm going to actually solder it all up. For the moment it is all on a breadboard.
-
I've been learning about how to work with arduino.
I really appreciate all the work you've done. I see you've been working on this project for awhile.
The info in the links is a little over my head at the moment,... gave me some new terms to search, ha.
-
If all you want to do is just the VNT then it's actually pretty simple. The code on the site there needs some adjustments to actually work but nothing major (the PWM frequency is waaaaay too high for instance). All the OTHER stuff I'm doing with it gets a bit more complicated. I definitely don't mind sharing though!
-
If all you want to do is just the VNT then it's actually pretty simple. The code on the site there needs some adjustments to actually work but nothing major (the PWM frequency is waaaaay too high for instance). All the OTHER stuff I'm doing with it gets a bit more complicated. I definitely don't mind sharing though!
Hello guys,
I have a golf 2 1.6L turbo diesel that a hope to fit with an arduino-controlled VNT one day. Have been looking all over the internet on this and found 2 usefull sources:
1) http:// dmn.kuulalaakeri.org/vnt-lda/ (no recent submits from what I can tell)
2) https:// github.com/vanbcguy/vntlda (most recent source that I could find)
I was actually wondering if there has been any progress on this matter as I failed to find a fully operational VNT controller that acts similar to a stock ecu-controlled vnt (i.e. vanes fully open at idle and as open as possible at cruisin speeds for max fuel efficiency and so on, inputs being TPS RPM and MAP).
I'm new here and english isn't my first language so please bear with me.
Thank you in advance.
-
That's what my code on the github link there does...
-
That's what my code on the github link there does...
Do you have This fully and succesfully operational on your car?
And what extra hardware would be needed to have this properly working an IDI diesel since my engine is still stock (only gov mod, boost pin set steeper and K14 instead of K24 for quicker spool)
-
Yup, I'm fully operational. It's been working quite well. I could probably still get the boost control a LITTLE bit tighter but it's in the "good enough" camp these days.
Extra hardware... You need something to give a TPS signal. I'm using an injection pump that has one on it from the factory but you can probably find something that will work. You'll also need a tach signal. I'm using the TDI crank sensor for mine but there's other ways of doing that.
-
wow man that's nice, is there also someone with an arduino-controlled VNT on his 1.6TD IDI?
i'm trying to find someone who allready has done this on an IDI with a mechanical pump
to find out if this is doable for me because I don't have much experience
but i'm very very interested in this concept?
-
keep it simple, like its done in Portugal
-
keep it simple, like its done in Portugal
There are some very good reasons to not do it that way.
-
keep it simple, like its done in Portugal
There are some very good reasons to not do it that way.
How do you mean Libbydiesel?
-
Reasons to not do it that way include but are not necessarily limited to excessive back pressure, over-speeding turbo, worse fuel economy and boost spikes when transitioning from cruise to max acceleration.
-
I forgot, 'fail to damage' in the event of a boost leak.
-
Can you explain "fail to damage" in the event of a boost leak ?
-
Sure. With just a boost can, if you are at max boost and a hose pops off, boost pressure will drop to zero and the vanes will snap shut causing the turbine to accelerate at maximum acceleration and the turbo will most likely self destruct. Hopefully you lift your foot off the go pedal when you hear/feel the issue, but with just a boost can operating the vanes, that will have the sole effect of reducing fuel. With other better controls, the vanes will open as you lift your foot and you are far more likely to escape the situation with a turbo/engine that is intact. You then reattach the hose and motor on rather than replacing the turbo and possibly a bent rod or four... Better yet would be code that recognizes a significant disparity between pedal position, rpm and what the manifold pressure reading is currently and what it should be and opens the vanes - limp mode.
-
Thanks libbydiesel.
Btw, haven't you tried an Arduino controlled VNT on your IDI? What are your thought on this method?
-
I have not yet. I was working on a project very similar to vanbcguy's but did not complete it. I think it is an excellent way to go and am sure I will use it in the future.
-
Thanks for the reply, spoke to mate at work about this, he actually had the hose come off at full boost, and yes it lunched the turbo only, Not sure why the con-rods would bend from a turbo hose coming off, only ever seen bent rods from engines which have ingested liquids, but i will bear it in mind and ensure all hoses are well secured,
-
A common failure mode of the VNT15 is to snap its shaft which then causes pressurized oil to dump into the intake and hydrolock the engine. Bent rods from failed turbos are not uncommon on the ALH that came stock with the VNT15.
-
I think i was having a senior moment ! when you said if a boost pipe came off etc i was thinking of the pipes from turbo to inter cooler to engine, not the link pipe on the turbo to the boost can ! Duh ! I have a good quality piece pf tubing on my setup,
I could put a metal pipe with a banjo fixing and a real short piece of tube to further reduce the risk, but for now I´m happy with the set up i have, Been looking around at easier alternative to make it vacuum controlled, came across an Australian website
where they use a Dawes Valve to control the VNT. Anyone have experience of this setup ?
-
I think a Dawes valve is just a fancy name for a manual boost controller, which is basically a spring check valve with adjustable spring preload. I used to run one on my Corrado.
Edit: the Dawes valve would control the input signal to a WG diaphragm setup...I think it would have the same limitations and risks that libbydiesel discussed earlier.
-
If ANY boost connection comes loose anywhere either between your intake scroll to boost can OR from compressor outlet to intercooler to intake manifold, the same situation can occur. In the last decade or so I have had two different couplers 'pop' despite everything being attached properly.
-
@jmaddocks The boost controller you had on your corroda was for a waste gated turbo,
the dawes valve set up i looked at was usuing a vacuum can on the turbo supplied from the vacuum pump, the vacuum was regulated by suppling boost pressure to the other side of the valve.
@libbydiesel, the senior moment was oil inthe engine bending rods with the main boost pipes between scroll and inlet manifold, and thanks for the warning.