VNT20 M-TDi Questions

[Henk]

Good idea, yeah could do that. trying to re-design it to be a little tidier, got a manifold now so will make an adaptor in work when I can
No GT15, that's on the MK4 PD engines, mine was a K03 I think, but 'twas shagged, started scoring the housing on the cold side.
The whole engine has been stripped to be cleaned, painted and hopefully re-assembled. except now all the bolts have rusted 
So I guess either it will work, or I'll realise i was never destined to be a mechanic after all 

[keaton]

electronic VGT controller > mechanical controller...

see my sig for a cheap standalone controller, can even do LDA control if you want  

8x8 map for VNT & LDA control
 Use N75 solenoid based actuators
 Live configuration via USB interface and graphical map editor (no special software needed)
 Two VNT control modes:

    * Duty Cycle map - Uses RPM/TPS info to set VNT actuator position. Additional dampening can be added to improve engine response. No map sensor needed for this setup.
    * Target Pressure map - Specify requested pressure value on each control point. Vnt-lda will automatically move actuator until required pressure level is reached.

LDA table can be used to fine tune soot emissions and torque limiting on certain RPM-range.
Cheap and free :-)

    * RPM Engine speed sensor, use Honeywell GT101DC or alternator W-output as a source.
    * TPS Throttle position sensor (like Bosch "001")
    * MAP Manifold absolute pressure sensor for detecting amount of boost, for example 4bar Freescale MPX6400
    * Arduino AVR-microcontroller board, servos, n75 (or similiar) solenoid & vacuum actuators

[theman53]

Lord verminaard and I have been over this a long time ago, we might have even recommended the link to you, but this is NOT a stand alone controller. As quoted you have to buy lots of pieces to do the mechanical bidding, and the Arduino board. That site only gives you the program that is ready to run in the board. You will have several hundred dollars buying all new parts and then you have to make them work. You can't just buy a servo and bolt it on to the stock bracket...there is no stock bracket on these older cars.

It is a good start what you have but the picture painted makes it seem like it would be easier/cheaper than making a mechanical setup and I don't see how it is.

[keaton]

Lord verminaard and I have been over this a long time ago, we might have even recommended the link to you, but this is NOT a stand alone controller. As quoted you have to buy lots of pieces to do the mechanical bidding, and the Arduino board. That site only gives you the program that is ready to run in the board. You will have several hundred dollars buying all new parts and then you have to make them work. You can't just buy a servo and bolt it on to the stock bracket...there is no stock bracket on these older cars.

It is a good start what you have but the picture painted makes it seem like it would be easier/cheaper than making a mechanical setup and I don't see how it is.

 
that is just not right...
I was trying to make a standalone VGT controller and found this. not several hundred dollars to implement, your just wrong.
arduino = $35
Power FET = $4
LM317 (5v reg) = $2-6
Map sensor (not required) = $15 (or junk yard)
TPS=... junk yard...
RPM =... $40 brand new... or junk yard (BMWs) or alternator mod
N75 valve = junk yard... or brand new...
proto board = $2
software = free

vacuum actuator = already on the Turbo...
so how do you come up with several hundred dollars if you already bought the VGT and have the Vanes fixed or using a spring & boost can?

[theman53]

TPS and N75 and vac can are all used in your deal above. I did state for new parts. IF I go to all the trouble of doing this I would not buy used stuff and have to trouble shoot after hours of fab work when one of the used parts is acting up. I will provide links and not just any links with the highest prices possible, but the most economical links I found when I was going to do this.
http://octopart.com/1gt101dc-honeywell-7370636    Hall sender 23.00
http://www.thepartsbin.com/repsite/bosch~throttle_position_sensor~reparts.html    Bosch TPS 47.00
http://compare.ebay.com/like/290481934963?var=lv&ltyp=AllFixedPriceItemTypes&var=sbar&_lwgsi=y&cbt=y   N75 valve 55.00
http://metalmanparts.com/product.sc?productId=352&categoryId=52       Vacuum can  105.00

Still have to buy arduino too 35.00     .... BTW we are over 200.00....
and the linked site you gave says to have a map sensor to "Target Pressure map - Specify requested pressure value on each control point. Vnt-lda will automatically move actuator until required pressure level is reached" and they recommend a  "4bar Freescale MPX6400" which to my knowledge is NLA but they have a replacement and they are fairly cheap but I don't have a link for that sorry.

Now the fact that I have bought over 200.00 in parts and now I CANNOT just bolt it onto the car I have to fab up the control system as they didn't come with this stuff factory I add in a factor that isn't cheap for me...my time. This is how I came to the several hundred dollar conclusion. If I were going to get all used stuff I could do that with used all thread, used springs, and used hiems joints and have less than 20.00 in a mechanical setup. Probably just as much setup time.

Also, I have 2 VNT turbos, both had the Vac can robbed off of them, so you cannot assume that all VNTs will have them on when delieverd.                   

[R.O.R-2.0]

im far from sold on the electronic control methods..

i still think the mechanical methods are better for the mechanical engines..

you can do a mechanical setup for ALOT cheaper than 200 bucks. plus, you wont spend near the time building the linkage setup..

[libbydiesel]

I find the electronic controls interesting but if going electronic I do not think I would take the approach linked above.  The vac actuator and n75 valve was a kluge from the factory and the primary cause of the bad rap VNT turbos often get.  If going electronic I would use a stepper motor actuator from a Sprinter.

At the same time, I would not assume that the mechanical controls are easy or quick to build.  The parts are fairly inexpensive, but it is a fair bit of time/effort to get it set up exactly right.  Without everything moving just right, one will end up with vane sticking issues or inaccurate max boost control 

[keaton]

TPS and N75 and vac can are all used in your deal above. I did state for new parts. IF I go to all the trouble of doing this I would not buy used stuff and have to trouble shoot after hours of fab work when one of the used parts is acting up. I will provide links and not just any links with the highest prices possible, but the most economical links I found when I was going to do this.
http://octopart.com/1gt101dc-honeywell-7370636    Hall sender 23.00
http://www.thepartsbin.com/repsite/bosch~throttle_position_sensor~reparts.html    Bosch TPS 47.00
http://compare.ebay.com/like/290481934963?var=lv&ltyp=AllFixedPriceItemTypes&var=sbar&_lwgsi=y&cbt=y   N75 valve 55.00
http://metalmanparts.com/product.sc?productId=352&categoryId=52       Vacuum can  105.00

Still have to buy arduino too 35.00     .... BTW we are over 200.00....
and the linked site you gave says to have a map sensor to "Target Pressure map - Specify requested pressure value on each control point. Vnt-lda will automatically move actuator until required pressure level is reached" and they recommend a  "4bar Freescale MPX6400" which to my knowledge is NLA but they have a replacement and they are fairly cheap but I don't have a link for that sorry.

Now the fact that I have bought over 200.00 in parts and now I CANNOT just bolt it onto the car I have to fab up the control system as they didn't come with this stuff factory I add in a factor that isn't cheap for me...my time. This is how I came to the several hundred dollar conclusion. If I were going to get all used stuff I could do that with used all thread, used springs, and used hiems joints and have less than 20.00 in a mechanical setup. Probably just as much setup time.

Also, I have 2 VNT turbos, both had the Vac can robbed off of them, so you cannot assume that all VNTs will have them on when delieverd.                   

sounds like your not very resourceful and lazy. you will go to all the effort to swap a TDI into car, spend the time to convert an injection pump form electronic to manual ( usually taking an IDI IP and a TDI e-IP and making one) then complain about having to make a bracket or 2. you will also claim that its SSOOO expensive to do... lets see, a "$200" control system to prevent blowing motor or head gasket. how long does it take to replace a head gasket? material cost? now you claim your time is expensive, you can just guess a $/hr rate and do the math and then add that to the cost for your head gasket replacement...

if you want to make this super cheap you can get a Atmega238p with the arduino boot loader programmed into for $5 + shipping, so no need for a $35 dollar board, you then need some caps, a crystal a few resistors ($10 total)....

the MPX 6400 is still available for ~$10
http://www.digikey.com/scripts/dksearch/dksus.dll?vendor=0&keywords=mpx+6400

you can also go to freescales site and request an engineering sample for free. again only needed for target boost mode, not needed for duty cycle mode...
you can also use other analog map sensors, just some calibration is required & 99% of the time the math profile is given in the datasheet.
you can also go to Honeywell's site and request an engineering sample again, for free. or you can use the W-post mod to the alternator ~$2(wire) & 1hr

you don't need a Bosh 101 TPS, a 10k pot will do the exact same thing, or go to a junkyard and pull a TPS off any 3/5 series BMW form 1992-1999 with pigtail for maybe $5-$10 ( never seen one go bad, doesn't mean they don't)

if you need to buy a vacuum can, spend the few extra dollars and get the post 2000 on so you get a linear 10k pot for feed back and wire that into ATmega328p on an analog port for feed back... just gotta add about 6 lines of code. you can even add the PID library (free) and 10 lines of code...

again, this is an afternoon to build & mount, & a day or 2 to fine tuning.

your still gonna spend time tuning the spring, threaded rod(s), and vacuum can. electronic version I can change is minutes... gotta go to emission testing, slap a new map in there so you don't smoke, then go right back to a power tune map...

if you event want to get carried away you can add a LCD and mount it to the dash for live feed back for $3-$20 if you know where to look and what you want...
http://www.bgmicro.com/24x2lcd.aspx

http://www.sparkfun.com/products/709
(I have both of these, the Black on white one is very nice)

im far from sold on the electronic control methods..

i still think the mechanical methods are better for the mechanical engines..

you can do a mechanical setup for ALOT cheaper than 200 bucks. plus, you wont spend near the time building the linkage setup..

all engines are mechanical.
time is minimal. soldering is not that hard or time consuming

200 or less vs a new motor... your right, electronic feedback controls are inferior

I find the electronic controls interesting but if going electronic I do not think I would take the approach linked above.  The vac actuator and n75 valve was a kluge from the factory and the primary cause of the bad rap VNT turbos often get.  If going electronic I would use a stepper motor actuator from a Sprinter.

At the same time, I would not assume that the mechanical controls are easy or quick to build.  The parts are fairly inexpensive, but it is a fair bit of time/effort to get it set up exactly right.  Without everything moving just right, one will end up with vane sticking issues or inaccurate max boost control 

these is a added line of code that sweeps from open to close to open once upon start up.
the stepper is a nice option but requires more hardware (both electrical and mechanical).

I guess you cant please them all, this is kind of a retrofit kit using the existing vacuum can.

if you got a source of sprinter actuators (for free) i can help break into the protocols that move it... there are some CAN BUS ones out there that should be very easy for me to break into as i have done lots of work with CAN BUS and the protocols.

[theman53]

When I laid it all out as I did I hit over 200.00 with more to buy, and you had no comment on that. I wasn't saying the electronic option wasn't doable or a bad option. I stated it wasn't bolt and go and that is wasn't as cheap/easy as you are portraying. I was giving a different view of your perception, not knocking it.

This is in the TDI section, but you posted it everywhere. I am more thinking about the IDI's. Even the M-TDI, I guess. If you have the option to run the ECU and the setup exactly like it was factory then I wouldn't build another electric controller to reinvent the wheel. That is just me, personally I would rather run mechanical controls if I have a non ECU car.

You can call me names like lazy and not very resourceful, but my point was proven. You look less credible by doing that and that wasn't what I was trying to accomplish.

Oh yeah...this is your first warning, More name calling*me or others* and you will take a weeks vacation. If it continues you will get banned for longer. Have a nice day 

[keaton]

When I laid it all out as I did I hit over 200.00 with more to buy, and you had no comment on that. I wasn't saying the electronic option wasn't doable or a bad option. I stated it wasn't bolt and go and that is wasn't as cheap/easy as you are portraying. I was giving a different view of your perception, not knocking it.

This is in the TDI section, but you posted it everywhere. I am more thinking about the IDI's. Even the M-TDI, I guess. If you have the option to run the ECU and the setup exactly like it was factory then I wouldn't build another electric controller to reinvent the wheel. That is just me, personally I would rather run mechanical controls if I have a non ECU car.

You can call me names like lazy and not very resourceful, but my point was prove. You look less credible by doing that and that wasn't what I was trying to accomplish.

I did comment on your price layout, and told you where and how to get the parts cheaper. Engineering samples, I get them all the time for work, personal designs, and tinkering.

its fine, don't use it. I really don't care. using the argument of not "bolting on", having to make bracket(s) and it costs money is silly.. clearly the AHU with its fixed gate turbo and putting a VNT is just a bolt on out of a kit...

clearly, see the OEM welds on the IP throttle bracket, see the OEM cut metal with the bolt for a pivot point


yes I posted it in several place, not many people know about it. if not many people know about it then it, then its hard to use as on option...

Oh yeah...this is your first warning, More name calling*me or others* and you will take a weeks vacation. If it continues you will get banned for longer. Have a nice day 

sorry I upset what seems to be sensitive feelings over the internet 

The thought that an electronic control is required to prevent engine damage is nonsensical.  The only situation that would cause damage to the engine is vanes stuck in the closed position which is actually more likely due to the crappy vac can/n75 control.

electronic control is required to prevent engine damage is nonsensical? really? talk about nonsensical.... companies wouldn't spend millions in the electronic control systems if they didn't prevent damage... 

http://www.youtube.com/watch?v=v30q6VkphbE

thank you grammar police

[theman53]

Quote from: libbydiesel
The thought that an electronic control is required to prevent engine damage is nonsensical.  The only situation that would cause damage to the engine is vanes stuck in the closed position which is actually more likely due to the crappy vac can/n75 control.
electronic control is required to prevent engine damage is nonsensical? really? talk about nonsensical.... companies wouldn't spend millions in the electronic control systems if they didn't prevent damage..


I think what he is saying is the "electrical" part of that...not the vane control part of that.

I would lock this thread, but your intelligence is too awesome to keep locked up.

P.S. I did not get "upset" by your comments. I have many threads on this site if I needed to prove that I can make stuff. I am just letting you know as I have Mark, Toby, and several others the forum rules and the way things go around here. I don't ban first offense or perma ban after second as a general rule as some have done. Look in the General and FAQ sections, I think Vince posted something like : if you wouldn't talk to a child or grandparent like that then don't click post. I am just calling you on that. I wasn't the only one that noticed either.

I still never said it was a HORRIBLE IDEA...I said it is more work and cost than what you laid out. I know you "could" find some cheaper parts, but I have found in my bass guitar that using a pot made in some 3rd world country only lasts about 4 years of playing, and I don't use the tone knob that often. I would rather just buy a nice bosch part that was made for that kind of abuse.---- Again, my personal preference.

[libbydiesel]

Keaton, nice engine pic!  Beautiful.

Do you have links to some threads/blogs of functional completed projects using the electronic control?  I'd love to see pics of your vehicles that are running that control method and the details on which turbos, which parts you used, the necessary fabrication, etc, etc...  I'd also be interested in hearing how you monitor the vane motion.  Do you have one of the actuators with the potentiometer?  How many miles have you been running it and what issues have you seen?   

[keaton]

Keaton, nice engine pic!  Beautiful.

you should know, you posted the link 

Do you have links to some threads/blogs of functional completed projects using the electronic control? 

nope

I'd love to see pics of your vehicles that are running that control method and the details on which turbos, which parts you used, the necessary fabrication, etc, etc... 

well I guess I cant answer you, because I don't own a diesel or a mechanically governed one. I put one on a Benz 300D that my friend owns.
I even went to the point of laying out a PCB, etching it and soldering it so it plugs right into an arduino. I don't have pics of the build and i'm at work right now... but when I get home I can post up the pics & gerbes of the "sheild" with the schematic. put it in a fried ECU after removing the PCB ( water proof box and connector to harness)

TPS: used a left over BMW Throttle body and TPS.
RPM: cam sensor BMW 93-95, it creates a pulse that goes to in interrupt (pin2)

as an engineer and a person who truly enjoys designing electronics, system designs, building them and programming them I did all the electrical work and my buddy did most of the mechanical (doesn't mean i can't do it).

I'd also be interested in hearing how you monitor the vane motion.  Do you have one of the actuators with the potentiometer?  How many miles have you been running it and what issues have you seen?   

information is sent to a 24x2 LCD that sits on the dash. The actuator has the vacuum nipple on the side and the 3 pin electronic plug on the top, and i preform an analog read on he feed back..

issues: I'm always wanting to tweak the code and tune and he doesn't want me to touch the car because he is happy with how it works..... 

miles on the setup ~10-12k

working on writing a EGT protection setup, user sets max EGT and when hit, the vanes go full open and fires a 2nd N75/N18 valve that cuts the pressure to the LDA/ boost enrichment (goes to open atmosphere= no enrichment )

[libbydiesel]

The beautiful engine comment was tongue in cheek.

Do anyone know of an inexpensive source for the smart vane actuators?  The least I've seen them new so far is $240 from IDparts and they're hard to find used.

[CRSMP5]

http://www.xsboostturbochargers.com/ is who id call libby..