Somebody PLEASE do a direct PWM turbo controller already!!!

keaton

Veteran Member
Joined
Apr 21, 2009
Location
Mesa AZ, 85202
TDI
TDI-less ATM
keaton - thats pretty cool.

If we know what frequency the bus runs at and what node addresses the turbo actuator and ECU are, the rest is easy(ish). Record all can messages on the bus and pick out the messages to/from the actuator over a wide range of commands to get a feel for the data range and resolution of the messages and if there are any enable messages or checkup bits that need to be set.. The controller (yet to be designed) then just needs to take a physical input (from N75 command) and spit out CAN messages (which is the easy part).
iirc CAN runs at 250kbs & 500kbs... i would have to go back and look at the code header files.. I wanna say 500, i think that what i'm running at in all my stuff...

if someone has a CAN based controller they can send my i can start working on it. tried to do this once before the the shipping was like $60+...

also got a bunch of Attiny167 with build in LIN support & a munch of CAN chips. :D
 

chapelhill

Veteran Member
Joined
Aug 19, 2005
Location
Scotland
TDI
03 Ibiza pd13- 2260vk Turbo etc.., Merc E280cdi
If eventually the ecu is not going to be used for all the control strategy (probably more relevant to two turbo set up), then if we use boost to give some feedback, how is the boost signal to be processed. When logging boost, emp and AFR using a logger at 50 samples per second it is easy to see that the instantaneous boot pressure fluctuates by around 10% on 2260 at full boost, and a 65mm compressor wheel had nearly double that. As far as the control system is concerned this is noise.

I am not sure how the ecu deals with that, either through the I element of the PID, or if there is some other pre processing of the signal. Indications are that we may have substantially higher fluctuations than OEM.

As I have pointed out earlier I have seen some instances of boost instability, when pushing the performance envelope of the 2260, but to eliminate it has taken a much higher reduction on boost limits than I would have thought necessary.
 

Fix_Until_Broke

Top Post Dawg
Joined
Aug 8, 2004
Location
Menomonee Falls, Wisconsin, USA
TDI
03 Jetta, 03 TT TDI
J1939/11 and 15 run at 250k but that's primairly for off highway vehicles and diesel engines. I don't remember the on highway standard, but I'd bet it has the same physical layer, just a different message structure. CANOpen can run 125, 250, 500, 1M, etc but 250 and 500 are the most common that I've seen.

It sounds like the CAN based actuator is not the most useful one due to it's orientation.

iirc CAN runs at 250kbs & 500kbs... i would have to go back and look at the code header files.. I wanna say 500, i think that what i'm running at in all my stuff...

if someone has a CAN based controller they can send my i can start working on it. tried to do this once before the the shipping was like $60+...

also got a bunch of Attiny167 with build in LIN support & a munch of CAN chips. :D
 

Scott_DeWitt

Vendor
Joined
Apr 7, 2004
Location
Texas USA
TDI
2000 Audi A4 1.9TDI quattro
Wouldn't a remote mounted actuator be feasible? Mount actuator in a more convenient location and then run a cable from actuator to turbo.
 

Rub87

Top Post Dawg
Joined
Dec 10, 2006
Location
Belgium
TDI
Ibiza '99 90HP
If eventually the ecu is not going to be used for all the control strategy (probably more relevant to two turbo set up), then if we use boost to give some feedback, how is the boost signal to be processed. When logging boost, emp and AFR using a logger at 50 samples per second it is easy to see that the instantaneous boot pressure fluctuates by around 10% on 2260 at full boost, and a 65mm compressor wheel had nearly double that. As far as the control system is concerned this is noise.

I am not sure how the ecu deals with that, either through the I element of the PID, or if there is some other pre processing of the signal. Indications are that we may have substantially higher fluctuations than OEM.

As I have pointed out earlier I have seen some instances of boost instability, when pushing the performance envelope of the 2260, but to eliminate it has taken a much higher reduction on boost limits than I would have thought necessary.
Boost input is passed trough a PT1 filter, the time can chosen in the dataset
 

JFettig

Vendor
Joined
Aug 18, 2010
Location
Blaine, MN
TDI
B5 Passat, 2010 Jetta
^exactly what I was thinking.. a brief look at my 2056VK, it would be super easy to extend.
 

Fix_Until_Broke

Top Post Dawg
Joined
Aug 8, 2004
Location
Menomonee Falls, Wisconsin, USA
TDI
03 Jetta, 03 TT TDI
Remote mounting and cable actuation opens a lot of doors for actuator options. Heck, I'd like to have that on my vacuum operated actuator for easier VNT adjustments :)

There are millions of gasser cruise control actuators that could be used. A large number of industrial controls with position feedback options are available.

So if it's this easy, why don't the OEM's do this? Not knocking the idea, but it seems too simple to have not been thought of before.

The spring on the actuators is pretty stiff so you'd probably need a guided cable assembly, one that would be relatively low friction and take the heat. A shifter cable would be a great example - guided for push/pull, takes a lot of force, smooth/low friction - I'm not sure if it would take the heat or not?

Tangent thought - Attention vendors - build a remote actuator kit where the cable mounts to the existing actuator bracket on the turbo and you can remote mount your existing VNT actuator on the firewall or valve cover or wherever you want for easy adjustments. Sell it for $100 and you'd have people beating your door down to buy it.
 

Fix_Until_Broke

Top Post Dawg
Joined
Aug 8, 2004
Location
Menomonee Falls, Wisconsin, USA
TDI
03 Jetta, 03 TT TDI
My suggestion was off topic of this thread - sorry about that.

A cable, one bracket and maybe a jam nut or two should be able to be sold for $100, make a decent profit and cover customer support.

This is not a solution for the EDC15 ECU to drive the newer actuators discussed on topic in this thread.
 

Rub87

Top Post Dawg
Joined
Dec 10, 2006
Location
Belgium
TDI
Ibiza '99 90HP
It has been done driving such an actuator, ask T3-UN1 ( I think his nick was like that) then I would use the precontrol maps of a car originally equipped with such an actuator as starting point, as 100% means there vanes closed, while in edc15, 80% already means vanes to the stopscrew.

now one would need to use the N75 map also for stopscrew setting, or be careful with it, as when you command the actuator to pos x and it it his the stopscrew before reaching pos x it will stop working I think

Good and cheap PWM actuator can be found on the KKK bv50 etc from early audi 3l, turbos fail often so actuator should be cheap, ideally to be used with the cable idea
 

brum

Veteran Member
Joined
Nov 30, 2009
Location
Bulgaria
TDI
Passat B5, 1.9 TDI, AFN
The only situation where the factory clocking would nearly work is with a top mounted turbo and all custom pipework and oil lines, might need a funky layout to work well in a packed bay too.

VAG GTB's have the CAN-BUS actuator AFAIK
About the CAN-BUS actuators on VAG GTB turbos - I have two and they are with 5 pins, two directly wired to the motor and 3 wired to the position sensor. So these are for sure type 1.
 

Rub87

Top Post Dawg
Joined
Dec 10, 2006
Location
Belgium
TDI
Ibiza '99 90HP
not much to do with these as the motor controller aint there =) better take one off MB or volvo etc
 

Digital Corpus

Top Post Dawg
Joined
Mar 14, 2008
Location
Ontario, California
TDI
'97 B4 w/ 236K mi body, 46K mi soul
I case CAN operation is still on the table, I had someone point me over towards Microchip's PIC32 line of MCUs for a project of my own. They have up to 2 CAN busses per chip, though their compiler is where they make the money and there isn't a simple alternative.

Keep in mind that running a remote cable ot the turbo needs to be able to withstand a tremendous about of heat and the cable length can influence the maximum bandwidth. Shielding the cable in a sheath of metal weave can help with both.
 

engineered2win

Veteran Member
Joined
May 24, 2011
Location
Dublin, OH
TDI
MkVI Golf TDI
J1939/11 and 15 run at 250k but that's primairly for off highway vehicles and diesel engines. I don't remember the on highway standard, but I'd bet it has the same physical layer, just a different message structure. CANOpen can run 125, 250, 500, 1M, etc but 250 and 500 are the most common that I've seen.

It sounds like the CAN based actuator is not the most useful one due to it's orientation.
All the VG actuators I've worked (Holset and Garrett) with are 500k, which is the current canbus standard. J1939 is a subsection of canbus and runs at 250k and is used on over the road trucks because of the maximum backbone length. The truck ecm's have separate 250k and 500k bus' to make everything work. You can not plug a 250k baud device into a 500k baud backbone and get it to work, and vice versa. You will get erratic/abnormal update rate errors and it will hose up everything on bus. I've screwed that up before so I know what happens.

Also it will be difficult to come up with a remote mounting scheme that can withstand the heat. The Holset vg actuators are cooled with coolant, while the turbos themself are generally only oil cooled and lubed.
 
Last edited:

Fix_Until_Broke

Top Post Dawg
Joined
Aug 8, 2004
Location
Menomonee Falls, Wisconsin, USA
TDI
03 Jetta, 03 TT TDI
I know you must have the same frequency of all nodes on the bus - the higher the frequency, the less tolerant of wiring/shielding/etc the network is.

I don't think a remote mounting scheme would be that hard to cool, but then again, I've never done it so what do I know?



So - has anyone decided what actuator and what ECU need to work together?
 

Deus Violentia

Well-known member
Joined
Oct 12, 2009
Location
Germany
TDI
2.0TDI Polo 9N
Hey guys as far as I understood we should go for a hybrid actuator made out of the MB and VAG ones in order to use it as PWM controlled actuator right?

I just came across this as I plan to mount a GTB2260VK on my BKD which has EDC16U1, still wonder if the PWM for the N75 is some kind of comparable between the EDC15x and EDC16U1?

I´ve learned that the PWM should be at 140Hz right, what exactly is the modification thats done within the dump of the ECU?

Btw. got a original VAS6395/1 here which is the orig. VAG VNT tester for those turbos, with which you can adjust the VNT by connecting this handtester to +12v, GND and the plug at the actuator. I may take it apart and make photos of what is inside if someone is interested?
 

Rub87

Top Post Dawg
Joined
Dec 10, 2006
Location
Belgium
TDI
Ibiza '99 90HP
guess will be just a pwm signal with around 140-150hz.. for the solenoid it doesnt matter wether its 135 or 145 hz..
 

Deus Violentia

Well-known member
Joined
Oct 12, 2009
Location
Germany
TDI
2.0TDI Polo 9N
btw. anyone got the pinouts for those different actuators from MB, VAG and Ford(?) with PWM and(or) CAN?

I took a look at the actuators that are mounted at the V10 TDI´s with EDC16U1 and EDC16U31. Found out that those have CAN and Pin 5 at the actuator is also occupied going directly into the ECU. It´s labeled with a remark that this pin only exists till the end of 2002.

V10 TDI actuator is 6NW008412, if you browse for pictures of it opened up you will see that it has the ceramic board with the big controller known from the MB ones with pwm. I would bet that those are able to do CAN and PWM.

If so, I will hook one up to another EDC16U1 ECU which has the old style N75 PWM with vacuum actuator as you know. I bet that the pin from the V10 TDI ECU wouldn´t be occupied in a normal 4-cyl-EDC16 ECU which hasn´t to mean that the pin doesn´t do anything. If I get my hands on such a 008412 I will connect it on the table and do an actuator test via VCDS and see what happens.
 

Rub87

Top Post Dawg
Joined
Dec 10, 2006
Location
Belgium
TDI
Ibiza '99 90HP
I think you need to put a resistor instead of the N75, otherwise the ecu will detect open loop and don't activate the output

I tried the vdo KKK actuator from early v6 audi and it works fine using just PWM, feedback is also pwm at same frequency
 

Deus Violentia

Well-known member
Joined
Oct 12, 2009
Location
Germany
TDI
2.0TDI Polo 9N
Well I´ve got the full set of actuators (N75 etc.) here and can connect them. But thanks for the hint! Could you show me a picture of the actuator you mean?
 

Rub87

Top Post Dawg
Joined
Dec 10, 2006
Location
Belgium
TDI
Ibiza '99 90HP
Dont seem to have one, here is a pic of the signal

its found on all kkk bv50 on early 2.7 and 3.0 audi tdi, its says siemems vdo and has pwm in and pwm feedback
 

Deus Violentia

Well-known member
Joined
Oct 12, 2009
Location
Germany
TDI
2.0TDI Polo 9N
ohhh I remember! It has a aluminium housing with VDO written on the front right? Had one years ago! Just got my hands on two V10 TDI chargers with actuators, will test with them when they arrive here.
 

Rub87

Top Post Dawg
Joined
Dec 10, 2006
Location
Belgium
TDI
Ibiza '99 90HP
yep thats the one, it looks a bit different to the garrett ones, this one you can move by hand and goes by itself to the vane open position
 

jurgis_l

Member
Joined
Nov 20, 2009
Location
Latvia
TDI
MK2 TDI AFN
subscribed.
As my GT2260 is now working as I expected, the new GTB turbo is on the way.

Really like the idea about running cable from actuator to turbo and place actuator somewhere in better and wider place.

I have VW Golf 2 with four wheel drive (syncro) and there is very tight place between engine, big syncro gearbox, firewall and intake manifold. Its hard to fit turbo with vacuum vnt actuator there.
 

JFettig

Vendor
Joined
Aug 18, 2010
Location
Blaine, MN
TDI
B5 Passat, 2010 Jetta
I have an actuator off a t-reg 2056VK I can donate to someone for testing. This setup would be ideal for adapting GTBxxxxVK turbos to CR TDI's.
 
Top