Date:         Mon, 4 Jan 2010 04:49:02 -0500
Reply-To:     frankgrun@AOL.COM
Sender:       Vanagon Mailing List <vanagon@gerry.vanagon.com>
From:         Frank Grunthaner <frankgrun@AOL.COM>
Subject:      Re: Conversion? JETTA ENGINE?? Now Some TDi Comments (Shortened
              Version)
Comments: To: TDI-conversion@yahoogroups.com, BUZZ@NORULZCYCLES.com
Content-Type: text/plain; charset="utf-8"

 (My first version of this was rejected by Gerry because of length. Maybe this will pass!)

Scott,

You have posed a number of questions that hopefully will be of ageneral interest for those contemplating a TD or TDi conversion. I haveno ability to speak or type sagaciously, so this will probably turn outto be lengthy. I'm also carrying the post quotes so I can get throughsome of this without losing the thread of your questions.

First, let me clearly announce my prejudices here so that the offendedneed no read further. To begin with, I see no reason to embark on aserious conversion of gas to diesel if one intends to mount a turbodiesel, whether JX, or Eco-diesel or AAZ. Yes, I know many havepersevered and succeeded, and many are happy with their efforts.Nonetheless, the JX is still woefully underpowered and the gains of theEco-diesel are severely limited compared to the magnitude of the costand level of effort. The AAZ is a reasonably robust engine of themechanically controlled variety and if superstition or prejudice driveone to that conclusion, I hope you enjoy the result. If cost orbird-in-the-hand drives one to this decision, that is fullyunderstandable to me. I only caution you that the magnitude of theeffort may not return a level of satisfaction appropriate to the cost(your time as well as your dollars) after the cold light of dayilluminates your thoughts after the testosterone-induced flush ofsuccess wears off. If you do take on a TD-based conversion or if youchoose to add a turbo to a NA diesel (not the best idea), I wouldsuggest that you add an intercooler to the system. (see previousconversion comment below). If you need to wake up a NA diesel with aturbo I would suggest that you still target the low boost range of 5psi or under. 

Also, on my list of prejudices, I cannot understand the m-TDi approach.I have no fear of electronic control systems and modified wiringharnesses. I really don't see why anyone who can read a map cannot takeon the problem. Of course you can screw it up, but that applies to ahundred other steps along the way to completion. Removing theelectronic controls of the TDi design is like leaving a free gift ofApple stock just because you don't understand the basics of stockmarket trading. Yes, I know that Karl does em, and many of hiscustomers swear by his efforts. But he is one man and the m-TDiapproach leaves you locked to that source . (Yes, I know that there areseveral other individuals modifying these injection pumps, but thescale is still small and unique). But I say, sh*t happens to the bestof us and the number of possible numbers of mechanics and shopsavailable to delouse a previously functioning TDi system are large andgrowing even in the USofA. Besides, the TDi engine in its manymanifestations (TDi, PD-TDi, CR-TDi, etc.) is a revolutionary thing ofbeauty and a significant part of that beauty is the level of processcontrol that can be achieved through measurement and computer control.With the m-TDi approach, you also give up the possibility of chiptuning (as in Rocketchip) to the specific application. But enough, Ioffer my apologies to the mechanical TDi group. It could be I'm justblind to the issues. After all, I was never able to successfully writethe Hamiltonian for the solid-state NMR spectrum of cytochrome-C, soclearly applied Quantum Mechanics is not my forte.

With that preamble, lets address your first question - is the DK casestronger than the DZ or, more exactly, how does one solve the problemof a Vanagon transmission coupled to a TDi engine with demonstratedlong term reliability. The problem here is torque. The DZ transmissionis a variant of the original Beetle transmission design as is the DK,the DH, and a whole host of transmissions including the WBX versionsand those 5 speeds crafted for the Euro market. that original Beetleengine put out 68 N.m of torque or 50 ft#'s. The diesel vanagon engineput out 76 ft#, the 1.6l A/C Luftboxer put out 74 ft#, the 2.0L A/Cengine put out 103 ft#, the 1.9L WBX pumped out 103 ft# and thestrongest 2.1l WBX motor generated 128 ft#. The general folklore on VWVanagon or T3 transmissions is that they were designed to handle 128foot pounds of torque with a 50 to 100% overbuild margin.Unfortunately, the SA 2.6L engine put out 200 N.m of torque (147 ft#)and the transmission proceeded to break both gears and internalbearings. The SA engineers added oil deflection plates and the breakagestopped. Then, our European colleagues started installing TDi enginesand breakage started again. In the same timeframe, the Vanagon or T3transmission had become a favorite for the sandrail crowd and partswere breaking everywhere. 

The TDi based failures emphasized gear tooth shear loss and bearingdestruction particurly for the pinion shaft. Stronger aftermarket gearswere developed. Stress relief of VW OEM gears was attempted throughbead blasting. Consistant with drag and road-racing engineeringexperience, it was found that transmission internal oil temperaturesabove 120C lead to needle bearing loss and gear face pitting. Thesandrail hobbyists saw differential cage fracture in the two spydergear hub. They strengthened the differential hub with the addition of 4spyder gears. Engineering analysis of the loss or failure of gear teethhas always come down (for a particular ring/pinion design) to thenumber of teeth per unit diameter. Fine teeth (high tooth counts) shearin high torque loads more readily than more coarse designs. There arealso contributions from face contact area (depth of tooth cut and faceprofile) but to zeroth order, one can think of this as the girth of thegear. More girth, more strength. Most of the VW gearboxes we candiscuss have 7 tooth pinion shaft gears. Therefore the coarsest gearsets have the fewest teeth on the ring and the finest have the mostteeth. The final drive ratio is the division of number of ring teethdivided by the number of pinion teeth. The DK transmission has a finaldrive ration of 4.57 from a ring/pinion tooth set of 32/7. The DZtransmission has a final drive of 5.86 or a tooth ratio of 41/7. The DYtransmission is the same as the DZ. The DM has a final drive of 5.43for a tooth ratio of 38/7. The DM transmission was fitted to thenon-Westy 4 speed diesels and the DZ and DY transmissions were used inthe Diesel Westfalia versions. The DU transmission has a final driveratio of 4.86 from a 34/7 tooth ratio. I believe the 1.9L and 2.1L WBSunits had a 4.83 final drive ratio and this corresponds to a 29/6 toothratio. To conclude then, the strongest ring/pinion gear sets in the 4speed transmissions are the WBX 29/6 units, followed by the DK at 32/7and then the DU at 34/7. The weakest 4 speed transmissions with the 7tooth pinion are the DZ units with the 41/7 tooth ratio followed by theDM. I know of no strength difference in transmission cases between theDK, DZ, DM, DU or ACW units. In fact, I know of no torque induced casefailure for these units. Torque based case failure was seen on theolder Type 2 transmissions and girdles were developed to overcome theproblem. Not with these units. 

So what is the possible solution. TDi's should be run at lower rpmsthan the 1.8l, 1.9l and 2.1l gasser engines. The addition of the 0.70 ratio 4th gear is a solution. But coupled with the 5.86 final drive ofthe DZ, this gives a transmission top gear final ratio of 4.10 vs thefinal drive ratio of 3.88 achieved with the DK and the stock 0.85 4thgear. When coupled with large tires, you can get a 788 revs/mile outputfor acceptable revs at 75 mph of 4038 revs for the DZ and 0.70 combovs. 3821 revs at 75 mph for the DK case. These numbers are a bit high.For transmission longevity, I indicated that I built the DK I;m usingwith 4 spyder gears and glass peened OEM VW gearsets. I'm also adding atransmission fluid cooling recirculator with a 12V gearpump and an oilcooler. For diagnostics I'm adding a transmission oil temperaturemonitor to make sure the fluid stays below 120C. Another good butpricey choice is to regear the ACW WBXer trans for lower reves permile. I don't yet know if either of these solutions is adequate for thepower output of the AFN level of tuned TDi engine. 

Others have tried to solve this problem by regearing the ACW trans,adding the SA baffles in a rebuild and using hardened aftermarketgears. This still left several possible points of failure. I have beenexploring another solution but won't move on it until the DK breaks.Several individuals with deep pockets have turned to the Porsche G50transmission. Still others have looked at the Porsche Tiptronictransmission or the revamped version of the Audi automatic withcounter-rotating planetaries in the output shafts. Pricey, pricey.There are Mendeola Transaxles specifically built to handle the torquelevels needed. Their longest gearset has a 4.11 final drive and the0.70 4th gear for an overall of 2.877:1. This at a starting cost of$8K. While considering a solution I have also lusted after aposi-traction or limited slip differential. There is a possibleAmerican-made solution in the form of the Corvair Saginaw 4 speedtransaxle, It has the strength to handle 300 ft# torque levels and theoverall final drive ratio is 3.55 or 3.24:1, both available withPositraction. First gear is rather high for Vanagon applications butthe rest of the gearing is spaced nicely. Unfortunately the Corvairengine rotates in the opposite direction to the VW engines. There arereverse rotation modifications to the transaxle developed by theV8-in-sandrail crowd, but I don't know how costly. Thereare gear shift solutions and very sturdy adapters to the CV joints. Onehas a choice of VW, Porsche or Audi CV packages. Best, the 3.55:1 finaldrive is good for 3500 revs at 75 mph with the parameters used above.

Ok, enough, lets switch to your second question: Flywheels, clutches and TDi upgrade tuning?

You indicate that you are using the Jetta flywheel, pressure plate andclutch. There are several issues with this.  The shape of the Passatflywheel on the engine side is such that you will not be able to usethe dust/water splash shield and will need to either eliminate ormodify it. The TDC markings on the flywheel (1Z or AHU) will not becorrect for the alignment and observation hole in the bell housing. Youwill have to remeasure and remark. There is a clearance issue betweenthe pressure plate and the bell-housing interior. Finally, if you doany power mods, the clutch/PP will be inadequate. In the Vanagonapplication, the clutch face will see considerably more use than it wasdesigned for in the Jetta. I took the approach of opening the DVflywheel for the 228 mm. clutch and procured clutch and PP materialappropriate for the TDi load. This gave me the proper TDC markings, nosplash plate interference and no internal clearance issues. This is alldetailed on Alistair's website:

http://www.members.shaw.ca/McTavishPac/clutch/clutch.html

Reviewing the data on the website, I see that I didn't emphasize theimportance of surfacing the flywheel (both flattening and adding theturned friction surface). I also annealed the flywheel after surfacingand then had the combined unit rotationally balanced.

Third question: Efficiency improvements vs. fuel economy?

Whenever the flow efficiency (its just a hot air pump) and the thermalefficiency of the engine is increased the BSFC or fuel consumption isenhanced. Often not by much but modest deltas can be integrated. Thiseffect is often masked by the connection of the operator's reptilianbrain to the DYW foot control. The beauty of the diesel engine is thatin the case of excess air supplied to the combustion chamber, you cancarefully inject (or use) only the amount of fuel needed to generatethe power requested. Larger nozzles just let you add a greater level offuel than that permitted by the stock system. There is some suggestionthat the newer 520 injectors have a higher quality of nozzle machiningthan was the case in the original OEM units. This could lead to moreuniform and smaller diesel fuel droplets and therefore more efficientburn. I know of no direct proof of this at this time. If you changeyour nozzles it is critically important that they be flow balanced foruniform delivery. The PD150 intake manifold is the highest flow OEMmanifold available for the TDi engine that I am aware of. The extendedport branching changes charge flow resonances and interference waves inthe intake tract as well. The smooth curve radii from the plenum helpkeep charge velocities high as compared to the short runner/commonplenum arrangement of the stock manifold. The VW technical designpapers show extensive flow modeling calculations all aimed at givingthe highest level of filling in the chamber with the largest uniformityfrom cylinder to cylinder. This was optimized while keeping the overallintake tract volume after the intercooler as small as practical inorder to minimize turbo lag. Actual torque and power gains wererealized but the major effect was seen in the mid to high rpm range. Insubstitutions, the manifold change has been seen to lift the torqueoutput above 4000 rpm and to extend the operating range out to 4800before strong drop off. I have been particularly interested inextending the the useful rpm range of the TDi should the need for swiftescape arise. The technical papers do not attempt to attribute aspecific efficiency gain to the intake manifold in the PD enginedevelopment so there is no easy way to assess the impact on fueleconomy except by inference. Careful measurements of EGT vs. outputtorque should show a reassuring picture. 

Fourth question: VNT turbo vs. the KKK or Garrett wastegate turbo?

Not a fair comparison. The wastegate turbo is a brute force device.Hold the pressure as long as you can then blow the boilers. Allturbocharger designs are compromises in a very difficult parameterspace. For real world effectiveness with these small engines, a turbomust generate boost at low rpm levels where little heat is generated.Sized to do this, it quickly gets subjected to more heat and gas flowthan it can handle as the rpm runs up by integer factors. There arefour basic ways to handle this. 1) Design the turbo for the high rpmregime and give up low end boost - thereby listening to users bitchabout turbo lag. 2) Go to exotic material to support 400,000 rpmturbine shaft speeds and high temperatures (the users won't pay themoney and the Chinese knock-off will self destruct). 3) Use the wastegate approach, dump the excess gas at a given boost level and take anefficiency hit). or 4) Design the turbocharger with a variable loadturbine end to increase its range of high efficiency operation. The VNTturbo varies the blade attack angle for feeding the exhaust gas to theturbine blades. This variability lets the system operate at maximumefficiency from low to high rpm's at the price of complexity. Thesubstantial efficiency change of the AFN and ALH engines as compared tothe AHU is largely attributed to this turbo change. Beautifullydocumented by VW in the MTZ journal. I have the copies and have beenmeaning to post the english translations to Alistair's site, buthaven't quite finished with them yet. These papers document theimpressive design of the TDi engine and also the improvements added asthey went from the 1Z/AHU to the ALH and then to the PD systems. Tosummarize, the VNT or variable vane turbo is a superb advance inturbochargers and its addition to the 1Z engine makes possible theperformance of the ALH engines without the Vanagon fitment problems. Orstated differently, for a given torque output at a specific rpm, theEGT is significantly lower. Now the simplist way to use thisturbocharger is to latch onto a combo exhaust manifold/turbo from theALH engine group. To mount it in the Vanagon at the 55˚ angle, theturbo must be clocked to move the oil inlet and outlet lines to a morevertical position. Vertical +/- 15˚. This takes careful machining or avisit to a few turbo machine shops. The details are discussed on theTDiClub forums as well as others and are beyond the scope of thesecomments. To use the VNT turbo, the software maps used by your ECU needto be properly informed. Jeff at RocketChip routinely makes themodifications to turn your US spec ECU into a AFN clone or better forcontrol of the N75 valve. 

Fifth Question: Water injection for control of EGT's?

This has been recently on my mind and I have expressed the details atlength in a recent thread over on the TDiClub forums. You will findmore information than you want there. See:

http://forums.tdiclub.com/showthread.php?t=257484

Sixth question: Better nozzles than the 520's?

Not for my money with our weak-link transmissions. You need to reviewsome of the 20,000+ postings on the TDiClub site on the subject to geta flavor of the issues and suggestions.

Seventh question: Why Rocketchip?

There are many other tuners, but I've found him to be intelligent,honest and slow. He was willing to look at the real torque profileneeds and help me manipulate both torque and fueling within the pulselimits imposed by the application to our vehicles. So, sure there areothers, but his good, experienced and I trust him. YMMV!

Good luck and good night,

Frank Grunthaner


 

-----Original Message-----
From: Scott Greiner <greines@yahoo.com>
To: frankgrun@aol.com; BUZZ@NORULZCYCLES.com
Cc: TDI-conversion@yahoogroups.com
Sent: Sat, Jan 2, 2010 1:46 pm
Subject: [TDI-conversion] Re: Conversion? JETTA ENGINE??

      
                  
Frank Grun,
Thanks for the write up. If you don't mind, I would like to askyou some questions. I am looking to continue to improve my conversion.You mention that you are "using a DK with largertires and the rev range is acceptable. The 1Z with a PD150 intake,RocketChip tuning, 520 injectors and a clocked VNT turbo in a nicepackage and I'm completing a number of mods and updates including awater to air intercooler and water injection system for EGT control.Unfortunately, the TDi engine is pricey and the additional componentengineering requirements add more cost."
 
I have a 1997 Jetta AHU 1.9 TDI & a  DZ transaxles that wasre-built and re-geared, question is; is the DK case stronger that theDZ? My DZ is the 3 rib case, mine not that familiar with the DK, but Ithink I has 5 ribs. Since the DZ is from the stock 1982 Diesel Vanagon,wouldn't be stronger than the DK due to pushing the higher torque? Myre-gearing setup is stock 1st & 2nd, 3rd is 1.14 4th is .70 with a4.57 installed for the R&P. I'm running stock 14" wheels with 195tures. Planning on going 16" and adding stronger taller spring andbetter shocks.
 
I'm am using a 228mm clutch plate setup the Jetta's flywheel andspring. I was told that this clutch would be fine for upgrading the hpto 130, not sure what the torque would be about. Since the USA AHU 1.9TDI stock is 90 HP, I would like to improve that to the UK 110 hp or goup to the 130 HP. I have heard that improving the hp the fuel milagewill improve higher. What is your take on this and what is yourcomments about of improvements compairing why one would use the PD 150intake compaired to the stock 1997 Jetta's intake that came with themotor?
 
What is the comparison and/or different between the stock Jetta's Turbo and the Clocked VNT Turbo?
 
I am also interested in upgrading from the stock Jetta intercoolertha I am using to a water to air intercooler, I have not heard aboutthis water injection system for the EGT your mentioned, can youinlighted me about it and what improvements one would see?
 
I plan on using the 520 injectors, are there any other options other than the 520's?
 
And last what are some differents one can explain about  aRocketChip Tuning,. What other types of tuning are there and how isthis done? Why the rocket Chip?
 
Thanks in advance,
 
SOG
Mesa, AZ.
1982 Diesel Westy with a 1997 AHU 1.9 TDI
Just finishing up running my first fuel tank of Bio-Diesel and loving it.