Not a single good MPS in the whole lot...

[ubercrap]

Hey Uber, I sent you a PM with my address but I never received your PM?
Ray thanks for another great info post! It is hard to remember that the Djet was the first Injection system and only has 23 wires! Think of how much easier we should have it compared to new systems with a ton more sensors for drivability!! It is true with just a few hundred dollars you can set yourself up with a lot of test equipment to test a entire injection system and components! I have seen a CO tester in JC Witney for cheap, I think around 150$ or under! A multimeter can be picked up from Radio Shack or Sears for around 20 bucks and a fuel pressure vacuum setup in the same range. I think my combo that is a vacuum pump and gauge setup cost me almost 30$ and you can bleed brakes with it and test vacuum and pressure! I think a timing light is cheap too! I have had mine for years! I know though for a startup cost it can add up but you then have the equipment!
It is true that once you have these systems setup and checked they provide performance and good reliability! It's just a transition from a bug type setup that can be restored outta a catelog! I dont know how they check out with emmission test though! I hope Jen tries to keep her car till she can do some adjustments and more checks! I really think she is so close to success!! How much is it to bring a car in from Canada to here ? Sorry it wasn't easier for Jen and finding a mechanic that could had her going straight from the start! Isnt the Djet a common system also on Mercedes? I just never buy the oh its a 411/412 excuse!! If you are a mechanic you can find vacuum leaks/pressure check/ adj. and check componants! Take care everyone! Bill

I don't think it is that difficult or costly to bring over a car from Canada.

[vwbill]

Hey all, filed the head off the rivots of the MPS case with my tile carbide hand file then tapped the rivots out with a torque head screw driver. I found out why most of the MPS's have a vacuum leak! The copper diapham has a paper thin edge that holds it in place and the center seems thicker so I guess it is a miracle that any of the older MPS's still hole vacuum! I would believe humidity would help in forming corrosion along the thin edge and with the movement of the diaphram cracks the edge along the mounting plate circle. The one I opened is cracked 2/3 the way around the inner edge of the diapham. Imagine a accelerator pump diapham in a carb and how the area between the body and inner diapham rips. Now the question is can you repair via solder or use thin copper or thin rubber fabric glued to one side? I guess I still need to determine the total movement of the inner diapham to see if the movement would be too great for a rubber fabric patch and would it be too thick?? Would it only be good for a backup MPS? Too bad they don't make a replacement diapham! I could see a thin solder and fabric combo maybe working???
Anyone try this before? Have to visit Paul Ander's great site!
http://members.rennlist.com/pbanders/DJetParts.htm bill
p.s. Well Paul's site doesn't suggest trying to repair the diapham!!

[raygreenwood]

Actually the diaphrams are not thicker in the center. They are made from a single copper sheet. The range of thicknesses of the copper sheet has variations from time to time. The range I have found after examining about 30 MPS's over the decades....is between .0038" and .0052". There is no set variation that corresponds to any part # of MPS. The range of thickness has been found in everything from type 3 part numbers to the 039's from 914's.
This thickness variation only affects a couple of things. With a thinner diaphram, there will be a slight sensitivity variation to external atmospheric pressure.....but so slight you can only pick it up with a very fine vacuum gauge. It would make for only a very very slight difference in adjustment of the center screw on any given MPS.
Depending upon the direction of your thought proceses....a thinner diaphram may also either crack sooner.....due to being thinner....or crack less soon due to more ease of flexibility.
In the long run, it would also come down to the individual ductility range of a given grade of copper.
The grade of copper used is something I have not had the capacity to examine. Its very possible that the thickness variation exixts....because differnt grades of copper were used through teh years....and for no other reason.

The center only "feels"thicker....because of the ribs that are embossed into it. From the sections I have taken of broken diaphrams, the thickness is the same from center to edge. These are what give the plate...spring. There are very fine variations at the apex of the curves in the embossings. These are due to compaction of material on the die. Its common in any embossing operation. There are also manufacturing secrets that I just cannot possibly know. I can only speculate. It is possible....that the plate was heated during pressing to alter spring tension in the center....or even to make that spring tension take a particular set so it last longer at a more uniform rate.

There were also several rib patterns of subtle difference. Again, these do not appear to correspond to any particular part number of MPS....or thickness of raw material for that matter. They just used a differnt die.

In the few that I have manufactured, I settled on a plating thickness in the middle of the range that I listed above. I settled on .0045" thicknesss. I have to start with raw copper plate from rolled sheet....with a nominal thickness of between .0048 to .0052". I then attached that plate to a steel machine flat and hand lap it to between .0045 and .0048".
I have made numerous dies. The funny thing is, none have been machined. The level of machine work that needs to be done on a stamping tool like this....and it is a two sided tool...both sides different but perfectly meshed.....can only be done by a tool and die maker. The plant I work in....has its own tool and die shop with this capability. I have been working out a deal with the tool and die manager to see if he can make what I have in steel....when and if we ever get spare time.

Because of the thinness of the copper plate, its easy to emboss. The first tool I made...which only produced two uniform plates before it died....was of plaster and polyester backing. It proved they could work.
The second die was made of all cast polyester with steel plate backing. It worked better, but only in the first week of its life. Polyester tends to keep shrinking when it cures. It distorted.
That brings me to another manufacturing observance. Some diaphrams had 3 to 5 radial lines in them that crossed perpendicular to the eembossedrings of the diaphram. That is very similar to the technique used to make the barometric chambers. These are reliefs that keep teh plate from distorting when teh die is not perfect. They keep the plate flat. I use those now.
My third die....was made ....using JB weld . I had to make a small chamber from polyester. Then I put a block of wood inside with a 17mm hole cut in it so the plate with its center part would sit flat on the block of wood. I glued it down with hot glue, then sprayed the upper surface with micro powdered urtehnae mold release. Then put a vinyl ring around the perimeter of the diaphram as a casting seal. Then poured in the slightly warmed JB weld, covered the chamber...and vacuumed it down to get air out of the 1/4" of JB weld. Then, release vacuum....and take cover off.....and lay a disc of 1/4" steel plate on top of the JB weld, vacuum it to get the bubble out from under it....and let it cure for three days with vacuum off. I had to repeat this process for teh other side.
There are steel dowl pins in theour holes in thetock diaphram for register of thewo sides.

With the right lubricant, this mold will make quite a few plates before it dies.

The next problem...was how to get accurate center and small holes in the plate to keep from distoring it. I have a lot of die cutting experience. I have no found punches yet that I like the edge affect of and do not distort the plate. Instead, I made cutting burrs to grind the holes in the opper by dipping epoxy covered mandrels in silicone carbide dust and dressing them on a stone. These make sweet holes. This was my last installment in August.

I have some testing to make sure they work close enough to teh same as originals...still to do. Right now, the variation is less than 5%...from stock Bosch. I will be back on this before the month is out.

For now...yes...you colud dolder them, but the heat distortion kills the plate...and makes very a very erreatic adjustment. Silcone...very strong silicone like permatex copper works best. Ray

[wildthings]

I haven't had a MPS apart in a while, since I eliminated mine by switching to L-Jet, however is it necessary that they be made of metal? What about using plastic or the plastic covered cloth like they use in many other diaphrams. There must be some diaphram out there, metal, plastic or otherwise, that is readily available that could be adapted to the D-jet MPS.

[raygreenwood]

Ahhh...now you are thinking along the same lines. Parts I have modled are like this.....flat springsteel with cast on 600 degree silicone (very difficult to make but would last forever). The calibration is different.
The problem with a ribbed plastic diaphram...which it needs to be because it has to have its own self re-centering springinesss (there is no room for a spring on each side of the diaphram to re-center...so it must be a spring diaphram). The plastic must have the ability to not creep in shape...with temperatures up to 300f. Thats a tough one. You would need to us e a teflon based plastic....or a high temp polypropylene. It would have to be injection molded.
I have also been experimenting with one using a polyester "scrim" or internal mesh....that is under tension first them cast into the mold with high temp silicone. That one has the most promise.

But for now...the best thing to do is just make duplicate copper ones and new o-ring seals...and make them cheap. Ray

[vwbill]

Great info Guys!!! what about a dashpot diapham? Those look close to the same size? Also, you will laugh but I was thinking about a heat resistant material that is flat and was looking in the kitchen and found my SilPat made in France flexi cookie sheet!! They use them at SubWay for the bread molds so the bread doesnt stick. They are kinda tan/clear and have a grid reinforce pattern? It is alot thicker!
I really like Ray's molding ideas too! Have you seen those baking tin replacement German flexi silcone color things? That stuff is cool! So I can see using a high temp copper silcone now for a fix or maybe molding.
I have heard of dental molding material you can buy somewhere here and was wondering if you could make a two peice mold and make a silcone replacement?? I guess the cost of molding or die would be ok if you wanted to fix several MPS's but I wonder if the rebuilders in Cali would sell the diaphams they probably have made? I guess the cost they charge of 146$ I think is well worth it since they rebuild the MPS and calibrate the unit closes to your setup! I still like trying though myself and to see if I can make something to work like that Silpat or Dashpot.
My next question is does the center washer/ adj screw assembly could apart or is it pressed on the diapham? Bill
p.s.Question about the calibration, on Paul's site he uses a wavetec to calibrate the MPS; could you do it by graphing it and using a vacuum gauge and test the resistance?

[raygreenwood]

If you already have a die...the cost should be negligable. The stuff they use on thsoe cookie sheets is either teflon or a teflon based laminate. Its also not very springy.
Also, you have to remember, that whatever is used, must be at least similar in spring to what is already in teh MPS...or the calibration rate will change too much...and you can't adjust it.

The cost to make a steel tooling die....a real one...properly done.....and mirror polished of the proper steel and hardened....would be about $1200 at least.
Fortunately....it does not have to be so perfect. I will not be making millions of these. If I got a few thousand out before the die crapped.....then I think that could suffice. D-jet....though I love itso and will probably always use it on my cars that came with it....will eventually disapear except for enthusiasts and restorers. Its just too hard for most people to work on (these days)....to be worth their effort and attention span. If you have not already got one...sooner or later....you will ALL need a new harness to continue using D-jet. You will need a few spares of teh TPS and all sensors and a way to rebuild the triggers and MPS...to keep a daily driver for your life times. Thats a lot to think about. Ray

[MGVWfan]

If I can keep the Nomad on the road and finish the Duster off, I can finish experimenting with my electronic MPS idea using a MoPar MAP sensor and some analog electronic bits to simulate the CVT and its responses to vacuum changes. So far, it looks like I'll need 6-9 months of fiddling to get there after I get the 412 back on the road, but it appears possible to replace the MPS with relatively modern parts. D-Jet may be old, but once you eliminate the MPS, there's really nothing else that can't be fixed somehow.

In the meantime, Ray's replacement full-load diaphgrams and replacement O-rings will fix every bad MPS I have on the shelf right now (3, counting the one from a Volvo 164), so I too await Ray's dies getting made. Don't throw those bad MPS's away yet!

[raygreenwood]

Really...yes...you should be able to replace the MPS with a modern MAP. But also bear in mind, that the MPS...at least the late model diaphram types, are incorporating two seperate MAP functions. The early non-diaphram MPS system used a seperate full load diaphram that was not in the MPS. It was under the runners on type 3.
In the long run, those early D-jet types...ahd little or nothing to fail in the MpS. No diaphram. The seperate full load diaphram...was rubber and metal based and lasted a great long time until the rubber rotted.

You would need a wide range small encrement map to replace the the main MPS function. By wide range...I mean....not so much in the range of the MAP response...but in the sensitivity to vacuum. Then you would need another MAP (maybe two).....running from the same vacuum source to simulate the full load diaphram. The possible third may need to be a very sensitive small range one to give differential signature from ambient.

One other part I am working on that is actually further along than the diaphrams ...technically. I am printing a replacement circuit board for the TPS.

This one will be actually twice the size in all dimensions from the original. What it allows.....is a wider diameter arc of travel for the wiper arm.... . Think about that. What it allows, is the wiper to actually travel a longer perimeter....within the same number of degrees with slightly larger and better connecting contact strips (more sure contact)....and most importantly.....twice as many contacts. This would give twice as many injections with each swipe of the wiper arm. Sort of a high performance TPS . Ray

[vwbill]

Great ideas for the TPS Ray! So Wildthing you converted a 1.7L to L-jet?
Did you use the same injectors(yellow)?
Hey I measured the armature movement in the MPS and got close to the same as Paul's data from his great site. I got 4mm. He got 4-4.5mm. Did you see his chart for the change in inductance and hg? So when he calls the MPS a LVT does that mean it has a straight line change? Also why does he use inductance and not resistance or is that because it is a measurement of the two coils? It sure does seem like the MPS is an amazing sensor! Can't believe that 4mm of movement and the adjustment of that range can be used for different motor setups! Some great engineering! It seems like the D-jet would have better drivability once setup in different driving conditions? Funny how with only 4mm of total travel the diapham fails but I guess it could be moving in and out fast with variations in conditions? I dont know much about MAP but I had seen that they showed different bars types? I need to read Paul's site a few dozen more tiimes! bill

[raygreenwood]

Depending on what you are willing to do to that poor sensor...a lot can be done. I have actually measured as little as 3.5mm and as high as 4.8mm for stroke. The real key is how it is adjusted on both the full load diaphram stroke...meaning how far out the back plug is turned....and what mainspring was installed in the MPS.

An MPS I am playing with...has a modified armature ...or rod. The rod can only protrude so far in or out of the coils before it is out of range of the functional response of the coils. Hence (among other reasons)....the short stroke.

Looking at the armature...you will notice that there is no reason for it to be solid. It is teh steel outside that the coil is interacting with. I added an adjustable main spring to give slightly more load against the armature for when WOT or part throttle occours...and atmospheric rushes in to collapse the aneroid chambers.
This has the affect of allowing the armature to linger longer in that outward position....because it will require more than just partial vacuum to push the armature back...because of the higher mainspring tension.

The drawback this has....is that at idle...the armature is pushed out about .5mm- .6mm....from where it would normally been when properly adjusted at idle. That causes too rich of an idle.....and too much of a stroke when at fuill WOT.

So.....I "cored" the armature....shortened the rod tip ..and threaded the rod...so I can move the "can" section of the armature back and forth for adjustment. There is already an adjuster cone at the rear of the armature against the spring. After measure the distance to move...I can add or subtract at the back.

It will be delicate to tune...but should give good running. Ray

[wildthings]

The conversion to an L-jet is pretty easy. If you have a 4dr just find a complete L-jet off of a bus and bolt it on. If you have a wagon there won't be room for the bus air cleaner and you would be better off in getting at least the air cleaner and bellows off of a '74 Type 4.

You need to get every part of the L-jet. Fuel pump, relays, all the wires, hoses, sensors, throttle body, plenum, injectors, AFM, distributor, and the brain. Depending on the year of the L-jet you also need the breather box and if so equiped the O2 sensor.

It is pretty much just a bolt on, except for figuring out the air cleaner mounting. The air cleaner off of a "74 T4 would simplify this even on the 4dr. You will also have to figure out a new mount for the L-jet fuel pump and filter, and depending on the harness you end up with you will need to figure out where to mount the brain. My brain is presently in the engine compartment, but it is destined to end up behind the back seat in my 4dr.

Ray says the D-jet works better and gives a little more power, but I have found the parts availability for the L-jet to be way better, at least where I live.

[raygreenwood]

Yup...thats pretty much true everywhere from what I have found. But if you opt to go D-jet....as everyone ese mentioned.....get a couple of spares of everything (MPS, CHT, ambient air, TPS and trigger points). All else ...injectors, pumps, cold start etc....are simple to find. Above all....if you want to run a long long time and keep D-jet around......get or build a new harness. It will run for ages. Ray

[vwbill]

Hey, do the full load type MPS's only have the slits in the back case? I have seen some T3 ones that don't have any slits? So if the coils are basically the same is the only variation the full load diapham? So can you setup any MPS for a t4 as long as it has the full load diapham? I have found so many MPS's with failure from the diapham it seems! bill
I found a place that has the wavetec inductance meter for 62$?
http://www.testextra.com/protek_cl200.htm

[raygreenwood]

The MPS without the slot in the back....is teh non-diaphram type. It used a seperate diaphram switch to signal full load. That diaphram switch is just as elusive to get. It was early type 3. Late type 3 essentially used teh same set up as the type 4.
yes, there are differences in spring calibration on teh non-diaphram MPS....but they could be adjusted around. So...if you had an early type 3 system with a good seperate diaphram switch, it could be used and adjusted on a type 4. You will need the ECU as well. Ray