TheSamba.com :: View topic

Thu Sep 09, 2010 12:37 pm

I'll have to search stfu and find that trigger wheel to see how he did it. the problem is if do find a 411 air duct its made of plastic. I'm sure i could figure something out but it would be a shame to hunt for a rare part only to hack it up to mount a bracket.

I just searched and found some info on the stock type 4 inj. bosch pn is 0 280 150 112 vw part # is 023-906-031 and they are 185cc used in all busses, type 4's, vanagons, and 1.7/1.8L 914's as well as 77-79 bugs. I still haven't found any latency/dead time/dwell times for those injectors. I have a set of modern injectors just laying around that are 185cc as well that I can find the latency times for, they probably will run a lot smoother as well. I just have to make a custom braket to mount them and get clips for them which isn't a big deal.

Thu Sep 09, 2010 2:21 pm

supaninja wrote: I'll have to search stfu and find that trigger wheel to see how he did it. the problem is if do find a 411 air duct its made of plastic. I'm sure i could figure something out but it would be a shame to hunt for a rare part only to hack it up to mount a bracket.

I just searched and found some info on the stock type 4 inj. bosch pn is 0 280 150 112 vw part # is 023-906-031 and they are 185cc used in all busses, type 4's, vanagons, and 1.7/1.8L 914's as well as 77-79 bugs. I still haven't found any latency/dead time/dwell times for those injectors. I have a set of modern injectors just laying around that are 185cc as well that I can find the latency times for, they probably will run a lot smoother as well. I just have to make a custom braket to mount them and get clips for them which isn't a big deal.

see, that's what I'm getting at with the latency times. I don't think they exist, as numbers, for the stock set up. Why? because most of the "timing" for the injection pulses was done pseudo-mechanically. pressure? ECU assumes you have 28 psi, and doesn't give two shits about what the actual value is. Pulse was determined by the trigger contacts that were in turn opened and closed by a cam on the same shaft as the distributor. The times when they open are linked to the distributor, and by extension the cam, which also determines when the valves open and close. in that way, the timing was taken care of. Unless someone has physically sat with a timer to figure out how long that duration is at such and such RPM, I don't think you stand a chance in hell finding what those numbers are.

Honestly, I think you're drastically overcomplicating this. Just run the MS so it takes a signal off the stock trigger contacts (or if using L-jet, the magnetic points pickup doo-hickey) and whammo, injector timing is taken care of just as it would be stock. I've seen people in on STF do it and when I thought about going that route I was told by the guy over at DIYautotune that it would work, so I know it can be and has been done.

as for the mounting... even if its made of plastic, you could epoxy a mounting tab on :roll: or if you end up making it out of a stock steel t-3 shroud then the point is moot anyway....

Now the other option is to have them fire sequentially instead of in banks... or in multiple pulses of the same bank... you see what I mean when i say MS has a lot of options?

Thu Sep 09, 2010 6:46 pm

ya i know the stock FI setup is first generation fuel injection and arcaic compared to modern day fi. but latency times(or dead time or lag time) is calculated into the stock ecu and is one of the inputs required by MS. I got this from www.injector-rehab.com:

Injector "lag" time is the time it takes for an injector to open (OT) form the time it has been energized until it is fully open. There is also a close time that you do not hear about very often (CT). Some engine managment systems require you to enter this time and refer to it as injector lag, latency, delay or other terms.

If you have the wrong time in there the car will run like doo doo for sure. I don't expect to find the info on those injectors anywhere. I would have to send the inj to someplace that tests injectors like injector-rehab and they will have to put it on a test stand and hood a oscilliscope up and and determine the time it takes for the injector to shut off.

i don't have to run different injectors to get that number, i can always just try different times till it runs right, but they are 35 year old injectors. old technology and probably pretty much wore out. I have a set of modern injectors laying around, might as well make then fit and rock those, plus i can get the latency times for those other inj fairly easy.

This is all part of running a completely different management instead of the stock ECU. BTW you can chose sequencial firing. also the MS uses the negative on the coil to trigger the inj firing from what i can tell. i just looked at the input for the injectors, it has a 1.0ms default which will probably be good and it has a .2v voltage adjust, whatever that means so the megasquirt software i'm using is really dumbed down but will probably work fine.

My past build got pretty intense with all the info the management controlled/needed. if you got the latency times just a little wrong it ran like pure crap with all kinds of cold start and open loop issues. Lol, your right i am way over thinking this setup, it will work just fine as is. I should have it all hooked up this weekend and proabably next weekend i'll fire it up and start to make heads and tales out of this setup. I already know i need to figure out the resistance of the cylinder head temp sender at 85F so i can change the resistors in the MS to match that value.

Thu Sep 09, 2010 11:47 pm

supaninja wrote: ya i know the stock FI setup is first generation fuel injection and arcaic compared to modern day fi. but latency times(or dead time or lag time) is calculated into the stock ecu and is one of the inputs required by MS. I got this from www.injector-rehab.com:

Injector "lag" time is the time it takes for an injector to open (OT) form the time it has been energized until it is fully open. There is also a close time that you do not hear about very often (CT). Some engine managment systems require you to enter this time and refer to it as injector lag, latency, delay or other terms.

If you have the wrong time in there the car will run like doo doo for sure. I don't expect to find the info on those injectors anywhere. I would have to send the inj to someplace that tests injectors like injector-rehab and they will have to put it on a test stand and hood a oscilliscope up and and determine the time it takes for the injector to shut off.

i don't have to run different injectors to get that number, i can always just try different times till it runs right, but they are 35 year old injectors. old technology and probably pretty much wore out. I have a set of modern injectors laying around, might as well make then fit and rock those, plus i can get the latency times for those other inj fairly easy.

This is all part of running a completely different management instead of the stock ECU. BTW you can chose sequencial firing. also the MS uses the negative on the coil to trigger the inj firing from what i can tell. i just looked at the input for the injectors, it has a 1.0ms default which will probably be good and it has a .2v voltage adjust, whatever that means so the megasquirt software i'm using is really dumbed down but will probably work fine.

My past build got pretty intense with all the info the management controlled/needed. if you got the latency times just a little wrong it ran like pure crap with all kinds of cold start and open loop issues. Lol, your right i am way over thinking this setup, it will work just fine as is. I should have it all hooked up this weekend and proabably next weekend i'll fire it up and start to make heads and tales out of this setup. I already know i need to figure out the resistance of the cylinder head temp sender at 85F so i can change the resistors in the MS to match that value.

Ok -- I see where you're trying to 'fine tune' now. Unfortunately I don't really have any realistic solutions for you -- a time consuming but ultimately likely to be sucessful method is take an educated guess, and then pick a number suitably above or below, and then repeat the process in the opposite direction after you've done a run or two and logged it on the laptop. At that stage, you can interpolate roughly in excel and graph out a better approximation, and then just keep narrowing it down until you've got yourself a table of latency times that you can work with. but I stand by thinking that you may not need to worry about it too much, since you have a lot of adjustability in the other components. What do the injectors look like that you DO know the latency times for? if you have pictures, we might be able to help you rig up an adapter of some sort to run them on the stock L-jet plenums. Offhand, do you know if their flow rate is at least close to the stock ones?

Fri Sep 10, 2010 4:17 am

It will be tomorrow when i can grab the extra inj's and intake rail from the type 4 motor then i'll take some comparison pics. both injectors flow 185cc.

Sat Sep 11, 2010 12:18 pm

JSMskater- your right there is a wheel available for the pancake cooling on a type 4 motor. A super quick search of STF got me to this thread http://www.shoptalkforums.com/viewtopic.php?f=23&t=115786&start=45 and there is a kit that replaces the cooling fan spacer with a 36-1 trigger wheel and a brkt for the VR sensor. The kit is for the 914 crowd but they're cooling is identical to my bus motors cooling. I sent the guy an email and i'll call him monday. I just need that wheel, edis setup off ebay, and a ms2 daughterboard and i'll have a sweet ignition setup thats tunable. I'm pretty geeked out right now.

Sat Sep 11, 2010 1:01 pm

supaninja wrote: JSMskater- your right there is a wheel available for the pancake cooling on a type 4 motor. A super quick search of STF got me to this thread http://www.shoptalkforums.com/viewtopic.php?f=23&t=115786&start=45 and there is a kit that replaces the cooling fan spacer with a 36-1 trigger wheel and a brkt for the VR sensor. The kit is for the 914 crowd but they're cooling is identical to my bus motors cooling. I sent the guy an email and i'll call him monday. I just need that wheel, edis setup off ebay, and a ms2 daughterboard and i'll have a sweet ignition setup thats tunable. I'm pretty geeked out right now.

I do moon light as a darksider from time to time. :twisted:

Sat Sep 11, 2010 1:27 pm

so what would i be classified as with a type 4 and stock FI? Thats technically not a darksider, maybe a daywalker?

Sat Sep 11, 2010 4:05 pm

supaninja wrote: so what would i be classified as with a type 4 and stock FI? Thats technically not a darksider, maybe a daywalker?

If it aint stock-its dark. :wink:

Sat Sep 11, 2010 4:35 pm

notchboy wrote: supaninja wrote: so what would i be classified as with a type 4 and stock FI? Thats technically not a darksider, maybe a daywalker?

If it aint stock-its dark. :wink:

the pull of the darkside is too strong for me to refuse, master ;)

Sun Sep 12, 2010 1:18 pm

I just found the jackpot of info related to megasquirting (lololol, that sounds a bit naughty) a type 4 engine via the 914 guys. http://914megasquirt.ning.com/

After I get this rig running I'll make a very detailed list of everything I used to convert a type 3 to a type 4 with factory FI and megasquirt. I'll try to put it on the first post, might need a mod's help with that.

Mon Sep 13, 2010 7:39 am

JSMskater wrote: krusher wrote: I am going to go megasquirt and a subaru.

The megasquirt stuff adds up fast.

I think i will be in $700/$800 for it all :?

Did you get a stimulator to test your build,i may be interesting in buying it from you if your done with it.

Granted, i priced mine out for a T3 engine not a subaru, but I ended up at HALF where you are, all in, with the stimulator and a wideband A/F meter. :?

I can't really see where you're getting all these costs from, the subaru motor already comes with all the hardware you need... you just need to supply the ECU relays wiring harness and some time on the laptop...

Well for a start I will buy a ready constructed ms2 at $400 , plus relay block, wiring harness , ecu to relay block lead, stimulator etc.

Mon Sep 13, 2010 9:07 am

krusher wrote: JSMskater wrote: krusher wrote: I am going to go megasquirt and a subaru.

The megasquirt stuff adds up fast.

I think i will be in $700/$800 for it all :?

Did you get a stimulator to test your build,i may be interesting in buying it from you if your done with it.

Granted, i priced mine out for a T3 engine not a subaru, but I ended up at HALF where you are, all in, with the stimulator and a wideband A/F meter. :?

I can't really see where you're getting all these costs from, the subaru motor already comes with all the hardware you need... you just need to supply the ECU relays wiring harness and some time on the laptop...

Well for a start I will buy a ready constructed ms2 at $400 , plus relay block, wiring harness , ecu to relay block lead, stimulator etc.

THATS why. if its already built you shouldn't really need the stim... I suppose its helpful in some ways after but my understanding was its really more an assembly tool than a tuning one.

IMO alot of the wiring and putting together of it is the easiest part, which is why I wouldn't pay that much to have someone else do it. but for those without enough time or who aren't comfortable working with wires I can see how it'll get real pricey real quick. I think you save about 200 on that ecu if you make it yourself.

Mon Sep 13, 2010 3:32 pm

The unassembled version of MS2 is $250 but $400 for an assembled one would be worth it if your not comfortable with stuffing a PCB. From my research it seems most of the issues with MS are assembly related. Looks like Jake Rady is going to try MS2 on one of his rigs to get a better understanding of it http://www.914world.com/bbs2/index.php?showtopic=110664&st=20

The rest of my exhaust just showed up, I'll post some pics of what I have to modify on it.

Mon Sep 13, 2010 7:17 pm

I just paypal'd a guy for an 411 air duct, hopefully the second time is the charm.

Wed Sep 15, 2010 10:33 am

I see you want to change to silicone brake fluid. My experience with silicone fluid has been that it is accepatable in some ways and in others it has not been (for me). I dislike the fact that I do not have a rock hard pedal like I had with regular fluid. This seems a common complaint when silicone is used in old cars (maybe it's the lack of a brake booster). I love the non hygroscopic properties for longevity of brake components. I love the fact that it doesn't damage paint. It's cool that it's purple too.

Wed Sep 15, 2010 3:08 pm

I run silicone fluid in my "normal car" and i ran it in my old bike. The guy that rebuilt my master cylinder swears by the stuff, pm me if you want and i'll send you a long paper that wrote on the subject back in the 90's.

Thu Sep 16, 2010 12:16 am

JSMskater wrote: krusher wrote: JSMskater wrote: krusher wrote: I am going to go megasquirt and a subaru.

The megasquirt stuff adds up fast.

I think i will be in $700/$800 for it all :?

Did you get a stimulator to test your build,i may be interesting in buying it from you if your done with it.

Granted, i priced mine out for a T3 engine not a subaru, but I ended up at HALF where you are, all in, with the stimulator and a wideband A/F meter. :?

I can't really see where you're getting all these costs from, the subaru motor already comes with all the hardware you need... you just need to supply the ECU relays wiring harness and some time on the laptop...

Well for a start I will buy a ready constructed ms2 at $400 , plus relay block, wiring harness , ecu to relay block lead, stimulator etc.

THATS why. if its already built you shouldn't really need the stim... I suppose its helpful in some ways after but my understanding was its really more an assembly tool than a tuning one.

IMO alot of the wiring and putting together of it is the easiest part, which is why I wouldn't pay that much to have someone else do it. but for those without enough time or who aren't comfortable working with wires I can see how it'll get real pricey real quick. I think you save about 200 on that ecu if you make it yourself.

I would normally have a go at something like the construction of the MS2, but I don't want to drive down the road worrying if my first time electronic building skills are holding together.
Maybe I can skip the stimulator then, but i think it come in hand as a check tool if part of your engine is not functioning with the MS.

In my bus the vw long block was $3250 and I just got a JDM ej15 engine for $800 so there's a few $'s saved to buy MS stuff :wink:
If i spend $800 on MS stuff and $500 on an adapter i'm still over $1000 less than my vw longblock.

Thu Sep 16, 2010 5:51 am

mike sharp has a bunch of conversion parts for the scooby motors, including a shortened oil pan. http://www.sharpbuilt.com.au/sharpbuilt1/home.htm

Sat Sep 18, 2010 5:07 pm

I finally got my normal car back onto the road, now I have time for the notch again. Got good news, bad news, and some pictures (since everyone on here loves pictures).

Here's my $200 type 4 2L (I slapped the 411 air duct on so you'all can see how that works)

with it I got the entire fuel injection, harness, and alternator harness. The voltage regulator I bought for itplugged right in.

I removed the late bus exhaust, by remove I mean broke all the rusted up bolts off. I test fitted the setup I pieced together for the conversion.

good news, I don't need to shorten the header pipes. the extractor is as close as it can get to the shroud. Also notice all the room between the air duct and the muffler. I was a little worried about that all jiving.

With it mocked up I took a bunch of measurements and it should fit behind the rear apron...just barely though.

I got a new alternator for the seller for fifty bucks, and he threw in another intake runner since one was damaged. the motor didn't come with ignition, I can use the 009 thats on my type three motor or do a edis install, not too worried about that. I also measured the resistance of the chylinder head temp sensor @ 81F so I can dial in the megasquirt for the cold start function. This motor has every piece of tin, all straight too. It has all the bits for the thermastat, and I had to fix one flapper and it worked with the lighter test. I did a few measurements and I'll have to trim down the front tin about 2", I'm going to make a template from my t3 tin. shouldn't be an issue. The intake plenum is smashed up pretty good, I'm going to need a new one of those, they are $15-25 in the classifieds.

Now the bad news:
1. when I drained the oil I discovered it was full of water
2. on of the valves was being wierd(hydraulic lifter was sticking)
3. heard a clunking noise while turning it over.
4. and the number 1 reason I'm going to split the case...its got type 4 syndrome. someone over torqued the oil screen plate. The oil pick up tube and the bolt going through the middle of it moves about 1/2". I won't be surprised if the cam perch is broken.

Its not the end of the world, I was looking for a reason to get into the motor plus it was super cheap. Something had to be wrong with it. I've got a hot jake raby cam/solid lifters to stuff in it. Good used cases aren't expensive. and if the crank hasn't been machined I got a set of rod bearings, I also have a set of cam bearings and new rings too. I want to make sure the heads are good too.