TheSamba.com :: View topic - building a 1914, fuel injected daily driver. need info

Fri May 16, 2025 7:09 pm

Max Welton wrote: What would I do? Reach out to Mario at the dub shop.

https://thedubshop.com/

Max

esde wrote: I've built an MS2 unit and love it. If I wanted a turn key unit for daily driving, I'd get a Dubshop basic system with a micro squirt controller.

Before you let thesamba talk you into a blown stroker :roll: contact Mario. This is absolutely what he does for a living.

Max

Fri May 16, 2025 8:32 pm

That's a great idea.
Mario should be able to give you some realistic expectations about what this or that system can and can't do, at a professional level.
And set you up with something that fits your goals.

instead of all us goons who just tried everything for free.
And tell you the same thing anyway
You'll believe it a lot better when you are paying for the advice :lol:

Sat May 17, 2025 5:45 am

Funny thing to me is reading these comments about the difficulty in getting EFI to run properly on a stock 1600 under different altitudes and temperatures etc. and VW did it flawlessly 56 years ago. Makes me wonder about the serious lag in development in the aftermarket industry...

Sat May 17, 2025 6:18 am

Vw had electrical, mechanical, and automotive engineers, with millions of dollars to spend, and unlimited resources.

Now days it’s just someone with a summit racing catalog thinking they can slap a v8 system onto a vw and all the physics, thermodynamics, and heat transfer issues issues will just disappear.

VW also put the fuel injectors at the ends of the manifold where they should be.

Brian

Sat May 17, 2025 7:51 am

andk5591 wrote: esde wrote: I've built an MS2 unit and love it. If I wanted a turn key unit for daily driving, I'd get a Dubshop basic system with a micro squirt controller.

Tell me more. This looks completely different from the Sniper. Injectors at the end castings. Just emailed them for more info.
It seems the OP is on board and I'll be interested in how this turns out for him. That said, I'll offer my perspective as someone who has successfully done this.

I approached this with a type-3 car and a collection of stock type-3 FI parts ... minus the computer. The car was my all season daily for ~6 years operating in a range of altitudes and temperatures (Colorado). I started in almost complete ignorance of how megasquirt (and digital FI systems in general) work but eventually wound up with a reliable system that performed well.

https://shoptalkforums.com/viewtopic.php?f=66&t=132864&start=15

I found the process to be time-consuming but not particularly hard taken one step at a time. I am amazed at what VW was able to do back in the day before computers, both for design and implementation. The car eventually returned to the earth (rusted away) but the running gear lives on in a much better car. And it runs very well for a 1776 breathing through the straw that is the type-3 intake hardware.

https://www.thesamba.com/vw/forum/viewtopic.php?t=802152

To get here has been a time-consuming process. If I was doing this for someone else (a customer) this would be unacceptable. If someone had laid a worked-out solution (hardware and tune) in front of me at the start and someone was footing the bill ... it would be mostly a matter of assembly.

Using Mario should provide exactly this jump-start. The vast flexibility of megasquirt is its great advantage ... and biggest challenge.

Max

Sun May 18, 2025 12:52 pm

oprn wrote: Funny thing to me is reading these comments about the difficulty in getting EFI to run properly on a stock 1600 under different altitudes and temperatures etc. and VW did it flawlessly 56 years ago. Makes me wonder about the serious lag in development in the aftermarket industry...

Well....um...yes....no...kinda :wink:

So you say 56 years ago putting that at 1969....so you mean Bosch D-jetronic? ...which was actually released in 1967...but close enough :D

But, while some text's and people like Dr.-Djet over in other forums suggest that some D-jet ECU's had the altitude compensation function in the MPS (manifold pressure sensor) or in the ECU of some of the Mercedes with D-jet....that did not come along until a few years later and actually was not very accurate or effective.

Having been inside of every type of D-jet MPS there is...there is nothing in there that can "sense" the decrease of atmospheric pressure outside of what it is already doing to sense changes in barometric pressure inside of the manifold and change your injection pulsewidth.

More accurately....the later models of MPS had a beryllium copper diaphragm plate that reacted with changes in external barometric pressure......but that function....when barometric pressure drops....is actually an ENRICHMENT function.

More precisely that copper plate flexes for a fraction of a second to increase RATE at which the internal barometric chambers allow your ECU to enrich the engine).

It has no capability to DECREASE your pulsewidth and lean out the engine at higher altitudes which is what is required.

The other item on the very first models of D-jet with the simplified MPS wit hsingle adjusting screw had a hollow adjusting screw driving against the aneroid chambers inside of the MPS. This allowed the external chamber to have outside atmospheric pressure reference and the inner chamber was strictly operating in vacuum.
This was their first attempt at having some sort of altitude adjustment....but it simply was not sensitive enough to work and these still suffered from running rich at high altitudes.

Having that altitude compensation function incorporated into the later Mercedes ECU's....was accomplished by one of two things:

1. more of a function of when the MPS is reading enrich, when the throttle position and rpm is not dictating enrichment. Its an electro-mechanical "guess" and would be horribly inaccurate.

2. Having an extra barometric chamber widget wired into the ECU that increases/decreases the resistance by drivin ga variable resistor and most probably operates across both the intake air temp sensor (TS-1) and cylinder head temp sensor (TS-2).....both of which were notoriously only really good for setting the conditional "trend" of the engine tune (background fuel mixture baseline) and worthless for second to second adjustment of pulsewidth. It would work. Not be greatly accurate for teh sake of maintaining horsepower...but would generally work.

On that last comment...it brings us around to where Bosch DID make a concentrated effort to correct fuel mixture at high altitude. That would be with L-jet in 1974.

This is from the Clymer Porsche 914 manual in the L-jet section:

I have worked on L-jet on hundreds of L-jet injected VW 411/412, Porsche 914 a few Ferrari, numerous Japanese cars and more than a few VW bus. At one point in time I picked up a factory service tech certification for D and L jet......and never in my life have I seen one of these compensators ...either in a parts bin or installed on an engine.

But, from looking at what little I can see of the shape and the three pin output, it is obviously a barometric chamber driving (probably) a rheostat that changes the output resistance with the change in barometric pressure...and adds and subtracts resistance from both injection channels in the ECU....shortening pulsewidth.

I have no idea how well it worked but can tell you that unless your injection was perfectly tuned in the first place....its just additive and not exact unless you have one of the later models of L-jet that have a downstream 02 sensor for correction.

So yeah....having a MAP sensor circuit in modern aftermarket injection is a damn good thing!

Ray

Tue May 20, 2025 6:33 am

Max Welton wrote: andk5591 wrote: esde wrote: I've built an MS2 unit and love it. If I wanted a turn key unit for daily driving, I'd get a Dubshop basic system with a micro squirt controller.

Tell me more. This looks completely different from the Sniper. Injectors at the end castings. Just emailed them for more info.
It seems the OP is on board and I'll be interested in how this turns out for him. That said, I'll offer my perspective as someone who has successfully done this.

I approached this with a type-3 car and a collection of stock type-3 FI parts ... minus the computer. The car was my all season daily for ~6 years operating in a range of altitudes and temperatures (Colorado). I started in almost complete ignorance of how megasquirt (and digital FI systems in general) work but eventually wound up with a reliable system that performed well.

https://shoptalkforums.com/viewtopic.php?f=66&t=132864&start=15

I found the process to be time-consuming but not particularly hard taken one step at a time. I am amazed at what VW was able to do back in the day before computers, both for design and implementation. The car eventually returned to the earth (rusted away) but the running gear lives on in a much better car. And it runs very well for a 1776 breathing through the straw that is the type-3 intake hardware.

https://www.thesamba.com/vw/forum/viewtopic.php?t=802152

To get here has been a time-consuming process. If I was doing this for someone else (a customer) this would be unacceptable. If someone had laid a worked-out solution (hardware and tune) in front of me at the start and someone was footing the bill ... it would be mostly a matter of assembly.

Using Mario should provide exactly this jump-start. The vast flexibility of megasquirt is its great advantage ... and biggest challenge.

Max

Yeah MaX, I am on board. I like the dub shop system since the injectors are where they should be. Have reached out to him, but have not heard back. The one thing I don't know is if he has a cold start injector or something that deals with that. As far as the controller goes, self learning would be preferred. I do have access to a chassis dyno, and plan to use it for this particular instance, BUT want to find a system that I can use in the future as well. Once again, I am a shop. Spending a ton of time fiddling is not practical, and this particular customer is about 3 hours away, so popping by to tweak later is not an option.

Tue May 20, 2025 6:47 am

No need for a separate injector for cold start. Megasquirt has some built-in parameters that deal with cold start and enrichment that are all indexed to coolant temp (CLT). Mario has a nifty device that bolts to the head to provide CLT for this purpose.

https://thedubshop.com/modified-coolant-temp-sensor-package/

Max

Tue May 20, 2025 8:21 am

Yeah I used neither an extra injector nor IAC valve. All cold starts were handled with about 25-30 degrees of idle timing and extra fuel mapped to the output of the CLT sensor Mario sells. It would have been better with an IAC but the simplicity of my EFI setup was insane. It was such a clean install after I loomed the wires to all the components.