1/4 scale V8, first project.

[LongRat]

Superb progress here! This is what I have been waiting to see.
Keith, what made you fit the tube in the path of the injector? I wonder how it would run without that. Seems to me that there is sufficient runner surface and intake tract for fuel to stick to (and keep your engine running for many revolutions after the injector pulse), that adding this tube in the airway might only serve as an air restriction and the extra atomisation might be of little consequence. Or I might have completely misunderstood, what to you think?

Also, how are you pressurising your fuel?

 

[keith5700]

Just done a progress report but lost it again before I could click the button.
Will try later. Is it me or this forum??

 

[///]

Just done a progress report but lost it again before I could click the button.
Will try later.

Damn! I'm always hanging out to see news on this build.
Really looking forward to the new update.

Is it me or this forum??

I have lost a couple of posts too. I have now started highlighting and copying the message before hitting submit. If it fails I only need to paste in the copy and then try submitting again.

 

[gbritnell]

Hi Keith,
There's something wrong with the forum. Yesterday I wrote a long answer to some of your questions. I made sure that I was logged in before doing so. When I got ready to post I thought "better copy this because I've had trouble before". Sure enough when I hit the reply button it said I wasn't a registered user but yet at the top of the page I was logged in. I had to relog in before I could post the message. This has been happening since the changeover and nobody seems to have an answer. It sure is frustrating to have to copy everything you want to post just to make sure you don't loose it, especially when it doesn't happen anywhere else.
gbritnell

 

[happykingkong]

I'm surprised at you step by step to complete the engine manufacturing,Shocked me !

Cheering for you!~

Greetings from China
happyKK

 

[johnny1320]

did you get your smoking problem repaired Keith?

 

[keith5700]

I had it running again tonite. It starts so easily with the injector.
All the top plenum is now fitted so the twin butterflys are restricting the airflow.
There is quite a bit of smoke still!
The reason isn't clear yet though.
George, only one side of the engine has the top oil pipe working yet, for the cams.This side does smoke more than the other.
but also one cylinder doesn't seem to be working on that side too, as the exhaust pipe is a lot cooler than the others.

So at the moment it could still be either oil or fuel.

I am using a full sized Walbro pump for the fuel. It is running at about 40psi.

I'm not sure a 555 timer type trigger would be much better than using a microswitch off the points cam?
What it needs is a trigger off the crank, or off a cam, to fire once every 2 revs, and then a potentiometer off the throttle shaft to vary the 'on' time of the pulse.
Of course this is the basic operation of some full sized engine ECUs, so that's no surprise.

I have no idea how to wire this up, so I'm currently trying to sell the Rivett lathe, so I can buy a Megasquirt ECU.
I am familiar with setting up and programming full sized car ECUs, so this seems to be the best way to go, for me.
It will also control the spark advance curve, etc.

I tried to purchase a 38g/min. injector from MBE motorsport, but they are out of stock.

Thanks for all the replies since my last posts, I take it all on board.
Cheers, Keith.

 

[johnny1320]

Hi Keith, if the side that has the oil feed connected is smoking, is it possible that oil is getting under the tappet and is being injected past the guide by the down force of the tappet? I am not sure of the clearance you are running around the tappet. The injection system sounds promising, when are we going to see a video of this jewel running, I don't care about a little smoke.
Cheers

John

 

[gadabout]

Keith have you seen this site?
http://www.f1-2000.co.uk/index.php?f=about_this_project

regards
Mark

 

[keith5700]

John, when I designed the tappets I came up with a way which didn't hydraulic any oil down into the intake. I'll try and get a drawing up.
Mark, I came across that site a while ago and contacted the person involved. His name is Roger, and we have been in contact ever since. He's a proper nice bloke, and I believe he is a registered user on this site.

Some of his machining makes my effort look like I belong in the Stone Age, so I hate him really!

Keith.

 

[gadabout]

Keith,
Ah good, thought he might be able to assist with the injection side of things, his workmanship is superb isn't it? Much like yours for sure!
regards
Mark

 

[keith5700]

Well I managed to sell some bits (not the lathe) and have bought a Megasquirt ECU MS2 Extra, which will control the injector,
and also, as a nice by product, will provide full spark control, rev limiter, etc.
I've just about got my head round the instructions, and have got the laptop talking to the ecu.

I am going to get all the sensors working on a dummy engine, (electric drill) before installing it all on the engine.

i have got the crank sensor working off an 8 tooth trigger wheel so far, although the coil packs don't seem to want to fire yet.
The built in MAP sensor works ok, so I need air and water temp sensors and throttle position sensor.

I have a narrowband O2 sensor but don't think that's worth installing on a small, smokey engine?

Hopefully some progress at last.
Cheers.

 

[Lakc]

Its not likely that the O2 sensor will ever stay at the constant high temperature necessary for proper readings. Plus, you have to have an absolutely leak proof exhaust path as well.
Looks like you have a whole lot of fun ahead of you.

 

[FannBlade]

Even the engine stand is awesome!

 

[jwcnc1911]

This is over the top for a first project. Wow!

 

[LongRat]

Keith, please post a video soon! This looks absolutely amazing. And very neat and tidy. Personally I have never seen a model engine properly controlled with a real ECU and I think it may be a first. This is the way to go, it will be the first of many. Great achievement!

 

[keith5700]

Jeff, the O2 sensor is heated, so it may work, but judging by the amount of oily crap in the exhaust boxes I think it may bung the sensor up. If the ecu works well and I start to like the engine again then I'll strip it and make new pistons with an oil ring,
And maybe try and seal the valve guides a bit better.

Longrat, I am definately going to do a video, when its all together properly.
Hopefully in the near future.
I would like to see if its been worth going the injection way first. It may turn out to be an expensive waste of time!
If it works like it does in my head, then my next engine will have individual injectors and fuel rails, etc. that should look really cool.

 

[Lakc]

Jeff, the O2 sensor is heated, so it may work, but judging by the amount of oily crap in the exhaust boxes I think it may bung the sensor up.

I think that heater is more of a "helper" rather then providing 100% required temperature, but your really into uncharted territory here.

If you want to test it, plug the heater in on the bench, and hit the sensor tip with some argon or helium or any gas that has no oxygen in it. (propane?(be careful)) It should produce 1 volt based on the difference in oxygen between the pipe side and outside.

If it works, great! Feel free to carefully remove the metal shield and squeeze the sensor into rather a long and skinny oval cross section of exhaust pipe. Thats going to be one seriously odd looking weld bung, however.
Remember, absolutely no exhaust leaks allowed. Even the smallest leaks on full size motors always seem to wisp a little atmosphere right across that sensor. I think they were designed by gremlins....:wall:

 

[robcas631]

Terrific engine!

 

[keith5700]

Well, it was going ok till I powered up the coil packs. As soon as the sparks are working the laptop loses communication with the ecu.
After loads of searching and a few e-mails it turns out I need to be using resistor spark plugs, otherwise there's too much interference for the ecu to handle.
I've changed to carbon ht leads, which are 13k ohms, but these made no difference at all.
If I could figure out where to connect my oscilloscope leads I'd see if I could see the voltage spikes, and try and reduce them without having to somehow make resistor plugs.

I'm still trying to find out why you must have resistor plugs, which seem to be all around 5k, when I already have 13k leads.
Maybe the resistor has to be physically next to the spark?

Anyway, at the moment I'm so disappointed with this Megasquirt I really wished I hadn't bothered with it. It seems they're mega sensitive to voltage spikes, after reading the never ending pages of internet questions on the same subject.

I've no idea how I'm going to make resistor plugs, but at the moment it looks like either make the plugs, or sell the ecu and try and make some crude injector controller, and re-fit the dizzys.
What I don't want to do is spend ages designing and making new plugs, if they still don't work. That would be the last straw and the ecu would probably get a hammer through it.

You can probably tell I'm a bit hacked off with it all.
I think trying to see the spikes on the oscilloscope may be the best next step, and then try and cobble up a quick resistor plug and see if it reduces the interference.

Cheers.