Inliners International Forums General Discussion Blown and Injected snowman4839 turbo build

So how is having a J&S system that will for example pull a degree or two out of a couple cylinders knocking so that total timing for each cylinder looks like 34-32-33-34-34-34 to stop any knock
Or instead, having that indicator I was talking about showing I have knock at a flat total timing of 34 then retracting them all to 32 until any knock goes away.

Is that extra 2 degrees per cylinder really worth it?
This same question applies for the MSD BTM box.

Why does any of it matter as long as your total timing is going to be the same at WOT under boost?

Isn't there a general rule of thumb to get you in the ball park on turbo timing? like start at 35 degrees total and take 1 degree out for every psi over 3psi or something?

The difference is being able to make the most amount of HP available & having your engine work in a safe detonation free environment.

If you retract timing in all your cylinders so it does not detonate, you will lose the most amount power output doing it that way.

There are two modes the J&S can work in, retard ignition in all cylinders even if only one cylinder is knocking, (safer that way)
or have it selected so it will only pull timing in the cylinder/s that is detonating. Not as safe, but will make more power.

Every engine is different, there is no one set valve for total timing for a turbocharged engine.

If you choose to use a MSD BTM ignition unit, that is great.
But as I have told you all along, you cannot always hear detonation, as your last two engines proved.

If you still plan on running an open pipe exhaust, no muffler, you are never going to hear the detonation that has been tearing apart your engines.

Having an extra 2-4 degrees of timing could be another 20-30+ HP.

I hope you do get a knock light to help w/your tuning. They are available.


You can use that (a MSD BTM ) & pull back a lot of timing so the engine does not detonate, but personally I do not like the idea of leaving available HP on the table that I am not using, it frustrates me, maybe you are OK with that concept?

Think of it as if you are only using 3/4 throttle, but you have another 1/4 throttle left, but you are not using it.

The problem with having just a knock light, is that you cannot always be looking @ gauges to see what it going on, it is unsafe to be driving that way anyways.
With electronics they can monitor what is going on w/your engine & automatically change the settings so your engine can operate in a safer environment, which will be for the better,, & safer it will be for you your engine & your passengers.

Just like what is available also, if your engine is detonating, there are devises that will reduce your boost levels also in case your engine is hearing detonation. It will make your turbos wastegate open & reduce boost pressure.


The MSD BTM box has only a certain amount of timing it can pull.

The J&S can pull more timing than a MSD BTM.

When I had the Paxton supercharger blowing through the 3 DCOE carbs on my 250 inline 6, I did not have a knock devise, (MSD B.T.M) a J&S safe guard or anything like that.

I just played it safe & had run low total timing, but I was also probably leaving another 30-50 HP available on the table.

I have used forged pistons & glad I at least had those in the engine because I am sure if I ran cast pistons I would have grenade my engine a few times from the bad detonation I had heard.

MBHD

Well I intend on building the engine and at least trying the knock light but the J&S seems like something after I get after the engine is pretty and back in and before alky injection. Especially since I'll have forged pistons in there to last me until then.

I know I've been riding the fine line (and obviously crossing it) with timing on these cheap engine setups. So how is it that these tuners and engine builders know where to add and remove timing on these engines without knock sensors before they give it back to a client?

There is no way it can be done without the aid of modern electronics. How can you combat something that you can't see or hear? Only electronics is going to be able to help out because it detects and corrects before you even know whats going on. You will either be melting pistons without it or having the timing so low that you can't possibly hurt anything, but leave 30-40 HP on the table. In that case, the biggest waste of money of all will be the turbo itself if you can't use it to its fullest potential.

Well then I'll try and factor it into my budget.

I've read about about people using special acoustic devices to hear or read knocks. Also I didn't know if there was some indication tuners could use like EGTs or HP relative to timing increase or something.

Hank touched on something that is beyond most people's control, and that is the quality of today's pump gas. How many times have you known of someone that has gotten water in their gas from a local gas station of just crappy quality from one batch to the next. Its times like that when electronics is worth their weight in gold, and can save your engine before you even knew there was a problem. I'm starting a killer turbo 250 build myself and I wouldn't consider it without all that stuff in place.

So what do these boxes cost? The MegaSquirt 3 can run individual cylinder timing trims as well as individual cylinder fuel trims. You don't have to use the fuel side to use the timing side. A know a couple people that use MegaSquirt just for timing and still have a carb on the engine. I haven't delved into it enough to know if it'll do feed back from the knock sensor to a specific cylinder, but this is stated in the ignition manual:
"Monitor per cylinder - when sequential fuel or spark are in operation the code can determine which cylinder the knock signal applies to. This enables the individual cylinder data to be recorded."

I've installed a knock sensor in my Buick 455 by enlarging a water drain plug hole, but I found valvetrain was too noisy to get any good readings. There is an updated MegaSquirt add-on chip now that supposedly makes it so it will only listen for knock during a very specific time of crank rotation so in theory won't hear the valvetrain but true piston knock, and sensitivity is scaleable by rpm. There are coolant temp trims too for engines that have valve trains that are noisier when cold, etc.
"This internal module allows knock to be monitored over a specific crankshaft angle 'window', allows gain to be adjusted and knock level thresholds to be set depending on rpm. This allows far greater tunability and should help to distinguish between engine noise and knock."

I might go that way eventually. I've already sunk the costs of the Mega Squirt 3 with fuel and spark control so it's just a matter of purchasing the add-on chip and a knock sensor.

I plan on doing a full AFR and EGT balance on my L6 engine to get the most out of each cylinder. This may eventually include individual timing trims, but that is getting down to splitting hairs and I'd need knock sensor data to back that up that kind of change (caused by something like rear cylinders running hotter, etc). I've also shifted my engine to an electric water pump with reverse flow cooling to keep the head cooler.

http://msextra.com/doc/ms3/ignition-settings.html

TheSilverBuick - a setup for a 6 is $595 if I remember correctly. Looks like that megasquirt 3 system is a hair over $400. I figure I'd just dish out the extra money to save myself some soldering and get a "bolt in ish" type setup just for ignition.

But that does get me thinking. If I did ever go over to a megasquirt system, what would be different beside all the sensors and the injectors? Would I still need the rising rate regulator since it'll have high EFI fuel pressure anyway?

Here is what a lot of the big turbo guys here in my area use.
Here
I'm fixing to do a feature article in my magazine on one of the local guys that has a twin 88mm turbo setup on his EFI BBC and runs 4.40's in the 1/8 in a 3200lb drag radial car. Not sure what this system costs, but most of the guys in those classes run it.

I always say going to MegaSquirt for a single feature is rarely cost effective, but the more features you use the more cost effective it becomes versus the cost of piece-mealing different features.

You'd have to upgrade the whole fuel system to a high pressure system. And most aftermarket fuel pressure regulators for EFI are by-pass return type with a port for vacuum/boost reference. Then the cost of the injectors, fabrication of the fuel rail if no one makes one, etc.

To speak just of the ignition system. You'd probably be about $500 in for the MS3 and knock module and wiring. Now that I think about it, I actually wonder how the J&S system works on identifying the cylinder? Does it come with a dual sync distributor to identify cylinder #1? I don't think the MegaSquirt will identify which cylinder to trim timing on without crank and cam sensors, but it can do global timing retarding on knock with just the simple tach signal.

IMO, just for the cost J&S box it sounds like the Mega Squirt could get you some pretty good ignition controls that will adjust for boost as well as do rev limiting, 2-step rev limiting, shift light controlling, high boost warning light, datalogging of AFR's, rpm, timing and boost etc. There are a dozen input and outputs you can utilize. All while leaving the carb/fuel system in place. The fully assembled MS3 is $559 plus $85 for the knock module, plut $140 in a ready made harness that you just have to shorten the wires and connect to the sensor plug. So it adds up to quite a bit more than the J&S box all in.

The only wires you'd need are the power, ground, tach signal (from points or pickup coil), ignition out to control the ignition coil, your existing O2 sensor in, knock sensor in and maybe a coolant and intake air temp sensors for data logging and if you want to adjust the timing based on coolant or intake air temp. Then a vaccum/boost line to the MegaSquirt. That's 8 wires plus a vacuum line. Well, maybe up to 12 wires as the sensors will want ground wires too. If you do this bare minimum you can probably just use an MS2 and an aftermarket knock sensor board. $435 fully assembled plus $67 wire harness and I'd have to dig up the external knock module. The MS2 does essentially the same thing as the MS3 minus full sequential fuel and spark control. More cost competetiveness and can be upgraded to an MS3 down the road if you decide to take your project in that direction. MS2's batch fire or semi-sequential works perfectly fine, especially when compared to a carb.

Just food for thought. I know it's not for everyone.

The more I read about megasquirt, the more and more interested I get.
My fuel system should support a fuel injection setup. I have a Walbro pump that should put out max 70psi with 209lph at that pressure. That should be plenty for a fuel injected setup. Correct me if I'm wrong.
I would need a new regulator though, mine is set up for a carb setup (only adjustable from 4-25psi base pressure)

Only big hurdle I'd have is finding all the appropriate sensors to install. Does megasquirt have recommendations on which ones to use somewhere? Plus that seems like it'd make up a third of the cost. $Base system + $Sensor array + $injectors

It can be calibrated to dang near any sensor.

I pull my coolant and intake temperature sensors from GM cars in the junkyard with the plug and length of wire attached. Likewise for the cam sensor and someday to be installed TPS sensor on my OHC L6. Most aftermarket throttle bodies come with a throttle position sensor though, which are usually a GM part. Likewise for the knock sensor, grab a GM one and you'll likely be good, but most are calibratable to an extent (most knock sensors are rpm/frequency sensitive). The MAP sensor is built into the Mega Squirt. That's all you really need unless you are going to sequential injection and distributorless then you'll need a crank and cam sensor.

And again, you can always do it piecemeal wise. Ignition now and fuel later. That pump does sound sufficient.

Does it just have the option to pick from a list of presets for whatever sensor you ended up using? If not, I can understand calibrating TPS (open, closed, then it does the math for in between) but how would you calibrate something like a MAF or coolant temp sensor?

If I wanted to switch to megasquirt for MPFI and timing control with my distributor, what all sensors would that involve? MAP, MAF, TPS, IAT, narrowband O2 (could I just hook it into my wideband), ...?

and lastly, how would you install a cam positioning or crank positioning sensor on a car not built with one. Is this something that you would put on the outside of the harmonic balancer and track with a magnetic pickup?

You don't want to run a MAF. The MAP is built into the MegaSquirt board. For fuel, you want the coolant temp sensor and the intake air temp sensor for sure. Getting GM sensors from just about any vehicle from like 1988+ work. Coolant temp sensors work fine as intake air temp sensor as well, particularly when running boost. You can find a point to thread the sensor into the intake charge pipe and you'll be good.

Your wideband O2 can be wired right into the MegaSquirt. Depending on model, it may support using both the gauge and hooking up to the MegaSquirt (two programmable outputs).

I don't even have a TPS on my OHC L6 right now and the only side effect so far is a slight hesitation coming off idle from a stop, otherwise it runs just fine. Though some things are harder to tune without the TPS and associated data.

You would only need a crank and cam sensor if you are going to sequential fuel injection or distributorless coil packs. You can control timing with out those by simply hooking the "tach in" wire of the megasquirt to either a points contact or to an HEI pickup coil, then the "IGN out" wire to the negative side of the ignition coil (with +12v on the positive side). You would have full ignition coil control at that point.

If you definitely want the extra extra precision of sequential or coil packs like on my OHC L6 project, you will need a trigger wheel and crank sensor, which I've bolted a universal one to the front of my engine and used a $25 hall effect sensor from digi-key, but any pickup sensor you can come up with works. Stock Ford VR crank sensors are rugged, common, cheap and popular.

The cam sensor can be made in a few ways from an old distributor. You can smooth out all the lobes except one on a points distributor so the points contact opens only once per distributor rotation and that becomes a cam sensor, or like my Buick I took an HEI distributor and ground all the pickup coil and reluctor teeth off except one each and that creates one pulse per distributor rotation and makes a good cam sensor. Or like I did on my OHC engine and cut down the distributor in a way to create a tab and got a hall sensor to read the tab as it spun by. The most difficult part of the MegaSquirt is having sooo many choices.

Alright so in general the sensors don't seem that hard to find.

Are TPS sensors something that are integrated into a throttle body or is it something I can work into a cut up carb that I would use as a throttle body?

Why would you need BOTH a crank and cam sensor for sequential fuel injection since they're directly related? Couldn't you just tell the computer that the cam is half the speed of the crank?

If you do need both then wouldn't you HAVE to switch to coil packs/coil over plug setup so that you could use a hacked dizzy for the cam position sensor?

Is sequential worth the extra money/time/complexity over TBFI?

The TPS I've found to be a pain to make work with a carb, but it's not impossible and I know people that have done it. My OHC L6 uses a Holley carburetor body with the float bowls removed.

The reasoning for needing a cam sensor is to tell the computer if the cylinder is on the compression or exhaust stroke. You can run semi-sequential, where you pair the injectors and pair the ignition coils, then you only need a crank sensor and it'll spray half the fuel amount twice per revolution (so one full fuel load when the intake valve opens) and it'll fire the ignition coil pair on the power stroke of one cylinder and the exhaust stroke of another. I continued using the rotor and distributor cap on my Buick when I converted the HEI to a cam sensor. You don't have to do away with the distributor unless you want to.

Or you can just run the batch fire system and wire it up just as I described without a crank sensor and still be able to electronically control timing and use a standard distributor cap and rotor.

Batch fire and single coil will get you 98% of the HP in the combination. It's only really worth it if you are going for that last 2% of horsepower (from fuel trims and tighter ignition control), and that is usually only worth it if you are racing in a very competetive racing class.

This is what the HEI in my Buick looks like. I ran it another year or two with a rotor and cap distributing spark. Now it just sits under a cap since I went to coil packs on it too. This mod would work fine on an L6 HEI.

I have a feeling this will never happen. An EFI set-up. Nice to talk about it though.

You will need someone who knows how to use it after he would get the Megasquirt installed.
It is not easy if you have never done anything like this before.

He would be better off purchasing a new( Fast TM) self learning EFI system. And we know that will not happen. That would be about $2000.
But the good side is it will start right up & self tune.

They now make systems w/8 injectors for high HP set-ups.

With the J&S it can detect detonation from one or more cylinders & remove timing from one or more cylinders.

I do not think there is any aftermarket system that can do that.

The Megasquirt would work like the older knock sensor systems. if one cylinder detonates, it will reduce timing in all cylinders.
The J&S has had 20-25 years of development into it.
It is not a new concept.

MBHD

The MS has a bunch of options. Would have brought it up sooner but did not think Snowman would want it.
Timing control based on load/ boost is what I use on the ELCO. Works great on the street.

MBHD, I stand corrected if the J&J unit can do timing changes to diff. cyl that's great. I just would like to know how it can tell it's a knock and not another noise.

I am wanting to get a ms system i am a mechanic but have never done any tuning is this going to be overly difficult for me to get going? I want to control timing make a multiport injection now to steal a thread just is getting very informational

That's no spirit Hank!

I hadn't really considered any of this until the last week. I'm about to get this research job over the summer which will give me a lot of money for the project. Up until recently, I planned on a couple more disposable engine setups until I had enough to get a forged final setup. But with my job/awesome scholarships, I should have plenty for a new cam and valvetrain, set of forged pistons, the associated machine work. And now I can reasonably consider EFI and alky injection all within the next 8-10 months.

For almost the same price and benefits and options MS3 gives me vs J&S, I think it's a better idea. The timing control might not be as sophisticated but it offers a lot more in terms of info and control to ensure I don't grenade my big final investment.

You are right though, no way I would buy one of those TBFI systems for $2000. That's ridiculous. When I could build a SFI for $1200ish.
I'm a computer science major with a 3.94GPA that has been programming since 7th grade and messing with with microcontrollers since I was a freshman in high school. Computers are my life and the other half of my life's aspiration besides cars. I don't think a megasquirt install is what's going to beguile me.

The way it looks now, I work all summer and then come late july or august, I start amassing parts for the build. Already got the head rebuild kit from tom. Should be getting a price on the cam soon. After my first big payday, I can get that set of Ross pistons and the 7/16" studs from Tom. Then I can get ARP rod bolts and send the block and head off to machine the block for the pistons and install the pistons on the rods, resize the ends of the rods with the arp bolts, remove the old studs, mill the stud boss down, and drill and tap the boss for the new studs. Probably buy and have them install lump ports too. Already had new cam bearings installed in my first block so that's good to go.

Then comes the building the EFI system. TheSilverBuick gave me some inspiration and ideas. I'll figure that out over a couple months. Probably put the whole valvetrain in but leave the stock cam in there so I don't grind it down in case I have problems with the EFI the first few times around. Then once it's running right, just swap in the new cam, fire it up, and break it in.

TheSilverBuick - read through your whole thread. Great stuff! Awesome build! How are you planning on doing the injector bungs in the final build? Weld? Thread? How did you manage to build that fuel rail? Did you have clearance issues with that middle runner hitting the throttle body and/or the manifold itself?

They are tuned for different frequencies. I am not good @ explaining how it works.

All I know is it works to detect detonation not engine noises, valve train, piston slap etc.

MBHD

Snowman,
if you can afford EFI great! I was under the impression you did not want an EFI system & was out of your price range.


MBHD

Also keep in mind where the "siamese port" falls into the mix of things. It's not compatible with most EFI type designs....

You can't drill 2 holes next to each other per runner in a clifford or offenhauser intake for each intake valve? Like how tom's setup was on that Clifford EFI intake he used to have?

What's specifically not compatible about it?

Because you have 2 cylinders sharing just one intake port, and they have a tendency to rob fuel from the shared cylinder because of this design. You'll have one of those cylinders set to start drawing fuel into the cylinder and then the siamese cylinder will start to open its valve and draw fuel in from the other one, robbing it of its full amount of fuel meant for the first one.

I just went out and thought it through. I don't see any reason it shouldn't work. The neighboring cylinders are on opposite parts of the stroke. Like while 1 is on the intake stroke, 2 is on the compression stroke.

Snowman, this has been a problem that has been realized for close to 50 years with all "siamese" port engines, its there even though you can't understand it. Dividing the port helps a lot, Tom has done this with his turbo 292.

AFAIK Tlowes, Holley EFI is not really made for a Siamese port head.

Because a "siamese" port engine is such a different bird, I'd say that none of them really are.

Well I just read a little bit about it but it seems like the problem is centered around them opening at the same time like you mentioned? If they don't then what would be detrimental to performance?

The problem becomes that the cylinder intending to get its full dose of fuel is short changed by the next cylinder taking some of it and can go lean under high load conditions. Harry has fought this problem before also, that is why he no longer runs a "siamese" head.
Not that it can't be done or shouldn't be done, but know that the issue is there. That is just something you don't want to have to deal with when running 15 psi of boost under heavy load along with everything else to consider.

I understand what the potential problem is but *how* can it ever happen if neighboring intake valves are never open at the same time?

They are open at the same time, that's why it is a problem! There are cylinder pairs that are worse than others, and it is even made worse by the cam profile you use.

Also, keep in mind that most of Harry's success and advice are coming from his use of a non-siamese head. He can also tell you where the shortcomings from the siamese head are in this area if you ask him specifically.

:facepalm: I forgot about the Chevy having siamese ports. That makes EFI less ideal. Fuel injection still works, just not as optimized because there is fuel scavenging, even when running full sequential. Basically once over around 2,500rpm or at high throttle the injectors are open longer than the intake valve, so fuel scavenging/stealing happens in a siamesed port. Running batch fire will still run fairly well, just not perfect. I'm mostly pushing for doing the ignition side first and foremost if you are going to spend that kind of money on an ignition box.

I had all sorts of interference issues with the injectors and throttle body, and it is less than ideal for sure. My final intake will be a sheet metal intake and I'm leaning towards welding big nuts to the steel runners and running the aluminum bungs through a die and thread the injectors bungs in. BUT I have not seen if I can get a large enough die to see if it's even feasible. I may go another way with steel tube's of the right diameter or such.

I'm a geologist and I learned all this EFI from reading, playing with the software and asking questions when needed. No one has ever tuned my car's other than me, and sure it's been a learning experience, but it's not impossible.

Even so. I just went outside and turned it over by hand. Cylinders 3 and 4 are exactly 180 off on the cam. On the exhaust stroke on the 3, 4 should be on compression and vice versa. That should never be a problem

1/2 and 5/6 are different. One goes through the intake stroke then closes then after about 30* past BTC, the second one reaches TDC and begins opening the intake valve. Meaning the valves aren't open at the same time assuming the valves directly follow the cam.

The only explanation that made any sense was that the injectors would spray longer than the duration of the intake which WOULD steal fuel. Anyone care to explain why that would ever happen? Can't this be solved with larger injectors so that you could shorten the pulse to within the time the intake valve is open?

EDIT:So I just read some on the megasquirt site about batch vs sequential. Seems like it just preloads the valve at high rpms because the injector can't keep up. My question on a bigger injector with short pulse width still stands though.

Injector time is measured with Duty Cycle. My Buick with 232* of duration at 0.050" has the intake valves open about 32% of the time. So once the duty cycle goes over 32% the injector is spraying at a closed valve waiting for it to open. Once you are at 50% duty cycle, the injectors are open half of any measured time. In one second the injector is open .5 of a second, in one minute the injector was open 30 seconds of that minute. So the problem becomes, when injector 3 starts spraying, cylinder 4's intake valve is open and sucking the first part of that fuel and when intake valve 3 finally opens again it gets a lean charge low on fuel because cylinder 4 just ran rich with part of 3's fuel charge.

You are not going to get injectors big enough to escape the duty cycle scavenging issues. You will run into the issue anytime over around 32%.

**Originally I wrote 8%, I got the 8% number from the amount of time the intake valve is open AFTER the exhaust valve has closed on the overlap period. Which is determined by the LSA.

Alrighty then so screw fuel injection. That was a whole lotta nothin haha.

Aw well. I guess I'll stick to my controlled fuel leak. Looks like money just freed up for that J&S Hank.

I guess everything in that roadmap is the same except I'll be wiring that J&S instead of the MS3