Converting GM HEI to CNP (or waste spark)

I pulled the MSD billet HEI distributor out of my truck this weekend. I bought it new at least 10 years ago and it's probably seen almost 6-figure mileage in the interim. It's time for a cap & rotor, so I figured on a good cleaning and checkup. Upon inspection, the advance mechanism is flat worn out. Not unusual wear, just a lot of normal wear as can only be earned over 10s of millions of revolutions. Given the superior quality of build and materials, I'm a bit surprised, but really it's just a modestly improved version of what GM came up with when they rolled out the Delco points dizzy...what, 70 years ago? Difference is that the center cam on the MSD is TIG welded to the shaft, thus not replaceable. (Yeah, I know welding is also "only temporary") I'm not even going to try to send it to MSD for repair/rework. A new copy of same distributor is currently $530 and I don't need the module or coil. So, the rampaging Project Squirrel in my mind immediately got to thinking about Options.

First, let's list some reasons a classic GM HEI distributor is a good pick:

-Easy to find parts for and nearly ubiquitous.

-Easy to work on. Every self-respecting old-school wrench knows their way around one.

-Simple like rock. They just work, in that sweet spot of mediocrity, until the module fails. I do carry a spare MSD module since I need the dedicated square-wave tach signal for my Microsquirt TCU though.

-Cheap, in stock form. Someone will probably give you one from their parts stash if you ask nicely. I probably paid $2xx for my MSD version originally but the dollar is getting more expensive by the second.

-Decent performance, especially the fancied-up versions with hotter module and coil.

And some reasons they're not so great:

-Limited flexibility. You either run an HEI or something else, it's hard to hybridize or cross-pollinate (external coil, for example).

-Limited coil energy. This one isn't the HEI's fault, just a function of forcing any single coil to fire all an engine's plugs within a finite time window. After noticing firsthand the massive improvements in power and driveability converting a different vehicle from a single coil to waste-spark, I am fully aware the detriment of dwell-limited spark energy.

-Air makes a terrible conduit for that hard-earned spark energy. A spinning rotor sending spark across an air gap to the cap terminals is a fundamentally compromised design. Any mechanical distributor shares this handicap. Bonus demerits for the wear caused by arcing on said contacts over time, as well as the ionized dust that it generates and deposits in the airspace.

-Imprecision is inherent in any mechanical distributor, ultimately manifested as timing variation. There’s a lot of pieces, tolerances and clearances in play.

-Adjustability of advance curve is limited to what can be accomplished by geometry and Hooke’s law.

-Vacuum advance canister. Not that I have a better idea, but rotating the pickup coil is literally a "move the goalpost" approach. Ugh. 

-Central, fixed location means long and unequal length plug wires.

-Sensitive to environmental conditions. Raise your hand if you've ever power-washed one by mistake or hit a large puddle and had to dry it out to get home.

So what are my options? There's a damn good reason GM and everyone else has gone to COP/CNP and I'd like to bring those benefits to my old BBC.

1. Waste-spark setup. This would use two coils with 4x outputs each. Not ideal on a V8 since each one needs to pair cylinders by firing order (1 with 6, 8 with 5, etc) so you'd have to run plug wires to both sides of the motor from each coil. The up-side is that whatever you use to fire the coils only needs a basic 90° increment trigger with a 1x crank reference and doesn't care where the motor is in its cycle, so no cam position sensor needed. I keep thinking about using the existing HEI (or another distributor) just for a trigger. I don't have a particular candidate ignition driver picked out but there's got to be something suitable for not a lot of money. Uncertainty remains how to interface with it and how tunable the timing is. And what about a MAP sensor for vacuum advance? And man, that rat's nest of plug wires... As details are addressed, this whole approach starts to sound more like an overly complicated DIY project with no upside over...

2. Individual coils. Prior to the advent of off-the-shelf dual-sync distributors, the only way to do fancy multi-coil ignition or sequential fuel injection on a dinosaur-level engine was an add-on crank-trigger and a 1x cam signal squeezed out of something creative. This can still be done but compensating for that extra ¼” trigger wheel thickness on a 3x V-belt drive is a royal pain. ATI makes trigger shells for their dampers that could work (I’m running a 7” ATI now) but they’re about the same price as the whole plain version. As the aftermarket support for EFI conversions has exploded in the last 20 years, so has the availability of dual-sync distributors. Currently, MSD makes complete conversion kits for V8s from the Big 3. Their Chevy kit comes with said dual-sync dizzy, eight LS2-style coils, plug wires, an ignition controller/driver, harness, and the other misc bits to make it work. It currently retails for $1425. Add your own coil brackets and case closed, albeit in trade for a rather large bag of gold. The controller does have a built-in MAP sensor for vacuum/boost timing adjustments and even a temp sensor to make temperature-driven adjustments. It's tuned by PC over USB, with user-selectable table values. What’s the downside to this approach? Mostly cost, with the DIY itch not being scratched a distant second.

So after that lengthy preamble, what other options does the hive suggest to implement an individual coil setup on a classic BBC?

Why?

Are you spinning that lump fast enough that a distributor becomes unreliable?

If you're looking for hp per dollar spent, you're far better off putting your money and time somewhere else.

If you're just doing it because you can, I'd bolt a crank trigger to it, add Megajolt and call it done.

"Why" is a fair question, and arguably the right one to ask first.

This particular rat sees 6300 at the top of 1st gear, which is well within the RPM range of a good HEI. So no, it's not that the current setup was unreliable or inadequate when healthy (it hasn't been running quite right recently and now I know why). The real impetus here is that I need to buy new parts since the current unit is unserviceable and essentially unrepairable. I'm having a hard time with the cost of a new like unit, and Project Squirrel says that for that coin, I can do better.

The performance gains to be had with CNP are very appetizing, and although I'm a diehard carb guy, I do like the options afforded by computer-based tuning (of spark, fuel, or anything else). I also wouldn't mind NOT buying cap & rotor sets every couple years and occasionally scraping those terminals clean in the meantime. The fact that the dual-sync "distributor" is plug & play ready for a future EFI conversion is not totally irrelevant either.

Project Squirrel also has a say in this, although he does have to concede to practicality and the cost-benefit equation.

I'm a bit concerned about the MSD DIS setup too. Besides the choke-on-your-cereal price tag, MSD has encountered notable quality deficiencies in recent years associated with globalization of supply chains, component quality, and shortage of skilled labor (and engineering) just like everyone else. My Spark Disappeared is not the acronym I want to be shouting while stuck at the side of the road due to an unexpected failure.  And lifespan aside, do I really want to trust the roadworthiness of something I use to haul junk across multiple states to the purported non-volatility of that DIS memory? Probably moot since the trans is already electronic.

I'm a huge proponent of ditching a distributor if you already have an ecu on board.  It is unnecessary nonsense on an engine with a standalone ecu and is really just getting in the way.

However if you are a diehard carb guy, I can't see spending that sort of money for minimal gains.

If you are dead seat on it, I'd skip that msd crap and for about the same price (or less) buy an ecu from holley, megasquirt, etc and drive the coils directly off of that. Pick up some used coil racks instead of aftermarket ones.  And if you ever want to venture into this millenium you are most of the way there already.

This thread is relevant to my interests.  I have a 347 SBF I'm putting into a Fox Body with a Holley Terminator Engine Management, and to get it running and on the road, I'm using the TFI distributor.

But, long term, I want to go CNP using LS Coils, and I'd prefer a crank trigger over a dual sync distributor - namely for the same global supply chains and aftermarket parts failures you name above. 

There are a few places where you can buy 1/16" thick trigger wheels, which would be much less painful to adapt a belt drive system too.

I put 8 D585 coils on my SBC in wasted spark form run by an old MS2 and a 36-1 crank trigger wheel + Hall effect sensor. About the cheapest way to CNP. Works great but it's basically a dedicated autox engine. Had the LS coil brackets welded to a set of factor valve covers to make for nice short plug wires. 

Given the desire to put quite a few miles on the system, popping for 8 new-ish coils would seem advisable. That's $2-300 alone for decent ones. Plus at least a couple hundred for parts to spark them and I have a hard time getting to a lower number than a decent replacement HEI distributor. Might be more reliable and cheaper to repair in the long run, but could have teething issues too. 

Using a dual sync distributor to trigger CNP may be convenient, but it is missing a lot of the timing accuracy benefits of an OEM system. The eight tooth trigger doesn't have as much resolution to track rapid RPM changes, and timing chain slop and cam harmonics reduce its accuracy. I would get something crank triggered with a high resolution missing tooth wheel.

ShawnG said:

Why?

Are you spinning that lump fast enough that a distributor becomes unreliable?

It isn't necessarily a speed thing, and BBCs are a lot better than SBCs with their higher gear count on the oil pump and larger diameter cam cores, but spark scatter due to the camshaft twisting under load and harmonics is still a thing on the BBC.  The cam is driven at one end and the distributor and oil pump are driven at the other end.

Even just running a four position trigger wheel on the crank and using it to control a MS for spark timing through the distributor is an improvement.  (Should be easy if you use a 7 pin HEI module so the MS can easily control spark timing) But sticking a 36-1 on the front and turning the distributor into a cam sensor to run sequential spark with eight coils is even better-er.  

An MSD ignition is way better than using a module, too, and probably cheaper than going sequential.

Thanks guys, this is helpful.

I've done several megasquirt-based installs and am very happy with the results. I also enjoy the definitively DIY build & development process. This truck runs a microsquirt for 4L80E trans control, which is a dedicated FW build exclusively for trans control. It does have the ability to talk CAN which makes things easy if I do a MS EFI setup. Definitely on the table, but I'll keep the carb for now. That's an interesting suggestion to use MS for spark only though.

Megajolt looks appealing, but I'm not convinced it has any advantage over MS running spark only, especially for the money. EDIS is a solid option, but I don't have any history with it and/or any Ford parts lying around. Megajolt also requires a specific trigger offset which limits sensor placement vs trigger options.

HotNotch, are you using the EFI system at all to do fueling and interface with the TFI, or just running the TFI and a carb until you go all-in? Please post link/info for said super-thin trigger wheels. I found the ones from trigger-wheels.com but they are all 5mm thick.

JMcD, agreed on the new coils being a wise call. If I do go with used, there would be 2 spares on-board. Even for the same or reasonably more money, my preference is still to ditch the distributor. How are you running those 8 coils in wasted spark config? Did you just gang together pairs and drive them with a single BIP373 or IGBT? That would alter the effective primary impedance and possibly invite unequal coil performance. Or are you using the MS to run an add-on driver board?

I really want to NOT have a distributor at all. Besides a crank trigger being far superior for timing accuracy, I also don't want to compromise spark energy with the rotor -> cap air gap (plus parts wear). The only thing a "distributor" would do in the ideal end state of this project is drive the oil pump and seal the galley. The drive spud from the LT-1 (the '90s gen 2 SBC, not post-LS version) is too short and bolts to a boss in the valley, otherwise I'd be tempted. I'd probably modify a stock HEI instead for a 1x cam signal.

The primary reason I pulled the truck into the shop and started poking at it was actually a clicking noise coming from the front of the motor (heard best through the grille), along with a suspicious belt chirp. I found the distributor advance wear first in the course of a thorough once-over, but I also found this:

Folks, you've got to earn this. This is the original pulley from 1983. It's been on the truck thru 8 build iterations of three different motors for a total of 260k+ miles (odometer stopped counting when I put the 4L80 in in 2016--no speedo cable). Really glad I didn't use the A/C last time I drove it. Cracked completely thru and flopping at two bolt holes, third very close. Balancer face is marred but OK. 

It's like the truck and I both want a crank trigger. I looked at the belt drive & accessory brackets again, and I really don't want to add any spacing behind the crank pulley. That will snowball into a nightmare project, likely ending in snazzy billet brackets for everything or a serp belt conversion since at least two of the brackets can't be spaced out without fab work. I've spent a lot of time on the accy drive and pulley alignment, so it's sorted as-is anyway.

No aftermarket options exist for a factory-fit pulley with a trigger wheel or the ability to add one. I could get a replacement steel pulley and weld on a trigger wheel, but I really don't like the quality of replacements--every one I've seen has poor V-groove geometry and runout not at all meeting SAE J636 or my standards. I actually have an old chrome replacement for this, but it's terrible and I won't use it. March has a decent aluminum pulley, but all three grooves are 7" so I'd need to contend with drive ratio and other pulley changes.

I designed my own version in SolidWorks with factory sizes and offsets. Aluminum seems like the obvious choice for material, but how do I attach a ferrous trigger wheel? Even an intelligent mechanical design with clocking and clamping done well can't get around the CTE mismatch and potential for things to go South with temp cycling. Maybe machine it out of CRES instead, but that would be a 7.5lb pulley without drastic and expensive lightweighting.

So now what do I do?

Nockenwelle said:

 

...Even an intelligent mechanical design with clocking and clamping done well can't get around the CTE mismatch and potential for things to go South with temp cycling...

So now what do I do?

I don't think I'd worry about a properly designed shrink fit steel ring on an aluminum hub.  The aluminum has the higher CTE and should never be cooler than the steel.

Alternatives could be a flat plate/sheet metal steel trigger wheel trapped between the pulley and the balancer or aluminum with flying magnets.

In reply to Nockenwelle :

Here are some options for Trigger wheels

This one looks SBF specific, so of no use.  But may reach out and see if they can make one custom?

https://hgmanufacturing.net/products/sbf-36-1-trigger-wheel

Here are some other options

https://www.diyautotune.com/product/6-3-4-36-1-trigger-wheel/?gQT=1

https://www.ebay.com/itm/326423637009?chn=ps&mkevt=1&mkcid=28&google_free_listing_action=view_item&srsltid=AfmBOoo5CH4s7yK5R6X1KXjNeGdwu_YBMQEoK0_AX4IO7rkrrKlN1ZbbzRk&gQT=1

In reply to Nockenwelle :

I am using the Terminator X from Holley to run both the EFI and Ignition.  For now, I'm using Holley's TFI harness with the factory distributor, but running a 4150 Throttle body on an Wilson Performance EFI Converted and ported Dual Plane

Eventually, I'll do the crank trigger and an Explorer Cam sensor, and LS coils, but for now, I just need to get in the car and get it on the road

In reply to Nockenwelle :

Innovators West and ATI will both make balancers with flying magnets embedded in them, so no added thickness - just need to figure out sensors and mounts

ShawnG said:

I'd bolt a crank trigger to it, add Megajolt and call it done.

I have done this on several vehicles and it just works.  If you want to get real crazy take a 72 tooth spur gear from McMaster car and make it a to a 72-1-1 and mount it in the distributor. Then add a fc rx7 ignition trigger sensor inside the distributor and use it to provide the trigger for an edis controlled Megajolt. Mount the edis to the firewall and run wires to the plugs and wiring to the Megajolt that you can stash in the glove box.   This is the KISS principle for everyday ignition that just works and lets you have complete control of the timing. 
 

EDIT.  Putting the trigger inside your distributor lets you clock things anyway you want. Solves your crank mounted trigger problem. 
 

This is my adventure back in 2008 converting my 924s to edis using a cam/distribuitor triggered Megajolt ltjr  not exactly the same but the principals are the same  

https://forum.autosportlabs.com/viewtopic.php?t=2026

In reply to dean1484 :

Wouldn't that cause trigger issues because of the distributor rattling around due to the cam harmonically twisting and the lash in the timing chain and distributor drive teeth?

When GM went to multi toothed engine position measurement (as opposed to having a controllable ignition module reading the 8 teeth on the reluctor) they went to crank trigger, or a sensor on the front of the cam.  GM being cheap, they would have just bunged a trigger wheel in the distributor housing if they would have thought it'd work well enough.

At the very least, I wouldn't go with a 72 tooth wheel, I'd go with a 24 maybe.  That'd work for an 8 cylinder engine.

Note that even Mazda took the trigger wheel off of the distributor and stuck it directly on the crank, for rotaries.  

Trust me it works and no you can not use a 24 tooth with edis. 36-1 at the crank or 72-1-1 at the cam for edis. 
 

if there was that much slop you would have valve timing issues. Also remember your spark is already being controlled by the distributor that has the timing chain slop that you are worried about and it works fine. 

I missed the EDIS, I thought we were talking MegaJolt which just needs a multiple of half the cylinders for a crank signal.  (I think)

Distributors DO have timing issues, that's why when the automakers really began to care about getting precise timing control, they stopped triggering off of them.  SBCs are some of the worst because of the small ropey camshaft and the way the distributor/oil pump drive is at the wrong end of the engine.  

72-1-1 running on my Porsche where the distributor once lived  

stationary 

This was the second or third iteration I went to a smaller trigger wheel that was still 72 tooth to make it more compact.  The sensor had no issues with the resolution   

In reply to Pete. (l33t FS) :

It is low tec for sure but it would just work and again it is stupid simple. 

In reply to Nockenwelle :

Yes, MS2 with two coils fired at the same time by each of the 4x BIP373 (looks like it has been superseded by a new part). No issues with wasted spark up to 7500rpm using D585 coils (used GM takeoffs). Nothing fancy on the dwell settings, just the recommended values for those per the msextra documentation.

In reply to dean1484 :

If you could put it on the front of the cam drive (like an LT1) then it'd work. Like a Ford trigger wheel mounted to the cam sprocket and have two separate pickups mounted in the timing cover.

The issue with the SBC is the oil pump and camshaft harmonics that get introduced to the party with the location of the SBC's distributor, and the gear lash in the camshaft drive.  It can make the pattern of closely spaced teeth get a bit "fuzzy" if mounted in the distributor.

People have had decent luck with smaller tooth counts, but then this wouldn't be EDIS friendly.  You'd need two EDIS for a V8 anyway, no?

Yes I think that is how they put edis on a v8. Two EDIS v4 coil packs but I could be wrong.  It has been a while since I messed with it. 

V8 EDIS had its own module and two four cylinder coil packs.

I wouldn't mess with EDIS at this point; the supply of Ford modules is pretty thin on the ground. A MicroSquirt firing logic level coils or an external ignition module can accomplish much the same thing.

I have a plan. First a couple clarifications: 

My original statement of option 2 was built around the MSD setup. v2.0 still has the major downside of relying on the mechanical distributor for trigger/timing, and there's no debate that's not ideal. Triggering off the crank is always going to be the most accurate option, though that's a pretty clever setup Dean did on the 924. The situation for a Chevy V8 is not quite as dire as Pete makes it sound since the load from the oil pump keeps the gear lash loaded-out most of the time, but totally agree on the sources of error/slop in general. So I should have included an option #3 or at least 2.1: full distributorless crank-triggered setup, either waste-spark or individual coil.

I also forgot that LS coils are logic-level, so ganging together a pair and triggering them with a single input is perfectly fine as long as the driver can support the current draw of both. The BIP373 (or later ONSEMI IGBT that has superseded it) in the MS2 is a high-current driver for "dumb" coils and totally unnecessary for logic-level coils, but it doesn't mind or hurt at all. Still trying to find the hard limit on the MS/uS logic-level outputs just to prove it will be fine: uS does not have high-current ignition drivers.

For the record, EDIS only works with a 36-1 trigger wheel (or a 72-1-1 on the cam) AND a specific offset from missing tooth to TDC. Therefore Megajolt as an EDIS-only setup is likewise limited. Not insurmountable, but certainly a consideration for any Megajolt install.

APEowner suggested a shrink fit for a steel trigger wheel on an aluminum pulley. That's a better idea than the press-fit I already considered, but two cons here: I'd need to machine a precision ID on the trigger wheel (as well as OD on the pulley, but as a machined part that would be expected), and I'm concerned that the residual forces would dish the trigger wheel like a Belleville spring and maybe introduce (uneven) axial runout. I really want a more robust connection since a loose or warped or even MIA trigger wheel at speed & far from home would be catastrophic. 

I found out that MSD has a pretty solid repair option for their distributors. In my case, they just replace the whole main shaft rather than dickering with replacement of the center cam. Total for my rebuild including new module and vacuum can (which were both fine, they just replace automatically), main shaft & advance, and pickup coil was $194 and it's on it's way back now, so at least I have the baseline config as a fallback.

So the plan. I bought a good-used GM pulley off ebay. I've got a 36-1 trigger wheel on order from DIYAutotune. I don't currently have access to a lathe, but I'm going to get the wheel bored out to fit around the ~1" tall "root" of the pulley with a very close fit. Fixture the pulley and wheel carefully then TIG them together gently to avoid warping. Adding a Hall or VR sensor is trivial, lots of room and bolt holes to work with. No impact to existing pulley offsets or anything else belt drive. That should solve the pulley and trigger topics.

I bought a set of good-used D585 coils off ebay with harnesses and brackets for $120. I have like 5 loose coils in my stash already, so I'm covered for spares and the potential for a bad one in this new set (expected to be unlikely). They're the yellow/clear epoxy version with 3rd-gen heatsinks. Will mount the 4-packs on the inner fenders on both sides of the motor allowing short and equal plug wires NOT running to the back of the motor and also avoids the wire knot running to traditional 4x waste-spark packs.

I'll be buying a Microsquirt for this, running spark-only with the coils triggered in waste-spark mode--no cam signal needed. The setup/config/compatibility with the rest of this setup as described is well-proven and routine. The uS will be capable of running batch-fire on this in a future project, which is fine as I have no need for full sequential on this build. The existing Microsquirt running the trans needs a tach signal, currently supplied by the HEI module but it can accept a tach signal from the ECU over CAN instead, so that's handled. EFI source has one available for $370 with an 8' harness. I'll probably mount it on the firewall right behind where the cap used to be since it's a protected & relatively cool location and also something of a statement. 

The distributor will have no function other than driving the oil pump and plugging the galley, as desired. I'll probably put the freshly rebuilt MSD unit on the shelf and toss in an old GM unit, stripped. I'll cut the old cap off about 1" up and glue a cover on it. Still leaves the option to cut 7 teeth off the mag pickup for a cam signal, but no need unless I go full sequential with a different ECU. I don't think I've ever seen a distributorless BBC in the wild. Will be weird to see one running "without" a distributor or plug wires but still sporting a big Holley; I'll have to post some pics.