Discussion and questions related to the course EFI Tuning Fundamentals
My interest is in the 700-3000 rpm range of ignition timing with a desire for fuel efficiency. The vehicle is a 1982 Ford E350-based 11,000 pound 24 ft. motorhome with no possibility of dyno tuning in Nevada. The engine is the classic Ford 7.5L 2-valve per cylinder pushrod hydraulic tappet engine with long-tube headers feeding a single 3 inch exhaust, an RV towing cam and cam timing set to 2 degrees advanced. Most engine operation is between 1500 and 3500 rpm. For some time I have been confused about the purpose of the two stage mechanical advance curve used on pre-EFI big block engines. Andre's discussion of the interplay of spark advance timing to compensate for pressure wave front delay vs. rpm and the variation of optimal timing with load and volumetric efficiency was the first credible discussion that suggests a low-to-mid rpm non-linear timing advance curve. In my road trials (flat Nevada Desert and GPS based timing trials), at wide open throttle, the engine likes 32 degrees advance at 3,000 rpm and above. From engineering data from the Ford Laboratories, the BSFC curve shows a minimum at 1900 rpm. The factory engine torque curve is flat from 1500 to 2400 rpm. Can anyone suggest a two step (slope) starting timing curve for this application?
You may be conflating two quite different aspects of ignition timing.
Your distributor is using a double acting vacuum purely for emission purposes, most likely to reduce oxides of nitrogen and/or maintain 'converter heat, and will indeed be killing advance as under some circumstances there is a vacuum applied to the 'back' of the diaphram to pull timing and, as you say, overall engine efficiency is reduced.
You should be able to source a distributor spring and weight kit for your engine, and there will probably be a range of vacuum canisters for different versions of the distributor for different applications.
Initially, I suspect you're going to have to go 'old school' - figure out a good manifold vacuum (might work with a ported source on the carb' if available) and use that on the advance side of the canister, or at least disconnnect the retard hose and plug it.
For timing, you will probably be looking at a total of around 30-35 degrees total mechanical - you suggest 32 degrees - and 10-15 idle, both with vacuum disconnected, and as much in between as practical - but this is something that you will need to experiment with. Then use a vacuum canister that gives as much vacuum advance on top of that as possible without pinking.
With a motor home, especially one that's rear engined, heat is always going to be a problem and the more you can do to reduce the heat in the engine compartment the better, and spending some time and effort to ensure the engine is getting as cold air as possible to the carb' will also help - you may need to feed from a warm area under the engine cover in your winter, but when it's above freezing, cold air is your friend.
Thanks for your reply. I wasn't being adequately specific. I'm not referring to vacuum advance contribution. I'm referring to the fact that "old school" centrifugal advance mechanism distributors (with vacuum advance canister) have two different advance rates. The first running from idle to about 2000 rpm which advances quickly at about 1 degree per 100 rpm and a second slower rate operating from 2000 to 3000 at which point it is "all in". The two different ramps are chosen by a combination of two springs and two weights. In my case for a mechanical distributor, the base timing at idle (650 rpm) is often 10 degrees BTDC, another 17 degrees is added from 650 to 2000 and 5 degrees is added from 2000 to 3000 rpm. The mechanical vacuum advance canister adds up to an additional 12 degrees at 15 in. manifold vacuum.
What I was hoping for was an idea about what rate of advance to add to the idle base timing up to an inflection point (perhaps related to the start of the torque max or the BSFC minimum - which I assume is the maximum volumetric efficiency) and then what rate of timing advance to add up to 3000 rpm. I had also hoped for some advice about potentially reducing the total and/or no load timing above 3000 rpm driven by an understanding of the large bore (4.36 in diameter), the low compression ratio (8.2:1) and the almost complete lack of quench in this stock (except for cam and headers) emission design engine. BTW, I am adding, at this time, a programmable ignition with a Crane Hi-6S multispark inductive ignition and after debugging this system with the Edelbrock 1406 carburetor, I will be adding a FiTech throttle body fuel injection system.
I got that - I'm old, and have spent many hours, back in the day, with a dissy machine and a selection of weights, springs, bushings, canisters, etc. - even some welding to limit the travel. However, that was after testing with manually adjusting a locked dissy on a dyno', so the required curve could be mapped out - on paper, none of this modern electronic stuff - so I knew what I was aiming for.
Some will use a mixture of parts for each side of the mechanical advance, to try and fine tune it, but that can cause uneven wear and even sticking. There is also the shape of the cams the weights act on to consider.
However, that doesn't really address your query, and as it is very unlikely someone is going to have exactly the same engine setup as you, you're going to have to do some experimentation for your particular installation.
I specifically mentioned the mechanical side first, as that will apply under load and when there is zero vacuum advance - I believe you got that - but the vacuum canister, and how much timing it adds, is CRITICAL for light loads/part throttle operation as it adds the timing required for the less dense charge which burns slower. Different canisters will advance different amounts under different amounts of manifold vacuum.
One option, which might be worth considering, is to convert to a mappable 3D ignition, as that will allow you to more easily tailor the timing to suit, especially with some means of monitoring for knock and instantanious fuel consumption.
From your description, I believe you have the 460 CID version of the 385/LIMA engine? Hmmm, I know it's been quite a while, but I would have thought the dissy kits would still be available, but seems not? Might need to change to an aftermarket dissy as they should still have weights, springs, etc. available and some have adjustable vacuum canisters (IIRC, I used one on an ACCEL) , so you can just tune it rather than trying several.
if you haven't already done so, I would suggest asking on some of the forums/clubs more focussed on that vehicle/engine, such as www.460ford.com where at least one chap has a similar enquiry - https://www.460ford.com/threads/autolite-mechanical-advance-curve.151798/
Ah, changed search terms - https://www.summitracing.com/search/make/ford/part-type/distributor-advance-kits
Good tutorial, if you missed it, here - https://www.racingjunk.com/news/performance-tuning-your-distributor/ - might seem granny-eggs, but worth re-visiting.
Gord, good find on the tutorial, and I'm very familiar with 460Ford as I have posted there often, read the group at least weekly. For some time I have been trying to understand the classic two-stage mechanical advance curves. The linear advance curve seems intuitive given the purportedly fixed combustion rate and the need sync the peak pressure to the best mechanical advantage point in the piston travel cycle past TDC. However, many experimental papers demonstrate the complexity of the flame front with AFR ratio, fuel droplet size, turbulent inlet flow and initial pressure just prior to the ignition point. Part of this is mapped by the ignition delay after spark generation and much of this is modified by the recent advances in combustion chamber design. I am locked in with a combustion chamber degraded from its initial competitive quench design, by the frantic smog emission mandate triggered chamber redesign which dropped the compression ratio by lowering the piston face relative to the head essentially nullifying quench.
My main problem is the heavy load presented by pushing the 11,000 pound motorhome (front engine, C6 automatic with Gear Vendors overdrive and rear drive) up and down hills and mountains, not to mention air resistance and drag on the level salted plain! This motivated my use of an RV cam (Melling MTF-3), better exhaust flow with the long tube narrow pipe headers and 3" high flow diesel exhaust. The carburetor is a Edelbrock 1406 that I had retuned by a carb expert to support the details of my use and application. I intend to carefully monitor the carburetor AFR response to operating conditions and use that data to inform my interpretation of the self-tuning of the Fitech 400 fuel injection system
To clarify the ignition issue, I am moving from the current mechanical Duraspark II distributor (mechanical with vacuum advance) to a 2D programmable ignition which maps manifold vacuum for load and engine rpm. The ignition map can be tuned in variable increments, and our current discussion is to help me design an ignition map for this ignition system.
Going back to the linear vs. two stage advance curve issue, Andre's discussion in the webinar began with the linear-to-about-3500 rpm analysis of constant flame front velocity vs. time to 12 degrees after TDC. But then he added the discussion about flame front velocity modifications with cylinder filling given by volumetric efficiency. He showed the flame front propagation velocity reaching a minimum at the point of maximum volumetric efficiency and then commented that optimum spark advance curve shows a dip in this rpm region and the optimal curve departs substantially from the simple linear relationship.
I'm trying to get an understanding of the value of the inflection point and the slope or rate of change from idle to and after that point. For example in the Autolite distributor discussion you cited, the initial curve was 1.6 degrees per 100 rpm to 2,000 rpm and then 2.4 degrees per 100 thereafter. Many of the stock Big Block distributor curves are more like 2.4 degrees per 100 rpm and then 1.6 degrees per 100 rpm with the inflection point between 1800 and 2400 rpm. After this decision I will try to refine the values with road tuning but I had hoped for a well reasoned starting approach!
Thanks much for all your help!
Ah, so many $10 words, Frank. I think you are WAY overthinking it*, especially as it isn't directly relevant - 'cept for the general rule that advance is added with rpm and as the charge density is reduced - basically the mechanical advance and vacuum advance functions - with the first being most critical for power and the latter for economy.
I am not familiar with such types of 2D mapped distributors where vacuum is used, the 2D ones I am used a simple advance:rpm model and the 3D versions used a TPS (Throttle Position Sensor/Sender) for (implied) load, which was the equivalent of vacuum advance, to further modify the advance.
Again, you are going to have to work out, from practical testing, exactly what is going to be the best for your engine. I would suggest also picking up some form of knock detecting headphones, or other method, to avoid possible sub-audible detonation/pinking which could, over time, damage the engine.
Uh, you DO realise that the '2 stage' curve is just a slightly better way to approximate what is actually a curve than a single 'stage'? Again, not something to overthink.
*You don't have nearly as much information as is needed to try and work it out from theory and, anyway, empirical trumps theoretical.
Gord, You're right, the programmable ignition system I'm switching to is a 3D map with advance, rpm and load as parameters. I saw rpm and advance as variables, hence my suggestion that it was 2D. Fortunately, due to the focus on the 460 Ford engine on 460Ford.com, there is a great deal of empirical information on optimal total timing for this engine in both performance and towing applications. 32 degrees total in by 3000 to 3200 is a documented value for this cast iron head running 7.5 to 8.2 compression ratio with my "towing" application. The general empirical suggestion from the mechanical distributor guys is fast advance to 2000 to 2500 rpm then slow to 3K . Anton's curve from the EFI fundamentals course suggests a timing advance needs to get low rpm timing largely in by the torque max which starts at 1500 rpm and drops at 2400 rpm with BSFC minimum at 1900 rpm. I was just trying to see a dyno curve that really optimized low rpm timing for a Ford or Chevy large bore, low compression towing application.