Programmable MSD 6AL-2 questions

[SSLance]

So, a year later I am revisiting this topic. It was interesting to read back and see what I bailed on last year and attempting again this year with the new engine.

Last year I disconnected the MAP sensor and just ran the run retard graph in the box and it did pretty good all year. With the new engine, and new train of thought...I'm trying to program in a "boost retard" curve once again.

My engine builder concentrates on race engines and we dynoed this engine at 32* locked out 0-6500 RPM. I initially started with the run retard curve in the MSD that I ran last year that pulled 10* out at idle ramping up to all in at 2000 RPM. (22* at idle, 32* 2000+RPM) It ran pretty well but the idle was a bit unstable, most wouldn't have noticed it but while trying to tune the carb it was kind of giving me fits. I finally took the builder's advice and locked in 32* 0-6000 RPM. That cleaned the idle up a lot and I was able to get the carb dialed in nicely. It pulls about 17" of vacuum idling at 850 RPM with an AFR of 14, cruise AFRs are right around 15 and it gets a bit fat under WOT around 12-13 depending on what RPM I start the pull at. It runs very smooth, makes great power, and is safe if I get a bad tank of gas somewhere.

The only issue is it will ping if I lug it down under 2000 RPM in 3rd, 4th or 5th gear. I have to try to make it ping, under normal driving circumstances I can not hear any detonation.

So I have put the MAP sensor back in place and am working on using the Boost Retard function to try to pull some timing out under low vacuum situations.

Here are the dials at idle completely warmed up.



It idles around 5.25 PSI, will drop to 3.5 PSI under full engine compression like downshifting coming to a stop and the highest I've seen is about 13.75 PSI at WOT in 3 or 4th gear. I created this Boost Retard graph and drove it.



It's basically pulling 10* of timing out (32-10=22*) when vacuum drops under WOT pulls and stopped the pinging. The thing is, at WOT the PSI gains to 13.75 and the box stay at 5* retard (27*)...never letting the timing get back up to the 32*s it needs.

I'm thinking about pulling the right dot that is right now at 14.5 to the left and putting it at about 13, that way at just less than WOT the box would put the full timing amount back in, only retarding the timing between 10-13 PSI.

Am I thinking correctly here? I can't seem to find a way to pull timing out at WOT (zero vacuum) at low RPMS, but not pull the same timing out at WOT at higher RPMs where the timing is needed. So my hard head is thinking to just tell the box to pull the timing at just about WOT...

[68Cuda]

I think you have the right idea, you are just at the point where you are going to have to try it and just see what it does.

[SSLance]

I went back to the MSD forums to read around a bit more to make sure I was on the right track and I think half of the time the techs there that answer are completely backward as well.

Trying to compare vacuum advance with vacuum in inches to Boost Retard with instead of vacuum in inches, manifold pressure...makes one's head spin.

What I need to do is decide if it's worth the effort to lock the distributor in at say 45* crank timing and plot both run retard and boost retard graphs to pull enough timing out with either or both to get the timing in perfect spot for all scenarios? I could probably end up with total timing in the 40+* range at idle and cruise RPMS and use the MAP sensor\boost retard to pull timing out under a load and WOT.

Really, for my use...it runs pretty dang good like it is as long as I don't dog it down in higher gears and apply massive amounts of throttle. Seems like it might be safer for me to stay the course instead of trying to reinvent the wheel...

[68Cuda]

Originally Posted by SSLance

Trying to compare vacuum advance with vacuum in inches to Boost Retard with instead of vacuum in inches, manifold pressure...makes one's head spin.

What I need to do is decide if it's worth the effort to lock the distributor in at say 45* crank timing and plot both run retard and boost retard graphs to pull enough timing out with either or both to get the timing in perfect spot for all scenarios? I could probably end up with total timing in the 40+* range at idle and cruise RPMS and use the MAP sensor\boost retard to pull timing out under a load and WOT.

Yeah - kind of confusing... theoretically atmosphere is 14.7", so anything under that is a vacuum. It is just whatever your reference is. You just have to unplug to figure out what your atmospheric pressure is. The crazy part is the fact that your scale is flipped. If 14.7 is atmosphere, that is what you think of as 0 vacuum. The numbers don't really matter in reality, it is all just a reference. Sounds like your "0" is WOT at 13.75, so be it.

OK... two different camps on this. Manifold vacuum and port. Based on the problem you are describing let's try port vacuum, this may give you the effect you want. With port vacuum you should only see vacuum at partial throttle. This will eliminate the extra advance at idle and at full throttle and stabilize the rpm at idle. The "port" you want to plug into should be above the throttle blades and should read 0 vacuum at idle when measured with a vacuum gauge.

I think you are still thinking of the vacuum curve backwards.

OK, two different curves! One for RPM (aka centrifugal advance) and one for the "boost" retard (aka vacuum advance). For the RPM curve let's say you want to go from 8 to 32, all in by 2800 rpm. And, for the "boost" you want 10 degrees advance max on the ported.

Your base timing with no curves would be 8+24+10= 42 degrees (pretty close to the 45 you stated). Remember that everything on the MSD is backwards, it is all retard. So, the RPM curve would start at 24 degrees retard. If your idle is 900, then maybe you start adding timing back in above 1200. Your "curve" would be 24 degrees retard from 0 rpm to 1200, then progress to 0 degrees retard at 2800. The effect is timing would be 8 degrees at idle and at 1200, it would then increase gradually to 32 degrees at 2800 rpm. This is my example... set it for power at what works best for you! I used "8" as my idle timing just as an example.

OK, for your boost curve (connected to port vacuum), if 13.75 is WOT, then you want 10 degrees retard at WOT to give you 32 total above 2800. So, left to right you put 0 retard to about 4 then increase your retard to the full 10 by about 12, leave it at 10 total all the way to the end of the graph on the right. At idle, or with the port vacuum line unplugged even, the 10 degrees retard will be pulled out and the timing should be whatever you want it to be, it should not jump around.

So add these together... say you are running 2800 rpm with the throttle barely cracked and a little downhill, you could then be running as much as 42 degrees advance. No problem under light / no load. As you tip into the throttle and start to load the engine the vacuum will drop and your timing will move to 32. Cruising speeds around 2000 (half way between 1200 and 2800) you would be running 18 advance due to RPM, and if there is no load, as much as an additional 10 which will help mileage.

Attached charts make sense? Boost:

[68Cuda]

RPM "advance":

[SSLance]

Thanks Michael, I really appreciate the time you have taken to help with this. You are one of the few I've found that gets how these boxes work and are willing to jump through the hoops to trick them into doing more.

I like your curves and understand what's going on much better...I think I'll try them to see how they work.

Today I pulled the boost retard back to below 13.75 psi and it worked as far as getting 32* at WOT again, and it helped with the ping...but I could still make it ping at cruise at highway speeds if I tried. I could live with it like this but I'm curious just how much having the extra timing in at cruise and at idle would help.

The using of a port vacuum is also an interesting thought. I actually talked about that being an option with my friend that built my carb last night.

I need to procure my friends timing light once again and charge my laptop up...then see how it works.

[SSLance]

Just to make sure I'm clear, one question. If I set the distributor at 42* with no curves, then put the two curves in you showed above...my timing at idle would be 18* correct? Run retard pulling 24* out at 900 (42-24=18).

If I wanted more timing at idle, I'd do your curves above except raise the run retard at idle to say 10* (42-10=32*). The question then would be what to do with the run retard curve from 1000-2500 RPM?

The Boost Retard curve you posted would pull 10* out when vacuum drops, how much more would we need the Run Retard to pull out at cruise to keep it from pinging is the million dollar question...

Cruise is generally between 1300-1800 RPM just FYI.

[SSLance]

I've been tinkering and picking brains a bit more and I think I've come up with a plan. Please take a look at this and make sure my thinking is straight before I implement.

Distributor is locked out at 45* advance,

Run Retard is set as below,
Idle timing 0-600 RPM -8* (37* actual)
1000-2000 RPM -18* (27* actual)
2800+ RPM -13* (32* actual)




Boost Retard is set as below

Full vacuum
17"-12.4" at idle or cruise -0* (37* idle, 27* 1000-2000 RPM)
9 psi (12.2" vacuum) -1* (26* actual 1000-2000 RPM)
10 psi (10.1" vacuum) -3.2* (23.8* actual 1000-2000 RPM)
11 psi (8.1" vacuum) -5.7* (21.3* actual 1000-2000 RPM)
12 psi (6.1" vacuum) -8.4* (18.6* actual 1000-2000 RPM) (23.6* 2800+ RPM)
13 psi (4" vacuum) -10.9* (16.1* actual 1000-2000 RPM) (21.1* 2800+ RPM)
13.5 psi (3" vacuum, WOT) -12* (15* actual 1000-2000 RPM) (20* 2800+ RPM)




On the idle side, the only thing I can think of is I might need to adjust the Run retard between 800 and 1000 so it remains at a steady state there as my idle fluctuates in that range and a moving timing mark will probably make it fluctuate more.

Maybe something more like this. Is that too sharp of a transition between 1000 and 1200 RPM?



On the WOT side, it appears that I need to make the Run Retard more like this so I can see full 32* timing 2800+ RPM



12 psi (6.1" vacuum) -8.4* (18.6* actual 1000-2000 RPM) (33.6* 2800+ RPM)
13 psi (4" vacuum) -10.9* (16.1* actual 1000-2000 RPM) (31.1* 2800+ RPM)
13.5 psi (3" vacuum, WOT) -12* (15* actual 1000-2000 RPM) (30* 2800+ RPM)


I'm not sure how to get those last three lines closer to the desired 32* timing at WOT past 2800 RPM? The variable is vacuum inches, not RPM...so it's almost back to my other method of stopping the Boost Retard right before WOT.

[68Cuda]

Lance, I would leave the RPM based curve flat in the idle area and just do a straight ramp from 1500 to 3000. Why do you want so much timing at idle? It does not do much. Factory these cars usually ran 10 or less at idle with the vacuum advance unplugged.

As far as the pinging, I would tweak one curve or the other so it gives you a little bit less timing. Your boost curve might be a little too aggressive, try pulling it a little to the left. For example set it to the full 12 degrees from 10psi to 15psi and start the ramp down a little earlier, maybe 6psi instead of 8. In other words, at 6psi it would be 0 and then ramp down to 12 degrees by 10 psi. Try that first.

There are so many variables in the way the engine is built, load, the quench, the camshaft, air density (temperature, pressure, humidity) that no one answer will work you just have to tweak it until you are happy with it. On my 440 when I put the mechanical advance MSD distributor I set it by ear and drove it. Later in the week we put a light on it and it measured 35 total. I have been running big block Mopars since the 80s, I know what they are supposed to sound like...