C300CA

I tried regular and premium gas only my new 94 Altima, 96 Maxima, 00 RX300 and 02 QX4, no difference in acceleration, MPG, idle smoothness, etc (I probably would if some type of meter was used). so I used regular.

When I took my new C300 delivery, sales person put in regular gas just to save a few $. After that I always put primium. I don't even want to compare regular vs premium.

Samaral

+1

speedgoat

I've been running both my Mercedes on regular 87 octane only for over a year now. I guess I'll report back here if they blow up but they have been running great so far. I was not able to tell any difference between 87 and 91 octane.

aamirani

I don't see why you would buy a car that you wouldn't be able to afford to put the required gas in it... Why buy a Mercedes if you can't even afford to spend a few extra bucks on premium gas?

speedgoat

For me I just didnt see any benefit with 91 vs 87. Why spend even a dollar more on gas if there is no benefit. I also have a 99 Camaro with engine modications that is tunned to run on 91. It detonates with 87. On that car I gladly pay for Shell premium 91 octane. The mercedes get the cheapest Arco 87 I can find cause they run the same on all gas I have tried.

If your car runs that much better on premium then I'm sure the extra cost is worth it to you.

TexFinn

I don't think it's a matter of affording the few extra pennies for anybody here.

I know for a fact that it's not that for my wife. It is due to her at some point seeing on TV or Consumer Reports or what ever something about that you do not absolutely have to use it and what ever blahblahblah. Since then she has stuck to her opinion and since there is no way to change her mind, I have given up.

Not thatI completely disagree with her, either, but I still use premium when ever I get to fill up her car. Or either of our daughters' cars (01 330i and 05 C230K).

Glyn M Ruck

I remind people what I said in post #11

OP ~ All you really need to know is that using regular fuel will keep the timing of your engine retarded as the knock sensor detects the onset of knock. This will adversely effect engine efficiency & consumption.

kevink2

Glyn, I have a data point for another engine timed by OEM for 87 AKI gas, that suffered greaty when I tried 93 gas, and hoped you could shed some light on this, as octane effects have been discussed here.

It was an 83 Saab Turbo that had a step change of boost retard of about 6-8 degrees that was dumped in at 3 psi, and it was like throwing out a boat anchor ... the retard killed power and elongated boost rise.

If I tried using 87 octane, vs the specified 91, it made the pre-boost and boost rise much more frisky, but would knock and reduce boost at half max boost. But when I ran 93 octane, it just added lag to the boost rise, and mad off boost performance sluggish. I could only assume that the 93 octane slowed the burn rate which delayed the compression peak off the optimum 10-15 deg ATDC. Hope you have some thoughts.

.

speedgoat

This is a good example of the way things used to be. I provided a very similar example with my 99 Camaro above. Your Saab was specifically tuned to run 87, my Camaro for 91, and they have very little ability to adjust for anything different.

With newer cars that is no longer the case. They seem to run fine on whatever gas you put in them.

I think that eventually we will see more OEM's doing what Ford has started doing. The 2013 Focus ST with the turbo 4 engine is rated at 252 hp using premium gas or 244 on regular. Ford has pretty much left it up to the owner to choose if 8 horsepower is worth an extra $0.30 per gallon. To me it would not be as I would never feel the difference 8 hp makes. To others in this tread it obviously would be worth it.

I just don’t agree with everyone who thinks that my Mercedes will automatically crash and burn because I use regular...

Glyn M Ruck

Kevin. You have partially asked this question before & I have explained the fuel blend potential influence or not. Octane boosters like toluene delay flame front propogation but the nasties used by the fuel companies don't to any degree & neither does lead. (TEL, TML)

I'm going to have to give this some thought because I don't know what the ECU/Turbo boost controller were doing.

Knock sensor out ~ Did the standard advance curve look like a two stroke with a big retard dip in the middle of it? i.e. initial advance at low revs going into retard at medium RPM & back to advance at high RPM.

What was the rev range & at what RPM do you estimate the knock sensor was detecting knock. I presume we are talking high speed knock here? e.g. 4000RPM

To the best of your knowledge was the ECU programmed to reduce boost & retard timing at the onset of knock?

It sounds to me as if the standard EPROM map had an advance curve poorly matched to the fuel being used.

Give me some answers & I will scratch my head. I have minor knowledge of Saab.

kevink2

I know you have an old Jag (wish I kept my scca prep'd Triumph) , so this should not sound foreign to you. The '83 predated ignition maps. It was centrifugal advance weights, and vacuum & boost/retard combo dashpot on the distributor. You had max centrif advance for the curve at 3000 rpm. Manifold vacuum would increase advance at idle and light load cruising. When cruising, if the manifold vacuum dropped to zero, you lost any vac advance you had. If you started boosting the manifold, the boost retard would fully kick in at about 3 psi.

The knocks sensor worked with saab's pattented APC boost control system, which is now found in most oem car turbos, where you have a Duty Cycle controlled dump of the boost signal in steps. The triggered knock sensor would only drop boost 1-2 psi, but if another knock occured in a time window, even more boost was dropped the next increment. This system was also used on the Masarati Biturbo, as I discovered when working on one.

The rpm range limit for knock control may have been limited to about 3000 - 3500 rpm. Back then they had trouble distinguishing knock from noise at high rpms.

As I mentioned before, the knock sensor worked with the boost conrol system, to reduce knock by backing off boost.

With this electro-mechanical system, it was sluggish but ok with 91 octane. I cured the lag by changing the boost retard "boat anchor" by using a MSD 6A CD ignition for more general pep, with an MDS Boost Timing Master. Rather than dumping all retard at 3 psi, the BTM allowed retard proportional to boost, with adjustable gain. I also used 93 octane, and increaded the basic timing advance by 2-3 degrees. The result was no lag, it drove like a torquey 3L six, and I damaged bearings in the tranaxle from all the torque it produced.

Thanks for your time Glyn, and your patience OP.

.

Glyn M Ruck

Kevin,
I have had a brief look at the Saab APC system. Standard vacuum/centrifugal distributor. Old style inaccurate Knock sensing at high RPM. Wastegate dump arrangement at the onset of knock at low RPM & boost reduction at high RPM where knock sensor is inaccurate. While it might have been innovative for the day the boost control was poor & the ignition timing always less than ideal. A workaround at the end of the day & the driving force for the more advanced DI/APC.

It is difficult to know what the system was doing at any specific time so I'm not going to guess. The timing was almost always compromised. The boost was controlled by RPM, manifold pressure & knock inputs. The boost was tailored to reduce at high RPM due to poor knock sensing. Lower CR exacerbates turbo lag.

This system was designed to protect the engine rather than achieve optimum performance.

To achieve optimum performance one would require to run the highest octane fuel available & advance static timing to suit while accepting that the boost controller was working against you at over 4500RPM.

With modern electronics DI/APC & way beyond that ~ things are a lot easier.

~ Optimised timing map for fuel used
~ Accurate knock sensing (the sort of thing Benz used in the Diesotto)
~ Optimised timing retard profile at the onset of knock
~ Optimised boost dumping in conjunction with timing retard
~ Optimised boost right through the rev range without high RPM profiling to compensate for poor knock sensing.

The old system was simply too crude for the job at hand but this was in fairness 1983.

Hope this helps.

glj12

I humbly suggest that this thread becomes a sticky, somewhere, given that people ask this question fairly often around here. Agreed? Glyn gave a very pleasant tl;dr of MB's post. Thanks, guys.

kevink2

Thanks alot Glen for your efforts, but I think you missed the question, and I'm sorry that I was not clear about the question.

In my post prior to your reply, I gave pretty much the same info about how the APC system, mech'l advance and retard worked together. I had reverse engineered the whole system, meaning I had several analog voltmeters in the cockpit to observe what duty cycles were, and the effect of adjusting potentiometers on the APC control box. For that saab forum, I was an expert in this area, at the time, and the MSD-BTM that I mentioned, introduced to replace the "step change" pressure retard, turned the car into a lagless monster.

The question was, forgetting the boost retard system, octane level changed the "little to no boost" early acceleration from a stop. With low octane this phase of acceleration was best, with highest octane it was the worst. I think normal fuel was 91 AKI. My thoughts were about flame propagation and whether the low octane was creating optimum peak cylinder pressure (peaking about 10 deg ATDC), and better than 93 AKI in this specific condition.

.

Glyn M Ruck

Kevin. The whole problem here is that you don't know anything about the fuel you were using in your testing other than octane rating. You don't know how the octane rating was met by those fuels either. What octane boosters were used etc. Some octane boosters like toluene & similar compounds do slow flame front propagation. Most modern octane boosters have little influence on flame front propagation.

If you had used reference fuels of known blend, density, distillation curve, octane booster type & treat level etc etc we could probably make some deductions.

In your case with unknown 87 & 91 AKI? fuels the results you saw might have been due to your particular timing set up & rate of flame front propagation of the fuel you happened to pick up from some gas station. We will never know. The horse has bolted.

What I'm saying is that you could get identical flame front propagation 87 & 91 Octane fuel from a gas station today.

So I suspect that your timing was not ideal for the 91AKI fuel you used hence what you saw. This might have been due to flame front propagation. But I'm guessing. It's one plausible answer with what was a pretty crude system on the old 83 Saab.

Did you ever play with more advanced ignition timing and see if the situation reversed between the two fuels?

kevink2

Glyn,
Thanks for the reply. I can say that I always went to the same gas station, so that variable is nixed. And I repeated it a few times, all with the oem system set to OEM spec'ns.

For my "optimized" set-up, I did add about 3-4 degrees basic advance to the timing, ran exclusively 93 AKI fuel, eliminated the awkward boost retard and went with the MSD-6A CDI with the MSD BTM for the linear addition of retard with boost (with adjustable gain).

As modified, it would outrun the later SPG 16V Turbo in 50-70mph top gear runs. But my project to maximize low end torque cost be a transaxle


Do you mean from the same station at the same time?

I agree the oem timing was not optimised for the 91 octane, in this narrow part of initial acceleration where there was no protection from the knock sensors, with little to no boost to control with.

Yes, a crude OEM system, but focusing on a defined interval during acceleration, before enough boost to see any knock response.

.

Glyn M Ruck

Yes.

B5chung

So how do you explain this finding.
http://www.cbc.ca/news/canada/story/...emium-gas.html

wildchild

Any thoughts about using octane booster? where i live in NYC premium is as much as 30-40 cents more per gallon, but i can get a bottle of octane booster that will treat my whole tank for @3 bucks, plus put me over the 91 octane of premium. if my tank is 17.4 gallons i would be saving around 7 bucks on reg vs premium then subtract 3 or 4 bucks for a bottle of octane booster. This should still be saving me a couple bucks and be well over the 91 octane so it should be a win/win...

perrina

As far as explaining the differences in compression ratio... the compression ratio is just ratio of volume in the cylinder from piston fully down to piston at tdc although this is a start for what grade of fuel it can be deceiving 10.7 is not very high there are engines with higher ratios that recommend regular fuel. This is because the compression ratio is necessarily correlated to the amount of pressure in the cylinder at tdc. another example is turbo engines normally run a lower compression ratio but forces more air in to the cylinder causing high pressure and temperature and greater chance of detonation.
For the original scenario the mercades benz will have better flow in and out of the cylinder so all though the amount of mechanical compression is the same you will start with more air and end up with a higher pressure.

Glyn M Ruck

Caveat ~ don't confuse volumetric efficiency with compression ratio.

Mikephinney

I thought it was interesting to hear that the Test Accord had lower numbers when filled with premium fuel. My C250 is a weekend car which I choose to fill with premium; my '01 civic on the other hand has 340,000 miles on the original engine and MT tranny. I run premium through it every three tanks or so just to clean things out, when I do I notice that the fuel economy is quite a bit less. I average 38 MPG on regular 87 and about 33-34 on premium. THe dealer did fill my C with regular once and I did notice it to be a tad more sluggish.

Glyn M Ruck

I suspect what we are seeing here are fuels with & without alcohol or summer vs. winter grades otherwise it makes no sense.