I have a Honda Civic with a naturally-aspirated Honda D17A2 (9.9:1 compression ratio).


I want a modest horsepower gain (say +38 hp) in the lower-mid-range powerband; I'm not after the maximum horsepower potential. My goal is better overall engine efficiency.

Initial Thought:

I initially thought there should be a way to bolt a small, aftermarket turbo onto my engine without having to switch from Regular (86-89 AKI) unleaded fuel to Midgrade (88-91 AKI) or Premium (91+ AKI).


I stumbled across a previous question (by Annath) and answer (by Bob Cross) that gas me questioning my initial thought. It was my understanding that:

  • There are numerous factors that can affect the maximum allowable effective compression ratio (ECR) in relation to the fuel octane rating.
  • It's best to have an ECR set as high as feasible without encountering detonation at the maximum load condition.

New Thoughts:

I began to think that the D17A2's 9.9:1 compression ratio might not be the immediate drawback most would think it to be in regards to forced induction and the 87 AKI fuel.

The factors that I'm aware of that influence the compression ratio include:

  • fuel anti-knock properties (fuel octane rating)
  • boost pressure(psi)
  • intake air temperature
  • combustion chamber design
  • ignition timing
  • valve events
  • exhaust backpressure

So, if you can lessen or modify one or more of the other factors in an attempt to balance the increase from chamber pressure, then perhaps a safe offset will occur.

Thanks to many modern normally-aspirated engines having well-designed combustion chambers and ECUs (or PCMs) providing appropriate self-tuning, I thought my D17A2 engine could handle a modest amount of boost with no significant change to its compression ratio. Am I wrong in thinking this? By making sure that the turbo I installed was small, wasn't set above say... 5 psi at max, had an intercooler installed for decreasing the intake air temperature, I believed my car's ECU could enrich the fuel/air mixture so that it would run even cooler yet, plus modify the ignition timing all in order to prevent premature detonation and provide a modest horsepower gain. Plus, thanks to the size of the turbo the power gain would affect the power band between 2000 - 4000 rpms, where most of my daily driving takes place. On top of that I thought I might even see a slight increase to the vehicle's fuel efficiency as the engine's cylinder's volumetric efficiency increased thanks to the forced induction. Am I overlooking something or is this a practical outlook to the potential of adding a turbo to my vehicle while maintaining the use of 87 AKI octane fuel? Do you think my Civic's ECU could handle the changes that would come with installing a turbo as I described or would I need to have the ECU flashed and re-tuned... my fear is this would be a potentially disastrous step unless under taken by a true specialist who understood the installed turbo and the D17A2 engine. I worry that this would mean that it would be very expensive, too expensive, to get the tuning just right, but maybe the default ECU setting would be sufficient. Anyways, what's your thoughts and is there a way to mathematically determine whether these aspirations of mine are even feasible?

4 Answers 4


I can get you halfway; here's the formula for the effective compression ratio of a forced induction motor:

Effective Compression Ratio (ECR) = sqrt((boost+14.7)/14.7) * Static Compression Ratio (SCR)

So, for 5psi on your 9.9:1 SCR motor:

ECR = sqrt(19.7/14.7)*9.9 = 11.46:1

However, I have had quite a difficulty trying to find a good way to get ECR to "what octane gasoline do I need", as it seems no simple matter. Some starting reading is here and here.

Off the cuff, 9.9:1 seems rather high even for 87 octane, and 11:46:1 seems to be square into the 93 octane range, which I've been told has a usual maximum of 12:1 compression (though again, no simple equation for it). It seems some motorcycles can get away with 14:1 on premium, but this seems not to apply for most cars; it probably has to do with the smaller piston size.

As a reference, most turbocharged vehicles use closer to an 8:1 static compression ratio; my Probe turbo used 7.8:1 and I could only get about 21psi on it with pump gas (premium 93 octane, 12.15 ECR). If I wanted to use 87 octane and not trip the knock sensors, I had to limit the boost to about 6-7psi (9.36 ECR).

  • "I have had quite a difficulty trying to find a good way to get ECR to 'what octane gasoline do I need', as it seems no simple matter." - Ehryk Right, which is one of the reasons why I posted my question. I've read elsewhere (sorry can't find the article I previously read) that said that engines running high compression (over 11:1) have been known to get away with running on Regular unleaded fuel and that they were just more efficient for doing so instead of running a Premium fuel.
    – Luke
    Jun 26, 2012 at 13:46
  • I'd say that's more the exception than the rule. faqs.org/faqs/autos/gasoline-faq/part3/section-1.html#b details most of the factors that affect your octane necessity, and I'd also say if you did this anyway that some form of water injection would be your best 'bang for the buck' to lower your octane requirement. I've 'heard of' 14:1 on motorcycles as well (which usually require premium), but I think there are other factors that make it possible on those setups.
    – Ehryk
    Jun 26, 2012 at 18:12
  • If you can pull it off, more power to you, and post details; you can also log your knock sensor activity and see how close to the line you can push it if you do. However, I don't see anywhere that would give you the 'ok' beforehand, because 9.9:1 is above what some cars can handle on 87 octane already. Also, what if you do it and the limit of 87 octane is 2 psi with so much timing pulled you're making only 5 extra hp? Would it have been worth it?
    – Ehryk
    Jun 26, 2012 at 18:15
  • My car's engine has 10.9:1 CR, and the manual makes a big point of saying that the use of gasoline with an octane rating below 91 is strongly discouraged.
    – TMN
    Jun 29, 2012 at 19:46

I have a 9.1:1 compression car that I normally boost to 16 psi as the intercooler maxes out there on hot days. We've run it up to 18 psi no problem on a normal day. At 21 psi the intercooler is totally saturated in a hurry and the ECU starts to flatline the timing due to knock. I've also got a 9.6:1 NA car that knocks on anything less than premium (apparently due to the 10% ethanol fuel we have, runs fine on regular on the rare occasions I can get pure gasoline)... So, in short, there's a lot of variables, too many to truly quantify. It's really an experimental process that must be undertaken with great care and the realization that you're increasing the liklihood of breaking something, possibly something quite expensive. If you can find someone that's already done it with your exact make/model/configuration, that's a good starting point (but without the normal safety margins, even basic wear and tear can make the car react differently from what would otherwise be an "identical" car)! :-)


You're safe with 5 psi and 9.9:1 on that car except for when:

  • You put the pedal to the floor with full boost at less than 3000 rpm
  • Go up a hill in a higher gear
  • Drive around with 3-4 other people in the car
  • Tow.

In any of those cases, use the highest octane you can find.

I run 9.5:1 static compression with the same dish piston as your D17. I run 87 octane up to 8 psi and get no knocking. My 1.7L at 8 psi makes about 220 bhp.

I have run 85 octane above 5000' elevation. The car can boost at 4 psi up long hills with no overheating.

  • Just to compare these numbers with the ECR figures, above - 9.9:1@5psi = 11.46:1 ECR - 9.5:1@8psi = 11.8:1 ECR.
    – Ehryk
    Oct 27, 2016 at 6:28

I think you'll find that the fuel/air ratios are going to be outside of the range your stock ECU can compensate for. Your best bet will be to find a bolt-on turbo kit that has either a chip or a reflashed ECU. The companies that make the kits spend weeks getting the mapping right, I doubt you'll find anyone willing to develop a one-off for your specific engine/turbo combination for a couple hundred bucks. Plus, rolling your own will mean fabricating your own brackets and plumbing, are you comfortable with that? Not to rain on your parade, but bolting on a turbo is a lot more involved than most simple speed mods. Have you looked into any other options? Maybe a low-restriction cat and a chip?

  • Already have done some performance upgrades. Exhaust / Intake. Nothing to the chip or anything of the like. I'm beginning to agree that it's more trouble then I'm willing to get into, but I was still curious as to find a mathematical formula which could determine the plausibility of such a task.
    – Luke
    Jul 2, 2012 at 13:30
  • 1
    @Luke Ultimately that's going to require some complex math and engineering. pV=nRT is the core equation you're going to need. Along with some accurate intake and post turbo temperature readings. Bob Cross has an excellent writeup linked in your question. There isn't a simple accurate rule of thumb, because it's a complex situation.
    – Leliel
    Feb 12, 2016 at 1:37

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .