|Home||Before you Buy||Understanding the RX-8||Maintaining your RX-8||Troubleshooting||Aftermarket Mods||Feedback and Support|
|The Engine Failure Story|
|Designed to Burn Oil|
|The Flooding Myth|
|Engine Core Warranty|
|Is an RX-8 right for ME?|
|Mandatory Compression Test|
|Year and Trim Variants|
Yes. And it can not be done with a standard analog compression tester for piston engines. A typical cheap analog tester just plugs into a spark plug port and as the engine cranks, the needle moves, and you can easily see where it's swinging to every compression stroke. 1 spark plug, 1 cylinder, 1 compression number.
The rotary however is 1 spark plug, 3 rotor cavities, 3 compression numbers per revolution. There is no way for a simple analog tester to record each face of the rotor independently as it spins, updating each number correctly as it compresses.
Piston engines also have relatively the same compression number regardless of how fast the engine is pumping. If there is a loss of compression due to a seal or whatever, there will be some loss as it compresses, and a slower RPM will read slightly less as it allows slightly more time to leak. Rotary engines have plenty of 'gaps', and there is a very real and measurable difference in compression depending on engine cranking speed. The faster it spins the higher the compression. You can even remove the apex seals completely, and it gets enough compression at 9,000rpm to stay alive, though not much below that it won't. (incidentally, this is an often over-looked dynamic of how the engine produces it's power curve, and how it doesn't have severe drop-off in power the higher you go in RPM the way piston engines do, and one of the reasons low rpm torque is so poor. Not the only reason, but a contributing factor). So, in addition to the 2 sets of 3 numbers for each of the 3 faces of each of the 2 rotors, you need the crank RPM for the test. A normalization chart based on RPM determines if you are pass or fail.
Many dealers also obtain a BARO voltage reading test to determine vacuum as another method of determining pass or fail. Vacuum is related to compression, but it isn't the same thing, and is possible for them to differ more than expected. There have been quite a few reports of Mazda techs returning BARO numbers that clearly fail and stating that they pass, so if it fails compression but passes vacuum, I'd question the techs quite a bit about making sure they did it right.
Only Mazda dealers will be able to do a compression test, unless you happen to live near one of the bare handful of non-Mazda owned rotary compression testers. A few RX7Club and RX8Club members have them, and most of the small number of rotary specific shops have them. Do some digging to see if there is one near by, but there likely won't be. In those cases, a dealer is going to be your only option.
If the RX-8 you are looking to purchase is at a non-Mazda dealer, your best bet is to simply ask the dealer if you can take it to the local Mazda dealer to get a compression test. If they decline, they ask if they can do a proper compression test. Be sure to be clear that an analog tester won't provide the individual rotor face numbers or the rpm of the test so it can be evaluated correctly. If they still decline, then they probably have something to hide and you should immediately cross that 8 off your list, no matter how much surface appeal it has.
No reputable dealer will object to having one of their cars checked over by a 3rd party, ESPECIALLY when it's a particular check that only Mazda dealers can do. A 129 million point inspection means nothing if they don't have a rotary compression tester to check the most critical part. An alternative might be to convince them to cover the engine with an extended warranty on THEIR dime. Use the effective tool of placing the responsibility in their hands in either case: "Hey, if you can't check the engine, I'm taking a risk buying it, and you need to cover it in case I'm buying a time bomb.". And then take it to get a compression test as soon as you possibly can.
You should have gotten numbers in a format something like this:
The general guideline for how good, or bad, the scores are, WHEN NORMALIZED TO 250 RPM AT SEA LEVEL!
This is largely my opinion, based on seeing hundreds of compression scores on RX8Club.com over the years. I am not using anything scientific to back it up. As always, your mileage may vary.
Here is a custom chart to make it easier to interpret the numbers clearly. Look up your numbers in the appropriate scale on the left four columns, then move over until you are in the column with your testing RPM. The value of the cell is in kg/cm2, which is the normal standard for measurement in the community. The black line is the failing line. Any 1 rotor face at a 6.9 or lower is failing.