Compression Testing Myths & Facts

[RotaryDiagnostics]

Greetings all,

I've been conducting frequent compression tests for several years now, mostly prototype testing or validation testing of the systems we build. That equates to hundreds of tests of our systems on various rotary engines. In reading threads on the various forums, I see a lot of advice on how to conduct a compression test. Some of the advice appears to be based on rotary lore that has been passed around on the forums for years.

This thread is intended to shed a little light on some of these myths and "facts" that I have tested at various times, mostly when testing prototype systems, but also testing systems prior to sale. I will add a new and different myth or fact to this thread periodically, so check back from time-to-time.

The first "fact" is: "Fully depress the accelerator pedal and crank the engine...". That is the instruction in the Mazda 2nd gen service manual. Well, is that true? And if so, just how much difference does it make? Is it so insignificant that one should not really worry about it?

The answers are yes it's true, it makes a huge difference and one should worry about it. Referring to the graphs below, you will see that the difference between depressing and not depressing the gas pedal can result in about a 20% difference in the readings. For example, peak pulses on Rotor 1 were about 100 psi with the gas pedal not depressed (first image) and 121 psi with the pedal depressed (second image).

All traces in these graphs were obtained from the same 13B Turbo engine, during the same test approximately 30 seconds apart, using the same tester. Incidently the engine was not warmed up for this test (another myth/fact).

So there is no doubt that when conducting a compression test. it is critical to have the gas pedal depressed for the duration of the cranking session.

The question of whether or not it matters if the engine is cold or fully warmed up for a test will be discussed in our next post. Check back.

[Phoenix7]

Great post! I will have to copy and sticky this into the General Rotary tech when it's done!

Now, why is it that the numbers change so drastically with the pedal depressed?

Keep it up and many thanks for helping clear up any incorrect info out there.

[Mobius]

Easily explained: Try breathing through a straw. The amount of air entering the combustion chamber is reduced when you have a restriction blocking the path of the flow.

Less air means there is less to compress, which follows that the overall pressure will be reduced.

[josh18_2k]

i thought you were supposed to press the gas to disable the injectors

[djmtsu]

i thought you were supposed to press the gas to disable the injectors

That is how you start a flooded S5.

For a compression test, I pull the EGI fuse on the strut tower.

[RotaryDiagnostics]

Great post! I will have to copy and sticky this into the General Rotary tech when it's done!

Now, why is it that the numbers change so drastically with the pedal depressed?

Keep it up and many thanks for helping clear up any incorrect info out there.
Thanks!

As Mobius indicated, I believe the difference is in the volume of air the rotor can "suck" into the combustion chamber--more air, more molecules, more pressure. The throttle plates when closed (gas pedal not depressed) will limit the amount of air available during an intake "stroke".

[JustJeff]

I recently tested compression and forgot about having WOT, first test were about 15-20psi difference between having WOT and not

[RotaryDiagnostics]

That sounds about right. What were you using for a tester?

[RotaryDiagnostics]

Often, the advice offered up by forum members and techs alike is that to get correct data during an engine compression test, it is important to always conduct the test on a fully warmed up engine. I conducted tests today specifically to attempt to answer the question of the impact of engine temperature on test results.

In short, the results of the tests indicate that the temperature of the engine has little, if any, effect on the actual compression numbers. Note the term "actual". By actual I mean the recorded numbers normalized to an engine cranking speed of 250 rpm in order to compare them with the specifications in the Mazda factory service manual.

Test conditions: The cold test was done on a 13b Turbo engine that had been idle for 2 days prior to the test. In other words, it was stone cold. Ambient temp was about 65 degrees. The warm test was done after a short road run with coolant temps rising to about 185 degrees and oil temps to around 170 degrees immediately prior to shutting the engine down. In other words, the engine was at normal operating conditions.

The setup was identical for both tests. Same engine, same plug holes, same tester system with the same Acquisition Units connected the same plug holes. Time between the tests was about 20 minutes. The warm test was conducted immediately after shutting the engine down.

So what exactly does the data reveal (refer to the attached images)? The max compression reading for rotor no.1 in the cold engine was around 126 psi compared to the max compression for the same rotor in the warm engine of about 129 psi. So there was a small difference in the recorded readings. However, the engine cranking speed for the cold engine was 237 rpm as compared with the warm engine cranking speed of 267 rpm. If the recorded readings are normalized to the same engine cranking speed, there will be little difference in the actual compression of the engine--cold or warm.

So conduct your test on a cold or a warm engine, but remember to consider engine cranking speed when evaluating the results. As a final caveat, by cold engine I mean an engine that has been run in the days prior to the test, not one that has been idle for weeks, months or years.