53 replies to this topic

[Jim Drago]

There will be nothing in this thread other than constructive items. There will be no mention of Daytona or protests etc.

 

 

 

If you have any info or constructive suggestions or ideas please enter them here...

 

 

 

Today, we were assembling a 2001 engine. I cc'ed the engine which we RARELY do at this point any more as we build all to the same formula. I think most engine builders do something similar.  

 

As many have stated, cc'ing or pouring the engine is not as easy as many think, it is also very time consuming. Imo, I think I could teach a fifth grader how to accurately measure compression with a whistler and be able to repeat the results 99 times out of a 100.  To get someone to cc accurately is 10000x harder and some could never do it right.  I hope someone better than me will write a detailed procedure here or maybe even do a video.  The main reason was to check the calculated compression on an 01 versus what I read with the whistler.

 

The readings I got by cc was 9.94, the reading with the whistler, valve cover removed was 9.9/10.0, which seems consistent with what others are seeing. 

Please share your experience and ideas here. 

 

 

People have asked how much the valve covers change compression. The valve covers themselves are not all the same, but the average change is listed below... Please comment if you have seen something different. 

 

 

 year               valve cover on               Valve cover off

90-93                 9.0-9.1                                    9.4

94/97                  8.7                                          9.0

99/00                  9.2                                          9.5

01/05                  9.5-9.6                                   10.0

 

Jim

[pat slattery]

please explain why the valve cover on or off makes a difference.  I have no idea

[Johnny D]

please explain why the valve cover on or off makes a difference.  I have no idea

 

You have to read the SIC Tech thread but long story short.

You get a tone, to tell you the CC, the VC muffles it and without it doesn't.

A true reading is with it off.

J~

[wheel]

The whistler is just like blowing on the lip of a pop bottle with liquid in it.  The less the volume, the higher the pitch.  The computer in the whistler compares the pitch with the piston at the bottom with the pitch with the piston at the top.  When it knows the cc of the engine, it can tell you, very accurately, what the ratio is between the cylinder with the piston at the bottom and with it at the top.  There are some factors, including calibration of the unit, temperature, incorrect engine displacement entry, presence of additional fuel in the cylinder, etc. that can change the reading.  When used properly, it is as accurate as an actual cc measurement.  The cc measurement, which has been pointed out, is not an easy thing to do, and has issues of its own.

wheel

[john mueller]

 year               valve cover on               Valve cover off

90-93                 9.0-9.1                                    9.4

94/97                  8.7                                          9.0

99/00                  9.2                                          9.5

01/05                  9.5-9.6                                   10.0

 

 

Do you have CC data to add to that chart?

[Mike Collins]

The previous and still sort of current procedure for whistling a car allowed for the valve cover to stay on and was considered a non-invasive procedure as per the manufacturer recommendation.  The tech bulletin, not procedure...calls for the valve cover to be removed on certain overhead cam cars.

 

My question becomes...other than at the runoffs who is going to post the bond to remove the valve covers?

[Johnny D]

Do you have CC data to add to that chart?

 

 

The readings I got by cc was 9.94, the reading with the whistler, valve cover removed was 9.9/10.0, which seems consistent with what others are seeing.

Please share your experience and ideas here.

 

 

 

J~

[Charlie Hayes]

A older engine sitting around the shop I CC'ed at 9.75:1 with the valve cover off it will flicker between 9.7-9.8:1 on the whistler.

A 99 engine I did math for would come in at 9.41:1 CCed and reads 9.4:1 on the whistler.

The flicker is the whistler reading within the hundredth and rounding up or down. It isn't that a valve cover muffles the sound it doesn't allow the probe to fully seat to the spark plug adapter which is then filling a bigger hole. The tubes on my whistler give a crazy low reading when valve cover on. That's wherr quality control comes into effect.

CCin it will give you the absolute number but as mentioned before minus the engine being in a shop, on a stand, with the correct tools and brain behind them it is a lengthy process. I have been there and helped with it it takes someone who has been building engines for 25 years a decent amount of time and measuring everything twice or more. Not sure if someone at a club race doing tech can perform such on 3 warm dirty engines with a bunch of chaos going on.

[Diller]

Next year will be my first year racing SM. Right now I am racing with in PTE with NASA so I don't yet have a dog in this fight but I have been doing my best to familiarize myself with the fine points of the series. One of those being test procedures to ensure legality.

 

I think one area that needs to be addressed and agreed on gauge calibration, gauge repeatability and reproducibility, and statistical variation. Anybody involved in quality control (specifically TS16949 or ISO9001) should know how easy it is to find holes in a suppliers gauging method or how hard it can be to defend your product via measurement data. Measuring CC is no different then quality control. I feel strongly that there will need to be an agreed upon standard for calibration, tools, and procedures before you can satisfy the test method. Without this understanding/agreement, one could always argue that the test method is meaningless with one wrong reading out of 1000. To be clear, if an automotive supplier shipped one bad product per 1000, the customer would find a new supplier rather quickly. For those familiar with six sigma, the bar is set at 3.4 bad parts per million. If we were to have a tool that could test a million engines and only 3.4 were a false positive/negative, I sure a hell hope that we, as functioning adults, could understand the statistical likelihood that the test method is still valid.

 

All of that being said, there is no way we can expect to get close to a six sigma level of measurement at an amateur level. Hell, even professional gauge makers struggle to keep that level of control. We can however collect data on each suggested test method and calculate how accurate it will be over 1 million test cycles. I believe that may help some to be more comfortable with one test method over the other. 

 

If somebody is willing to put in the time, I have access to Mini Tab that will help calculate a gauge R&R for any one method. In my line of work, we typically have three people test three samples thirty times. This will be hugely time consuming for somebody but I do think it will be worth the time if the class wants to put this particular issue to bed. 

 

As some have stated in the other thread, there are other ways to gaining CR other than just the heads. It sounds like a separate test would have to be used to isolate the variables discussed to make sure you know exactly what is being tested or to make up for the shortcomings of a particular method (whistler).

 

This matter (gauge R&R) may need to be a totally different conversation but I wanted to bring it up before things get too far set in stone.

 

Cheers and happy racing.

[Danny Steyn]

Jim, your numbers above replicate almost exactly what Mike Rossini has found with his whistler. And it also reflects the low numbers that Jim Creighton has observed over the years when using the Whistler on our cars with the valve cover on.

 

The part that we should be focusing on, and what Charlie and Jim corroborate above is that the Whistler is a VERY ACCURATE REFLECTION OF COMPRESSION RATIO when done properly and with the valve cover off.

 

As far as I am concerned we should absolutely change the procedure specifically for our class to reflect that all future Whistler measurements need to be done with the valve cover off.

[Bench Racer]

A older engine sitting around the shop I CC'ed at 9.75:1 with the valve cover off it will flicker between 9.7-9.8:1 on the whistler.

A 99 engine I did math for would come in at 9.41:1 CCed and reads 9.4:1 on the whistler.

The flicker is the whistler reading within the hundredth and rounding up or down. It isn't that a valve cover muffles the sound it doesn't allow the probe to fully seat to the spark plug adapter which is then filling a bigger hole. The tubes on my whistler give a crazy low reading when valve cover on. That's wherr quality control comes into effect.

The question was asked in the other thread as to how the valve cover or location of valve cover impacted the Whistler. Now we have another input on what impacts the Whistler. Do we have more input on what impacts the Whistler? I'v had issues doing leak down tests with a LongAcre tool which didn't seal correctly to the head like a spark plug washer does. I'v viewed the Whistler being used two times on a Miata production car.  I buy the impact point of the the Whistler probe not fully seating with the adaptor made by Mr. Hayes.

 

Point 7. of the SCCA Whistler procedure:

7. Determine correct spark plug adapter and install it in place of the spark plug (minimal torque is required). In some cars, especially overhead cam vehicles with spark plugs well down in the engine, it may be necessary to remove the valve cover to get an accurate reading.

 

If there is revised verbage to the SCCA Whistler procedure, please identify what and where.

 

If some one can explain how the valve cover casting/whetever or valve cover location impacts the Whistler readings, I'm ready to learn.

 

Thank you

David Dewhurst

[38bfast]

I have been involved in Motorsports for most of my life. Just about every form of motor racing seems to use a whistler to check compression. All that I have been evolved with have had the same issue. All have concluded that the only way to accurately measure an engine is to cc it. As we all agree that cc an engine in the feild is a pain in the ass. So the best way we can feasibly do this is go by the whistler with the valve cover off. If you are found noncomlyiant then you have the option to appeal and request the engine be cc. For the runoffs jump right to cc. If you have an engine that has pushed the limits too far you have exactly that.

[Dan Tiley]

The flicker is the whistler reading within the hundredth and rounding up or down. It isn't that a valve cover muffles the sound it doesn't allow the probe to fully seat to the spark plug adapter which is then filling a bigger hole. The tubes on my whistler give a crazy low reading when valve cover on. That's wherr quality control comes into effect.
 

 

Charlie... that's not correct.  I have whistled every single engine I have built... literally hundreds.  Try this:  Cup your hand around back of the whistler probe when you put it in your calibration jar and you'll see a reduction in the CR reading.  The valve cover on a Miata does the same thing.  I actually have the OHC specific probe for my whistler from Katech, which is different than the one typically used in tech.  It has the same whistle, thus giving the same numbers, but it has a steel extension tube that is longer than the depth of the spark plug hole so you can push the probe all the way down to the adapter.  Again... doesn't change the reading, but makes it easier to use.

 

The SM specific procedure IMO should address the following:

1) If we create a "shifted" spec that allows valve covers to stay on, I can guarantee we'll see some very creative new valve covers with just slightly smaller diameter spark plug holes.  From the standpoint of an engine builder... please remove the temptation!!!  

 

2) Another cheat is for a sleeve to be installed in the head that doesn't allow for the whistler to fully reach the bottom of the hole and contact the plug adapter.  Again, removing the valve cover makes this much easier to spot.  I plan to lobby for a line in the procedure to specifically check for this.  It's literally an eyeball check, and very simple with VC removed.

 

3) Competitors should NOT be allowed to install the tool.  I've talked to John Bauer at SCCA about this, and his concern was that a tech official could possibly strip the spark plug hole.  I much prefer that very unlikely risk, especially if started by hand, vs. a competitor not fully threading in the plug, or sneaking in some sort of shim.  The competitor may remove the spark plug, the official installs the whistler plug.

 

News flash... we're not all saints. On track, most drivers will slip their car into nearly any gap they can squeak their car thru.... close these gaps in the rules and tech procedures so engine builders (yes me too) aren't tempted to do the same.

[Dan Tiley]

Oh yeah... One more:

 

4) The Whistler is supplied with several spark plug adapters, 2 of which are the correct thread for a Miata.  One has the correct flat seat surface and crush washer, similar to our spark plugs.  The other has a tapered seat.  Guess what.... the tapered seat version reads 0.1 lower than the correct fitting.  Have I seen it used in tech sheds???  Absolutely!

[Parity]

SCCA, are you listening? These guys know how to use and misuse the tool. How about developing a detailed procedure with their input?

[FTodaro]

what is the conversion value you use for the whistler to convert to Cubic Inch?  is it 112?

[Dan Tiley]

what is the conversion value you use for the whistler to convert to Cubic Inch?  is it 112?

 

112 standard bore (actually 112.3)

113 overbore  (actually 112.9)

 

How many times have I seen overbores measured at 112?  Every time!

[FTodaro]

112 standard bore (actually 112.3)

113 overbore  (actually 112.9)

 

How many times have I seen overbores measured at 112?  Every time!

Thanks, that is why i think the protocol should spell out every detail.

[Charlie Hayes]

Charlie... that's not correct.  I have whistled every single engine I have built... literally hundreds.  Try this:  Cup your hand around back of the whistler probe when you put it in your calibration jar and you'll see a reduction in the CR reading.  The valve cover on a Miata does the same thing.  I actually have the OHC specific probe for my whistler from Katech, which is different than the one typically used in tech.  It has the same whistle, thus giving the same numbers, but it has a steel extension tube that is longer than the depth of the spark plug hole so you can push the probe all the way down to the adapter.  Again... doesn't change the reading, but makes it easier to use.

 

Mine must be bent then...the valve cover gets hung up on the tone tube not allowing it to fully sink into the spark plug hole on top of the adapter.

 

I have tried cupping my hand like that before and it can skew the results, but cuping your hand on top of the probe is more of an abstruction then the valve cover in my opinion?

 

Either way valve cover off gives a very clear result without the other probe and I think we can agree on that?

[Dan Tiley]

I have tried cupping my hand like that before and it can skew the results, but cuping your hand on top of the probe is more of an abstruction then the valve cover in my opinion?

 

Don't cap the top of the probe with your hand... shape your hand like a cylinder, (as if you were rubbing one out) to simulate the valve cover being there.  Or just hold a loose valve cover there.  Then turn on some Barry White music and measure away! 

 

Depending on how small you make your hand will determine how low the number goes.  My calibration bottle is 12.8, but I can make it read anything down to about 10 without pulling it off the bottle.  

 

You're right though.... it's really important for the probe to fully seat against the plug.  With the plug backed off just one thread (one rotation), it's almost 0.1 in the competitor's favor.