Chassis Dyno Testing - April 2003

Chassis Dynamometer Testing - April 2003

This article was originlly written for Rolling, the magazine of the Volvo Club of America.

I am frequently asked by Volvo owners, " How can I make my car go faster?" The question is usually accompanied by a list of modifications that have been made to the engine, mostly of the bolt on type. And it often includes the customer's idea of what he should purchase next. Even though I am in the business of selling performance parts, my recommendation is very often not the purchase of another "speed" part, but rather that the owner first do a little testing to determine the condition of the engine before making any further changes. The recommended tests would normally include a compression test, a leak down test, and a session on a chassis dynamometer.

Why would I turn down a customer's money, and recommend that he/she direct it elsewhere? The problem for me is that unless I have some information on the engine's current level of performance it is very difficult to make an intelligent recommendation as to what to do next. An item such as a header that would improve the performance of one engine might very well hurt the power output of another.  Is it making good power at low RPM? At high RPM? Is there a flat spot in the powerband? Is the engine currently making more or less power than it was supposed to in stock condition? Have the modifications to date helped or hurt? Have they gone in the direction the owner wanted? Answers to these questions will help me make an intelligent recommendation on what to do next.

To be effective, engine modifications have to be done in a systematic manner. The process of improving engine performance is really one of experimentation, where use of the scientific method is crucial. Control as many variables as possible and make one change at a time, carefully noting the results of each change, but keeping in mind that in some instances several matched parts need to be installed at once in order to produce positive results.  How do we know what the affect of a modification is? By establishing a data baseline and then testing to see how that baseline has changed after the modification. How is that done? By performing a series of quantitative tests. 

Although some forms of testing are out of range of the average car owner, the advent of the modern computer controlled chassis dynamometer has done three significant things.
First, and most important, it has made dynamometer testing readily available to the average car owner. Second, it has made regular "dyno" testing affordable. Third, it has increased the accuracy of chassis dynamometer testing to the point that the results can be compared to those  formerly only available on an engine "dyno".

A chassis dynamometer consists of steel rollers driven by the wheels of the vehicle and connected to a power absorber capable of controlling the load applied to the rollers. The car to be tested is simply driven onto the rollers and tied in place using typical trailer tie down straps attached to its suspension. In a state of the art system electronic instrumentation and computer  software control the individual runs and provide accurate documentation and interpretation of the results. With these controls and the ability to program the manner in which the load is applied, various different types of tests are possible, including steady state power, time to speed, etc., making it an excellent tuning tool. It also allows for the control of the main disadvantage of the chassis dyno, the possibility that unknown degrees of tire slippage will distort the results. The data produced will typically include corrected and uncorrected torque and HP numbers, both for power at the wheels and crank, as well as temperature, barometric pressure, boost, fuel usage, air fuel ratios, etc. 

Compared to the work and expense of testing an engine on an engine dyno, chassis dyno testing is a picnic. Testing on an engine dyno usually means at least two days of work. Much of the time is spent removing and reinstalling the engine in the car and hooking it up to the engine dyno. The cost of this process is usually well over $1000 if the time to remove and reinstall the engine in the vehicle is considered. With a chassis dyno most of the preparation and setup time is eliminated. A half hour to set the car up on the dyno roller(s) and tie it down, an hour or less for several runs, another half hour to review the results and get them printed out or copied on a floppy disk and get the car off the dyno. Total cost - $100 - $200 depending on the shop and the total time spent. But if you come prepared to do real testing involving repeated runs you can spend hours and costs will go up accordingly.

What do you get for your money? Depending on the particular dyno setup the results of a couple of runs can yield a tremendous amount of useful information. Probably most important is whether your car is putting out close to its rated power output. If it is not, that is a clear signal that what you need is not performance parts but basic maintenance, or maybe even a rebuild.
When making engine modifications for street or racing we usually do a dyno test to create a baseline before making any modifications, and then at several stages in the modification process.
We try to do modifications in steps when at all possible so that the results of individual changes or combinations of changes can be plotted. With engines that have already been modified, your first step is to see if the modifications to date have gone in the right direction.

A customer recently sent me the results of a dyno test of his B20 engine. This engine had a very typical downdraft Weber carb, aftermarket cam, header, and an overbore. Although peak torque was up from stock, peak horsepower was down about 20% compared to an unmodified high mileage engine we had tested. Worse, torque fell off rapidly at a 4th gear road speed of slightly over 50 MPH and HP did the same at 60 MPH, leaving no power for open road passing. Although the HP recorded at 60 MPH was only down 3 HP as compared to the stock engine, at 70 MPH the deficit was 15 HP and at 80 MPH the difference had grown to more than 50 HP!! At 60 MPH torque in the modified engine was up 10 ft.lbs. over stock, but by 70 MPH it had suddenly dropped to 40 less than stock. Unfortunately, this type of result is not untypical.

A further comparison can be made with dyno results from the stock engine used in the above comparison after it was supercharged. At 60 MPH at only 4 lbs. boost torque was up 18 ft.lbs. as compared to the "big bore" modified engine, with no decrease in torque through the 80 MPH mark, and was up 30 ft. lbs. over the stock engine at the same speed. Peak HP recorded, with stock muffler, etc. was 55 HP more than the modified engine and 42 HP more than the stock engine at 6 lbs. boost. With modifications to the exhaust port and exhaust system this same stock engine produced over 200 HP at only 8 lbs. boost, more than 100 HP more than the modified, big bore engine.  You can judge which engine characteristics you would prefer for your type of driving, but having this information on actual results of modifications certainly affects my recommendations.

What can you learn in a few runs other than getting peak HP and torque numbers? The list is almost endless. What is the best ignition timing setting for power and economy? Will going to richer or leaner jetting produce more power? Will adjusting the air/fuel ratio and/or increasing the fuel pressure on my fuel injected engine increase power? Is the air cleaner restricting engine power? Will settings that produce less peak HP but a broader power curve result in better acceleration.

Specific examples of what I have learned in only a few minutes on a dyno are as follows. Removing the stock air filter canister on an 1800ES will get you 5 HP. ( Doing a 2130 overbore would get you a smaller increase.) Changing the length of  the air horns on sidedraft weber carbs can affect the output by as much as 20 HP. Shortening the exhaust pipe the correct amount on a B20 race engine can increase the power by 38 HP at 4800 RPM.  In each case, I could have spent hundreds or thousands of dollars on extensive engine modifications to get the same power increases through guesswork.

Without showing you several full dyno plots, it is difficult to demonstrate the full range of information that is available from a modern chassis dyno. There is not enough room in an issue of "Rolling" to print out the full data from even one run, so I am going to try something different, something interactive.

I have done my chassis dyno testing on a Superflow chassis dyno. They have a nice website, www.superflow.com, which has much more information on dyno testing than can be included here. A full selection of testing manuals is available for download from the site, so I suggest that you go there for more information.  The site also has downloadable software which allows anyone to open and manipulate all of the data from a series of dyno runs, including the ability to display the data graphically and overlap different runs on the same graph for comparison.

The plan is this. If you are interested in seeing what the dyno results of stock, street modified, racing, and supercharged B20 engines look like, go to http://www.superflow.com/support/support-dynosoft.htm and download their "wyndyn viewer" program. This  will allow you to view and graph the data from any Superflow dyno system.  Then email me at jparker3@twcny.rr.com with the subject "dyno". I will add you to an email group and then email the whole group a series of dyno test results for review.

If you would like to do your own dyno tests, I suggest that you go to the websites of the two most popular manufacturers of chassis dynos: Dynojet at www.dynojet.com and Superflow at www.superflow.com. There they both have lists of shops that have their dynos, organized by area of the country. After you have done your own dyno tests, email me the results, together with the specifications of your car and any modifications. I will email them to the "dyno"group so that we can all compare and benefit from the results. In a few months I will summarize the results and publish them here. We might even find out if the "perfect beast" makes any real power compared to a supercharged B20.
You might also want to organize your own Volvo "dyno day", by area of the country or club, and then share the results. This has become a very popular car club activity and allows the cost of dyno testing to be reduced.

I look forward to hearing of many happy dyno runs and know there will be surprises, both good and bad.

                                John Parker
                                Vintage Performance Developments