PART 2 - Fluid Brake Dynos
By John Copeland
Last month we started talking about different types of dynos and some of the basics about what a dyno is and what it isn't. We talked about how important accurate data was if you hoped to get anything meaningful from your dyno experience. Over the years there have been dozens of outfits on the market that claimed to top quality information for a bargain price. Don't believe it!
While the basic design of a dyno is pretty simple, a really good dyno is a very sophisticated piece of equipment. If someone tells you they can sell you a perfectly good, top quality dyno for a few hundred dollars, well let me tell you, it just ain't so. Like in so many other things, you get exactly what you pay for. A bargain-priced dyno may work ok to break in engines on, but don't expect to get any consistent, worthwhile data from it. A good, useful dyno is the result of untold hours of sophisticated engineering. It must be robust enough to withstand the rigors of repeated high-frequency vibrations. It must deliver consistent, repeatable data. And it must be "user friendly." A dyno that is a pain to use....simply won't get used for long. Let's begin by looking at the basic fluid brake dyno.
If you remember from last month, a dyno measures engine output by comparing it to a known, or measurable resistance, called the load. In the vast majority of dynos in use today, either oil or water circulating through a pump provides that load. The concept is simple: the harder you make the pump work, the harder it makes the engine work. If you increase the load on the pump to the point that the engine can't generate enough power to drive the pump, the engine will slow down. Since that slows down the pump, it reduces the load on the pump and the system stabilizes at the new, lower RPM. Likewise, if you reduce the load on the pump, the RPM of the engine will increase until it reaches equilibrium between the power output of the engine and the load on the pump. In most cases the load on the pump is controlled by means of a valve on the output side of the pump. Take a look at the diagram in Figure 1 and you'll get the idea. By putting a valve in the line running out of the pump, you can control how hard the pump has to work. As you close down the valve, the pump tries harder to push the water or oil through the smaller opening. The farther you close the valve, the harder it works. Just a note here for you do-it-yourselfers, you need a "positive displacement" type pump for this application. That is, it must be a pump that mechanically forces the fluid through the pump. Typically, this is the type of pump that uses two gears that mesh together and force the fluid out of the pump outlet. A turbine type pump, like you might use to draw water from a well or to circulate water through a heating system, just won't work. If you close down the output side of a turbine pump, it simply spins idly and won't apply the needed load to the engine. With the proper type of pump, if you close the output valve completely, the pump may barely turn at all. It will virtually lock up solid!
As I mentioned before, the fluid for this type of dyno can be either oil or water. Water is more widely used, because it's cheaper, easy to replenish, and generally easier to handle. In fact, you don't really even have to recirculate it if you don't care to; just bring it into the pump from whatever source, and send the output down the drain. While that's pretty wasteful (and not too environmentally sound), it does avoid any problems with controlling fluid temperature. You see, as you force the fluid, be it water or oil, through the pump and then through the restriction of the output valve, it heats up. The more energy it absorbs from the pump, the more it gains heat.
This not only makes the whole apparatus hotter and more difficult to work with, but it can change your output readings as well. If you're going to use a fluid brake dyno for any serious work, you really will need to develop a system to maintain consistent fluid temperatures. Obviously, the larger a reservoir you use for the fluid, the easier it will be to maintain an even temperature. But you'll probably need some sort of cooling mechanism to circulate through, like a radiator core. There are several reasons to consider oil as the fluid in the dyno, rather than water. First of all, if you're looking at building something yourself, positive-displacement oil pumps are generally easier to find than positive-displacement water pumps. Worn-out hydraulic pumps have found new life on small engine dynos for as long as I can remember. They're relatively inexpensive, and the fact that they've worn to the point that some of the oil slips past the impellers is a plus in this application. That leakage inside the pump itself makes it a lot less "touchy" on the control valve. Oil is also self lubricating in the pump and does not create any rust or corrosion problems. On the down side, an oil pump system must, of course, be recirculated, a "closed loop"system. And, not only does the oil heat up even more dramatically than water, but, as it heats up, it changes viscosity dramatically. The hotter it gets, the thinner it gets. And as it gets thinner, it flows through the valve more easily and your control with the valve can get trickier. Temperature control with an oil-brake dyno is critically important. If you expect to get any reasonable use out of an oil-brake dyno, it must be fitted with an efficient, dependable oil temperature control system. Better quality units include an oil pre-heater to bring the fluid to a pre-determined temperature, as well as a thermostatically controlled cooling system to keep it there.
You'll recall that last month we talked about measuring the output of the dyno by measuring the torque applied to the pump housing as the pump load increased. The concept is pretty basic; the harder the engine tries to turn the pump, resisted by the back-pressure created by closing down the valve, the more the pump housing wants to twist. Measure that twisting force and you're measuring torque. Conveniently, the torque on the housing of the pump is always exactly equal to the torque on the input shaft of the pump. But there is another alternative for measuring engine output, if you're using an oil or water pump for the brake. By accurately measuring the output pressure, between the pump and the valve, you can gauge the torque on the pump shaft. But wait, it gets even better. Most hydraulic pumps are factory rated with the torque to output pressure figures. That means that you can directly convert the readings on the pressure gauge into to foot/pounds of torque. And, since this relationship is linear (i.e. if 10 ft/lbs = 50 PSI on a particular pump, then 20 ft/lbs = 100 PSI), it makes it a snap to set up. All you have to do is re-mark the face of the pressure gauge into foot/pounds based on figures supplied by the pump manufacturer and you have a direct-reading torque measuring dyno. See figure 2. For the home builder, the pressure gauge readout is a heck of a lot easier to deal with than trying to set up a spring scale or an electronic load cell.
There are several good quality fluid pump type dynos commercially available, as well as others that are not so good. Remember, the criteria for selecting a dyno has to be; 1) can it give me the quality of data that I need to make meaningful determinations about whatever tests I may be running? and 2) will it hold up to the hard, repeated stresses that my testing will subject it to?
For decades Stuska dynos have set the standard for small engine, fluid-brake dynos. These units are robustly built to withstand continuous commercial usage far beyond anything the average karter or engine builder is likely to need. The Stuska unit utilizes a variation of the pressure gauge system of data output, and load is controlled via a very finely threaded valve on the output side of the pump. However, unlike systems that directly read the output pressure of the pump, Stuska uses a "torque arm" mounted to the pump housing. At the end of that torque arm is a hydraulic cylinder, and the output of that cylinder is measured by a pressure gauge. Stuska believes that this system gives them the best combination of accuracy and repeatability.
They are also readily adaptable to electronic readout and data collection by simply substituting an electronic pressure transducer for the gauge. These dynos are available in a variety of sizes and configurations from big enough to test a motorcycle engine on to tiny enough to accurately analyze the performance of a model airplane engine. There are probably more Stuska dynos out there is use for karting applications than any other manufacturer. In most cases, these units are mounted on custom build test-beds designed to suit each users particular application, but the manufacturer can provide ready-made platforms.
There are other manufacturers out there marketing fluid-brake dynos specifically designed for karting use. Some of these units feature engine mounts and dyno stands already engineered to accept most modern kart engines. Most maintain the proven principles of the Stuska-type unit.
One of these systems is available from Davenport Dynamometers, Inc. of California. This unit features a sturdy platform that mounts both the engine and the water brake. The load control valve, as well as sensing and output hardware, are mounted on another framework, allowing these control functions to be operated remotely, away from the howling engine. While Davenport's unit is not remarkable from a mechanical standpoint, their computer interface and data collection and analysis systems really set this system apart. We'll talk more in another article about the critical subject of gathering data from the dyno and making sense of it.
Another fluid-brake system designed especially for karts is made by the International Dyno Corp. and marketed by Franklin Motorsports of Oak Creek, Wisconsin. Originally designed as a compact, all-in-one unit, at under $1500 it put owning a dyno within almost everybody's reach. But some serious users found the framework wasn't stout enough to withstand the rigors of daily use. Now International Dyno has a new model, more robustly engineered. The new Model 500 Kart Dyno features a motor-mount and hydraulic pump assembly that is separate from the rest of the unit. The controls, including an oil temperature gauge, mount on the water-cooled oil reservoir tank. The system is compact and, relatively, portable. Like the Davenport Dyno, the Model 500 couples the engine to the pump via a cog belt drive, like the ones enduro karts use. This system provides smooth, clean power transmission with minimal friction losses. The Model 500 utilizes "Pressure Gauge" output and is, in it's standard form, only suitable for manual data recording.
There is one fluid-brake dyno on the market today that can be driven either directly by the engine crankshaft or by the axle shaft with the engine still mounted to the kart. The DYNOmite Dynomometer system is manufactured by Land & Sea of Salem, New Hampshire. These folks make a variety of performance oriented products for marine applications, as well as test equipment for snowmobiles, chainsaws, and so on. Unlike the other fluid-brake dynos we've discussed, the DYNOmite unit uses electronic torque measurement. You'll remember that earlier we talked about measuring the torque output by measuring the twisting force on the pump housing. Well, that's exactly what the Land & Sea folks are doing. The DYNOmite unit has an integral "torque arm" built right into the pump housing unit. And partway along the length of this torque arm is an electronic strain gauge. A strain gauge is a device that measures the amount of force that is trying to bend the surface that it is mounted on. So, as the twisting force on the pump housing increases, it puts more and more pressure on the torque arm and the strain gauge measures that force. It is significant to note here that, unlike that other commonly used fluid-brake dynos, the DYNOmite unit, because of it's electronic output, and is automatically suited to computerized data collection. That's not to say that the other dynos can't be computerized. Quite the contrary. The Davenport Dyno is specifically designed to support automatic data collection. But that's not always the case. Most simple fluid-brake dynos are best suited to manual data collection. That is, you stabilize the unit at a given RPM setting, read the output information, write it down, and then adjust the load to a new RPM setting and repeat the process. Manual data collection is slow, and that extended run time makes it harder to "filter out" variations in data that occur as engine temps and other factors change. In general, the shorter the run time of your dyno runs is the better quality your data will be. That makes electronic data collection a high priority. In an upcoming article we'll get into the advantages and pitfalls of electronic data gathering on the dyno. And next month we'll look at totally electronic dynos; no water, no oil, just lots of "juice" of another sort running the dyno, providing the load, and measuring the output.
In the meantime, start asking around about what sort of dyno facilities your local engine builder might have available. If you're serious about your karting, you can't afford not knowing what your engine is doing. Being able to accurately predict where your engine will make peak torque and where the output falls off to the point that you don't want to turn any more RPM will work wonders with your on-track performance. Even more important is knowing what to expect when you change pipes, or carbs, or fuel mixtures. Oh, I know you can figure these things out by what happens on the racetrack, and I know that there is no substitute for track testing, but you can certainly make better use of the limited track time you have if you've done your homework on the dyno before you headed out. A good, accurate, repeatable dyno is an invaluable tool and no top-flight engine builder should be without one. As karting gets more and more professional and karters take their sport more seriously, more individual karters are getting into dynos as well. This is one area where knowledge really is power.