RFI Troubleshooting

Reducing problems with ignition or other engine noise interfering with sensitive computerized data-acquisition systems.

Basics | 07.20.20

How can I reduce ignition noise?

Electromagnetic Radio Frequency Interference (EMI or RFI) from the test engine’s ignition can be a major problem, when trying to capture data via sensitive electronic instruments running adjacent to a high compression spark-ignition engine. EMI symptoms include poor dyno tachometer readings, erratic servo operation, PC lock-up, and gaps or spikes in the captured data.

Interference worsens as the engine’s throttle is opened – this is due to the rise in spark voltage that accompanies any increase in cylinder pressure. Supercharging, turbocharging, and many race modifications that increase cylinder charging drive up ignition energy demands, often creating severe interference problems. As RPM increases (raising the EMI/RFI frequency) the number of interfering “pulses” generated goes up proportionally. The worst conditions often occur right around peak torque output (both cylinder pressures and RPMs are high) and the resulting EMI is often known to crash (blue screen) even a high quality PC running nearby.

Fortunately, there are steps that can be taken to reduce the interference at the source and also to harden the dynamometer’s electronics and your PC from this ignition “noise.” Taking one or more of the following preventive measures will generally allow testing of even the wildest combinations. EMI problems are not in any way specific to the DYNOmite, so your ignition system’s manufacturer may have other suggestions you can follow.

1) Run all ground leads to either the engine’s battery (or block if no battery is in use) or to a clean common junction connected by a heavy-gauge ground cable to the engine’s battery. Ideally this battery’s negative terminal and/or dynamometer’s frame itself should be connected directly to a quality earth ground (e.g. buried large-diameter 8′ copper ground rod) via a heavy-gauge wire (for a very low resistance connection). Avoid multiple ground points as this often creates a “ground loop” (where RFI can get onto the circuit).

CAUTION: Always hook up all the ground leads before hooking up any of the power leads for the DYNOmite data-acquisition module, its accessories, or the engine. Failure to follow this precaution can lead to circuit damage not covered under warranty!

2) Try using resistor sparkplugs (or caps) and also run graphite-core ignition cables (e.g. Accel Graphite Core wires). These components tremendously reduce the level of RFI transmitted out from the engine’s spark ignition. Subduing the noise at the source of generation is much more effective than chasing down every dyno harness lead that might be receiving these unwanted, random, signals.

CAUTION: Spiral wire wound “suppression” ignition cables are not nearly as effective (at reducing EMI) as, less-expensive, true graphite core cables. The cheapest set of graphite wires will “out-suppress” the most expensive wound metallic wires… period.

Tip: Keep an inexpensive set of graphite core cables in the dyno cell for troubleshooting problem ignitions. You don’t need expensive high temperature 8+ mm silicone jacket wires with high quality boots to survive on the dyno. Even installing just a graphite core coil lead may do the trick. New graphite cables will not reduce the engine’s power.

3)Keep the ignition wires (and coil wire) as short as practical and avoid routing the DYNOmite’s harness leads parallel or near to the ignition wires.

Tip: The EGT probes have extra shielding on them to allow operation near the ignition. Conversely, the Engine Temperature leads are quite sensitive to RFI, keep them away from the ignition if you can.

Tip: Temporarily install braided metal shielding, such as Aeroquip stainless braid hose, (grounded at one end to the engine block) over all the sparkplug and coil wiring. This trick was used on early Corvettes and many marine applications – where the fiberglass engine cowling provided poor RFI suppression.

4)If EMI is interrupting communications between the DYNOmite Data Accusation Computer and your PC running DYNO-MAX (via a RS232 serial port connection) the use of Land Sea’s special shielded and filtered cables may help. Symptoms include dropped data packets or random data spikes that worsen in intensity as the throttle is opened, increasing cylinder pressure and spark intensity. Avoid using USB adapters as they just add another point of EMI sensitivity – instead use a computer with built-in RS232 ports (or add an RS232 port bus card) so that there will be an uninterrupted grounded shield from the DYNOmite to the PC..

Tip: In extreme cases fiber-optic RS232 serial “cabling” is available. Such hook-ups use a converter at each end that translates the normal hard-wired connection into a light signal that runs along the fiber optic link. However, be sure to select models that do not require power from the PC or DYNOmite RS232 signal to operate! If you go fiber optic, order 100′ long leads so that you can run one test with the PC far away from your engine – if the problems go away during the remote test, it is your PC – see suggestions below.

5) Modern computers have gotten lighter and less expensive – and plastic housings are part of the reason. Unfortunately, plastic offers much less EMI shielding that the heavy metal cases used for early PCs. Toady, even many of the most expensive model PCs and laptops are very poorly “hardened” against the levels of EMI present near a spark ignition engine. If you can, try a few other models (or brands) to find the most robust computer.

6) Most modern PCs are no longer grounded via their power cords or chargers. In certain cases running a large gauge (#12 or 10) wire to ground the PCs case helps – try it both ways. Moving a problem PC further away from the engine is always a good idea. Sometimes, simply reorienting the position of the PC (rotating it 90 degrees), moving the angle of the power cord, etc. may uncover a less sensitive position (just as moving a radio’s antenna effects its ability to pick up a weak station). Installing inexpensive ferrite “RFI chokes” around the ends of the PC’s power and DYNOmite’s communication cables may help. These are available from most electronic-supply houses – in various diameters.

7) Place your PC and monitor inside an EMI shielded enclosure to harden it against electronic (and acoustical) interference. You do not necessarily need a $6,000 military specification EMI housing – something like the DYNOmite RFI/EMI Shielded PC Cabinet should do the trick.

Maybe it’s not ignition noise!

Here is a question, and our suggestions, regarding one user’s problem we received several months ago:
“Sometimes my laptop’s screen freezes up in the middle of a long hard pull on our Harley motors. This only seems to happen when I am both recording and have the throttle wide open. As soon as I back off the power, the screen comes alive again. I believe it’s RFI, but why does this only happen while I am recording?”

Your computer is probably sitting close to those engines. Laptops and newer PCs feature some of the lightest (per gigabyte) hard-disk drives ever made. This is great for reducing travel weight, but not so good for shielding their sensitive innards against the noise inside a working dyno cell. Under the extremely noisy conditions typical during a WOT pull, many new hard-disks refuse to transfer data – which, of course, is actually better than recording erroneous data.

Some newer hard-disk drives even lock themselves up on purpose – especially if it is a laptop drive! Engine exhaust noise falls right into the frequency range that the latest drive’s “shock protection” accelerometers are tuned for. Each time the noise gets loud enough, the BIOS parks the disk-drive’s heads – in anticipation of a bigger bump onto the floor.

So what happens to your real-time data? All personal computers provide solid-state (memory) buffers, where data queues up while waiting for the physical drive to respond (to Window’s write requests). Unfortunately, the buffers are small, intended to handle only short (one or two second) periods of drive unavailability. Once its buffers fill, Windows must sit and wait for the disk-drive heads to begin responding again.

If your dyno pull is fairly short (e.g. a quick inertia only test), the drive buffers may be large enough to temporarily store all the real-time data. Once the noise subsides, the drive catches up and so there is no problem. However, during your longer full-throttle pulls, or if the disk-drive’s buffers are small, the buffers fill to capacity and Windows hangs – patiently waiting for more room on the disk drive’s “loading dock.” During that period the PC’s screen appears frozen.

What can you do if your personal computer is a victim of this problem? Here are a few potential solutions:

Solution #1) Upgrade to DYNO-MAX 10.09 (or later). It includes separate programming threads and data writing buffers that are able to deal with many extended drive-outage periods.

Solution #2) Replace your old mechanical hard-disk with a modern 100% SSD (all solid-state drive) unit. These non-magnetic and non mechanical units rely on the latest flash-memory technology to store your data. Any new 100% SSD will be more robust when it comes to dealing with the conditions in a dynamometer bay compared to a mechanical disk-drive. For examples of top-notch SSD offerings, check out Intel, OCZ, Samsung or Toshiba units.
IMPORTANT: Do not confuse Hybrid drives (a conventional mechanical/magnetic disk paired to an oversize solid-state buffer). Maybe the buffer will take longer to fill and lock up – but why risk the possibility of doing the upgrade and not fixing the problem. A genuine 100% SSD drive is just not that expensive these days. Be sure to read the drive’s box carefully – to avoid buying a hybrid drive when you go shopping for a 100% SSD models!

Solution #3) A cheap/crude solution (while you wait for your 100% SSD drive) is using a Fast USB (2.0 specification) “thumb drive” to relocate your DYNO-MAX 2010 “TempRuns” folder into. Then set the Tools- Preferences tab to point DYNO-MAX 2010 to that Temp Runs folder location. Note: Your computer may not support the latest USB 2.0 specification. If it does not, skip trying to use the slower “1.1 specification” thumb drives – wait to buy a 100% SSD drive!

Tip:For DYNO-MAX 2000 you would have to place/run the entire DYNO-MAX 2000 installation onto the USB thumb drive. Unfortunately, while these USB-interface devices are less expensive, they are much slower, than a genuine 100% SSD drive.

Solution #4) Install “RAM disk” software to create a virtual disk drive in the PC’s memory. Then specify that drive as the location for DYNO-MAXs “TempRuns” folder. Note: Be sure to use a software package that provides “write caching” and includes automatic data synchronization with a physical copy on the PC. One, of many, such products can be found at www.everstrike.com/ramdisk and offers a free trial.

Solution #5) Avoid the noise. Buy, or build, an industrial enclosure for the PC. Or, if practical, move the PC (hard disk) further away.

Solution #6) If you have over $5,000 to spend on a noise-hardened military-grade laptop, check out www.ruggednotebooks.com. Once you chose to go this route, be sure to specify their 100% solid-state drive and additional RFI/EMI hardening options!

Similar Posts