Anyone have any tricks on what to insulate an AFR gauge wire with to eliminate RF noise coming from the coil and other electronics? My gauge keeps freezing on a number and not reading correctly. Autometer said it was RF noise but there is really no other way to run the wires safely other than close to the coil pack. Any ideas?
RF noise interfering with gauge
Started by
suck fumes
, 03-06-2012 07:59 AM
#1
Posted 03-06-2012 07:59 AM
2010, 2011 SM Sowdiv points champion!
2011 National points champion!
WWW.MOTORSPORTHEAVEN.COM
2011 National points champion!
WWW.MOTORSPORTHEAVEN.COM
#2
Posted 03-06-2012 10:01 AM
I have an AEM so, YMMV. Sensor is in the downpipe and the wiring comes up on the exhaust side and into the cockpit via the throttle cable grommet. excess wire harness is bundled under the brake booster.
#3
Posted 03-06-2012 01:01 PM
Eliminating RF noise from a coil system is notorously difficult. Harness relocation is likely to be much more effective than any sheilding / filtering methods. Even a few inches of additional separation can help a lot. If you can't relocate the wire, simply changing the routing can help alot. Much like changing the antenna orientation on a radio. If the wire is running parallel to the coil-pack, try running perpendicular to it, and making a 90 degree turn somewhere else.
If you absolutely can't move the harness, then I would try shielding the harness with tinned-copper overbraid. The overbraid MUST have a good connection to the chassis ground, to be effective. Use 14-16 AWG stranded wire, and keep it as short as possible to connect the shield to the chassis (less than 6 inches, 1 inch would be better). Make sure the chassis metal is absolutely clean and makes maximal suface area contact. If you have a ground stud on the gauge, that would be best. But, you still need to connect that ground stud on the gauge to the car structure in the same manner.
Typically, you will get best SHEILD performance with both ends grounded...but, that is somewhat variable. Often grounding both ends will degrade the sensor performance, so you may have to compromise by only grounding one end of the shield. Again, which end is optimal can change based on several factors. You may have to experiement with which gets the best results.
If you have a two wire sensor, you should make sure that the two wires are tightly twisted together....the more twists per inch the better. If you have more than two, then you want to be sure that you are twisting the right two wires together (Signal+, Signal- or whatever the names are...should have a + and -, or sigal/return). Any loops in the wire, or physical separation between + and -will increase your suceptability.
If that doesn't do it, you could try a ferrite beed around the sensor wires where they enter the gauge. Sometimes they have these at automotive stores for filtering the ignition whine our of stereo wiring. These are typically not very effective, but combined with the over-braid might get you enough noise reduction. There are some other countermeasures, but each option gets progressively more complex, requires more diligent engineering analysis, and ultimately more trial-and-error.
You might also check your engine ground strap, and its connection to the engine block, and the chassis. A poor connection here could cause excess RF from the coil packs.
Ignition systems can be VERY noisy. The very nature of a spark is generator is to create broad-spectrum noise, with releatively high amplitude. You may very well have to combine some or all of the above measures. As I said above, physical separation is likely to be your most effective option because the RF engergy falls off quickly with distance.
If you absolutely can't move the harness, then I would try shielding the harness with tinned-copper overbraid. The overbraid MUST have a good connection to the chassis ground, to be effective. Use 14-16 AWG stranded wire, and keep it as short as possible to connect the shield to the chassis (less than 6 inches, 1 inch would be better). Make sure the chassis metal is absolutely clean and makes maximal suface area contact. If you have a ground stud on the gauge, that would be best. But, you still need to connect that ground stud on the gauge to the car structure in the same manner.
Typically, you will get best SHEILD performance with both ends grounded...but, that is somewhat variable. Often grounding both ends will degrade the sensor performance, so you may have to compromise by only grounding one end of the shield. Again, which end is optimal can change based on several factors. You may have to experiement with which gets the best results.
If you have a two wire sensor, you should make sure that the two wires are tightly twisted together....the more twists per inch the better. If you have more than two, then you want to be sure that you are twisting the right two wires together (Signal+, Signal- or whatever the names are...should have a + and -, or sigal/return). Any loops in the wire, or physical separation between + and -will increase your suceptability.
If that doesn't do it, you could try a ferrite beed around the sensor wires where they enter the gauge. Sometimes they have these at automotive stores for filtering the ignition whine our of stereo wiring. These are typically not very effective, but combined with the over-braid might get you enough noise reduction. There are some other countermeasures, but each option gets progressively more complex, requires more diligent engineering analysis, and ultimately more trial-and-error.
You might also check your engine ground strap, and its connection to the engine block, and the chassis. A poor connection here could cause excess RF from the coil packs.
Ignition systems can be VERY noisy. The very nature of a spark is generator is to create broad-spectrum noise, with releatively high amplitude. You may very well have to combine some or all of the above measures. As I said above, physical separation is likely to be your most effective option because the RF engergy falls off quickly with distance.
-tch
Build: www.tomhampton.info
video: vimeo.com/tomhampton
Support: X-Factor Racing
I didn't lose, I just got outspent!
#4
Posted 03-06-2012 01:04 PM
PS: excess harness length is not your friend. I don't know how easy it is to shorten, but loops of cable are nothing but a giant antenna/transformer designed to maximize your sensitivity.
-tch
Build: www.tomhampton.info
video: vimeo.com/tomhampton
Support: X-Factor Racing
I didn't lose, I just got outspent!
#5
Posted 03-06-2012 01:07 PM
PSS: if you have to coil it up, at least coil it so that you have equal numbers of loops going in opposite directions (clockwise and counter-clockwise). Its not perfect, but its better than the alternative.
Sorry about the multiple posts, my browser doesn't like the post-modification editor.
Sorry about the multiple posts, my browser doesn't like the post-modification editor.
-tch
Build: www.tomhampton.info
video: vimeo.com/tomhampton
Support: X-Factor Racing
I didn't lose, I just got outspent!
#6
Posted 03-06-2012 03:18 PM
My cable is run over the top of the transmission, and then up along the chassis to the hole for the AC line (passenger side firewall). Extra wire is coiled indoors on the passenger side. Works nicely.
#7
Posted 03-06-2012 08:23 PM
Couple of additional tricks to try: Tie a couple simple knots in the line which creates a slight inductance coil which MIGHT be enough to knock down the noise. A little more complicated is wind it around a pencil several times, which does the same thing - you can add more by making more windings. This trick is worth a try since it is both simple to do and SM friendly (i.e., free).
Good luck, R1chy
Good luck, R1chy
#8
Posted 03-07-2012 07:49 AM
I use a Autometer A/F gauge as well and have no RF issues. Mine runs into the cockpit just like Dave Lewis describes. I did put a heat shield tube around the wire near the header but I doubt it is doing anything for RFI. I wonder if it is a gauge issue. I know that will be a tough sell with the Autometer folks.
#9
Posted 03-07-2012 09:55 AM
It has to be a gauge issue. Funny thing is the fuel pressure gauge I got too just reads zero and it's all wired up correctly. The autometer tech said to send them to him personally and he will test them on his own car! Pretty amazing customer service to say the least. This is the first time I have ever had issues with autometer. My electric water and oil and volt gauges worked fine.
2010, 2011 SM Sowdiv points champion!
2011 National points champion!
WWW.MOTORSPORTHEAVEN.COM
2011 National points champion!
WWW.MOTORSPORTHEAVEN.COM
0 user(s) are reading this topic
0 members, 0 guests, 0 anonymous users