Hello all,

My understanding now is that ALL digital A/F gauges (gauges with a numerical or LED readout) are wideband, and no number readout (which I want) gauges are available for narrowband.

Does anyone know of a A/F gauge that will mate to a3 or 4 wire narrowband sensor and present its readout in numbers? Hopefully in the $60 range but it takes what it takes.

Is it true narrowband ONLY does non-digital readouts? It was suggested to me look at the Autometer 3775 (digital but only wideband) and I liked its look until the narrowband/wideband issue arose. SO, I have looked at Autometer 5775 ...not digital, but costs $60 delivered.


Thanks
rock
'64d100

Rock,

I haven't seen any numerical NB O2 readouts. The reason is that the numbers don't actually help you that much. NB sensors swing very fast from "rich" to "lean" across 0.5V, and their curve looks like a titration curve. Thus, any voltage between 0.2 and 0.8 is almost meaningless (you are at stoich mix). Hard to get exact mixture anywhere on the curve.

I have used a Sunpro LED readout on two cars and like it a lot. $35-40 cost from various sources. Bought mine from Advance Auto locally.

If you need numbers, a high impedance digital voltmeter will work. I've plugged in my digital VOM before and it works fine.

Lou

_________________
Home of Slant6-powered fun machines since 1988

I use a DVM to tune fuel pressure at idle. The O2 sensor is stock GM 1 wire.

_________________
Yeah....Im the one who destroyed this rare, vintage automobile.....

A normal 02 sensor should swing back and forth across the middle or stoic range about 30 or more times a second. It might be slightly less at idle if it is getting cool, but the readings become less valid as well. It should perk right up as soon as the rpms and load come up. A sensor that does not is bad or going bad or mounted too far downstream. A heated sensor should not have this issue, but should still be mounted close to the action.

The readings with a DVM will depend on both the sampling rate and chance to some degree. As already pointed out, it is an average that needs to be interpreted, not a steady reading. None of the stand alone monitors I have seen show the cross counts, they really need to be watched as well to be sure the rich/lean readings are valid.

Thanks fellows,

GREAT info for me, taught me some stuff too. Truly, I was focused on the ideal ratio I have read (14.7) and how I could read it at hot idle for idle mix tuning. I did not realize the sampling process was actually akin to the problem I, as an old reel to reel audio mastering tape person encountered when the world went digital...that is, any alogorithm can most closely approximate a true curve if the sample number is infinite , and that there will be a sample numberpre unit time below which the data are not meaningful. (Analog tape, capturing a continuous waveform gives the best sound still, but those darn digital guys are pretty good too, just like in cameras!)

NOW I see why Lou and others here have found narrowband useful and I see its limitations in contrast to say, the Innovate wideband, which could also induce some interpretation problems for persons like me mainly interested in relatively static tuning rpm states.

Darn, now I have to go buy a book or two on EFI to see why or if wideband is a prerequisite for any EFI I may play with as time goes on, like Megasquirt. And learn ever more as a result of playing with slants.

Thanks again,
rock
'64d100

The HEGO switching frequency across stoich is a calibratable number in factory EFI systems. Absent a closed loop system, you cannot use a DVOM average from a narrow band sensor. You must use a wideband in any system if you want any accuracy at all. BTW, stoich for 10% ethanol blend - which is popular in many regions is 14.06.

Mitch

Nordskog makes a couple of gauges that give a numerical readout with a narrow band sensor, but I have not heard good things about their accuracy (Strangely enough, they also build some of the Innovate gauges, but those gauges are driven by a signal from Innovate's circuitry). The gauge can't be more accurate than its sensor, and here's what a narrow band sensor response curve looks like compared to the signal from a wideband controller:

http://www.megamanual.com/v22manual/wbvolts.gif

The split at the top is curves at two different exhaust temperatures. As you can see, it's good for telling if you're at 14.7:1, not so useful for anything else.
Actually, a lot of tuning on modern EFI systems consists of holding the engine at a steady RPM and load with a wideband in the exhaust while adjusting the mixture.

_________________
"Mad Scientist" Matt Cramer
'66 Dart - turbocharged 225
My blog - Mad Scientist Matt's Lair