Exhaust Gas Analyser for home use ?

[ghost123uk]

Can anyone suggest an Exhaust Gas Analyser that is more aimed at the home mechanic rather than the big expensive garage equipment types ? (or is this summat that is not really a sensible option ?)

I have looked at the Gunson Gastester Digital Exhaust Analyser - G4125 as per HERE on Ebay, but not sure if it will do what I (we) want, ie help getting a proper setting on an open loop BLOS / OMVL R90 setup ? (on a VW T25 by the way). I know I could just take it to an LPG installer, but I prefer to DIY stuff, that way I know it is right (having had varying degrees of satisfaction at garages re this subject !)

EDIT = just been looking harder at that Gunson one ^^^ and it seems it does not have the facilities we need, ie no CO2 measurement or Lambda, I thought it was too cheap to be any use !

[Gilbertd]

Why do you need CO2 or lambda? CO and HC are the only important ones.

[Tubbs]

Just go and ask an mot station for the use of theirs for 10 mins. You will get a lot more info out of a proper one. Hence why they are so god damn expensive.

[rossko]

Those little CO meters will allow you to do static setup on a carb, which is all an open loop LPG system is.

[ghost123uk]

Why do you need CO2 or lambda? CO and HC are the only important ones.

Ah, ya see I am still learning I did have a Tinley Tech Closed loop with the 3 LED readout thingy (for Lamda) on the other van and I got to understand how it all worked pretty well. That 3 LED Lambda readout is what has led my to believe that it was fairly important

So, it's not. Well, is there a CO / CO2 / HC set of numbers I should be looking for ?

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

By the way, I got ok at tuning by ear / performance when the open loop mixer is/was in place. My settings turned out to not at all far from wrong when I took it to Flintshire Auto Gas for them to check it. However, setting up the BLOS is new ground for me.

Just go and ask an mot station for the use of theirs for 10 mins. You will get a lot more info out of a proper one. Hence why they are so god damn expensive.

True, but as you can see, I am wishing to learn and I cannot really spend much time doing that at the local place. Plus, as mentioned above, I was misunderstanding the figures somewhat. I was trying to get the CO as low as poss, the CO2 high, while remaining as close to Lambda 1 as I could. (this was when Flintshire Auto Gas let me use their proper analyser for 1/2 an hour.

Those little CO meters will allow you to do static set-up on a carb, which is all an open loop LPG system is.

Ummm, I think I need more than that though, I think, or do I

Now as you know about more about what I am trying to do (ie set-up but learn as well and also be able to do it again if I swap summat around, which I am prone to) what would you recommend ?

Or would you say "forget it, get it set up by an installer and leave it alone"

[ghost123uk]

OK, I have found a local garage, the smaller out of the way "old school" type that specialises in classic vehicles. I had a word with them and they are happy for me to use their exhaust gas analyser for a small charge.

Question = could anyone tell me what (CO/CO2/HC etc) figures I should be aiming for (open loop with BLOS) ?


Plus, from a previous comment above from Gilbertd, (and I know he knows his stuff from reading a lot of his posts on here ) I take it that the Lambda reading, is not so important ? (which from my experiments with closed loop, now ditched, I always thought it was )

[rossko]

The lambda reading is useful, but not necessary i.e. if choosing a cheap analyser, you don't need to pay extra for it. Lambda is calculated from other readings, not directly measured. If you do have it, then tuning in the "1.0" area is about right.

What does your engine manufacturer say for a target CO? For a carb era engine, which is also inefficient in other ways, it is expected that a good tune for general use will be inefficient enough to make CO - usually in the 0.5% league. You might think you'd tune for minimum but that usually isn't the best compromise for driveability.

Likewise you can expect to see HC with no cats. If there's none, you are probably lean. If there's lots, you might be rich, or way too lean, or misfiring. The O2 output is likewise useful for diagnostic, but not really tuning, purposes.

The very best thing you can do is buy your man a couple of beers and go through it with him advising you. It's only a simplistic carb setup after all. The only thing that is appreciably different about LPG is a lower CO2 output, and that's not used for tuning anyway. You'll be able to tell him what the LPG specific twiddles do - the idle bypass, the reducer 'balance' (effectively output pressure), the Blos adjuster. He'll understand how an SU carb works, and soon grasp it.

[ghost123uk]

Thanks for that rossko. It all helps me understand, like =

Lambda is calculated from other readings, not directly measured. If you do have it, then tuning in the "1.0" area is about right.

I wondered about that. I figured the Lambda reading was perhaps obtained by the EGA blowing gasses over a heated sensor inside the actual unit

What does your engine manufacturer say for a target CO?

Ah, it varies then I must look that up ! The engine is a good condition VW water cooled 1.9l flat four on a Pierburgh carb, all running well (with no inlet or exhaust leaks) in a VWT25 Tin-Top Camper (which is in ace nick). I was working on the premise that the CO should be as low as is achievable

There are a lot of "oops" emoticans above aren't there

So am I right in thinking then, that basically, when setting up the mixture, the CO is the main reading the operative looks at and it should be as near to 0.5% (or whatever is appropriate for this engine) across the rev range as can be achieved ? Also thanks for the explanation re HCs, but what HC readings would one be aiming for ?

[rossko]

HC represents unburnt (wasted) fuel, so generally low is good.
Bear in mind that even with modern direct doodah super-doopa engines, there would be HC if there were not cats to eat it up. No engine is perfect.
Bear in mind that there are regions in tuning where you deliberately add excess fuel - throttle pump, full power - so you would expect elevated HC at those times, and it lingers around too.
There is no magic answer. On a well used carb engine, if you are getting HC in the low hundreds ppm you're probably doing well. If its in the thousands its probably sick.
It's not much use a tuning aid.

CO is about partly combusted fuel - again its waste, but generated as a result of conditions nearly being right. As such, its more responsive to tweaking to get things just right.

Work with your classic car man, see what he's interested in. This is about skills and experience, not dialling a number. I'm an olde farte but struggle with open-loop tuning because I wasn't doing this twenty years ago, when these systems were the norm.

Without a rolling road, there's a lot of estimation (guesswork). It is desireable for example to setup slightly lean in the cruise for economy, but difficult to achieve when you also want rich under full load. You don't necessarily have all the adjustments to make it so.

[ghost123uk]

Aye, once again thanks for the info

So there is no "try and get your CO to as near as xxx across the board whilst keeping your HC below xxx" etc then, That's a shame. T'would be easier for me to diy this if it was that simple.

Without a rolling road, there's a lot of estimation (guesswork). It is desirable for example to setup slightly lean in the cruise for economy, but difficult to achieve when you also want rich under full load. You don't necessarily have all the adjustments to make it so.

Yes, I always wondered about that. Holding 3,000 rpm under no load to set mixture is not representative of the throttle opening v revs when doing 70mph on the motorway is it !

It was for that reason that I bought the closed loop upgrade kit from TT, but I found the lag on the stepper was so great that under a lot of circumstances the mixture was wildly out, most importantly when under hard acceleration and when pulling away briskly from a standstill (traffic lights etc) For this reason I sold it on and got the BLOS. I like the BLOS, it is certainly more responsive than either open loop with mixer ring, or mixer ring plus closed loop kit.

I'm just not confident it's "right" at the higher end and was hoping to have some figures to dial in. I am told that the BLOS is pretty much independent of the bias adjuster on the vap, apparently because the BLOS "sucks" harder due to the variable venturi. But this leaves me not sure where to set that bias adjuster (yes, the original setting has been lost, and not by me )

My only problem with the guy at the garage doing it is, he did say I could use the EGA, but he also said "it's no good asking me about mixture settings for LPG though, I know nowt about LPG at all" So we are both a bit in the dark on that aspect I could hook it all up and stare at the numbers, now knowing what they represent (thanks to your help ), but not knowing what they should read

[Gilbertd]

It was for that reason that I bought the closed loop upgrade kit from TT, but I found the lag on the stepper was so great that under a lot of circumstances the mixture was wildly out

The idea of a lambda feedback system is that it fine tunes the mixture to keep it spot on, not that it adjusts it so it gets it right all the time. You have to set the mixer and vaporiser as if you are setting an open loop system in the first place so any lag on the stepper has minimal impact.,

[ghost123uk]

It was for that reason that I bought the closed loop upgrade kit from TT, but I found the lag on the stepper was so great that under a lot of circumstances the mixture was wildly out

The idea of a lambda feedback system is that it fine tunes the mixture to keep it spot on, not that it adjusts it so it gets it right all the time. You have to set the mixer and vaporiser as if you are setting an open loop system in the first place so any lag on the stepper has minimal impact.,

Yes I did all that. I had the closed loop fitted to a well set up vap and mixer but I was never happy with it's performance during acceleration from a standstill, the van felt sluggish and the indicator showed lean (up until a steady speed), no matter what settings I tried. Spot on during cruise though. Some other folk over on the VW T25 forum found the same. Anyway it's academic now because it's gone and I have the BLOS.



I spent most of yesterday afternoon researching setting LPG mixtures. I found some useful info including some graphs explaining the relationship between CO/CO2/HC/O2 and AFR. I learned that there is no such thing as a low priced EGA that is any use. I learned that a simple (Gunson et al) CO meter does not provide sufficient info and one needs a 4 gas jobby to do it properly. So the title of this thread has gone to the wall

Re figures. What would this look like:- CO = 0.2% - CO2 = 13% - HC = 4ppm - O2 = 0% ?

These are not my results, they are, from what I can glean, pretty good figures for open loop LPG on a 70's carb engine in good nick at ~3,000 rpm, no load.

Once I have something to work to, I can go to my local garage and use the EGA to achieve figures as close as I can to the ideal.

Thanks for letting me pick your brains

John

[rossko]

I think those figures are unusually "clean" for that class of engine. I expect it is set up a tad lean - but that could be the desired result, see earlier. Piling on RPM usually helps, so I would expect to see something different at idle - more CO etc.

[Gilbertd]

CO shows how well the combustion is while HC shows unburnt Hydrocarbons. So if you had an air filter that was clogged solid, the mixture would be rich so the CO figure would be high. If you then removed the plug lead from one cylinder, you reduce the CO by 25% (assuming a 4 cylinder engine) but the unburnt fuel from the one non-firing cylinder would cause the HC to go through the roof. HC will also go up if you are burning a bit of oil too so a figure of 4ppm on anything other than a brand new engine is expecting far too much. Anything in the high tens, low hundreds would be normal for an older engine.

[ghost123uk]

a figure of 4ppm on anything other than a brand new engine is expecting far too much. Anything in the high tens, low hundreds would be normal for an older engine.

Good info. Thanks