Tuning/setup the LPG system with no gas analyser
Adam Coombes
recourse to a gas analyser? Is it possible to tune by ear like a carb
petrol system? I've got some instructions here before (from Mr Shackles
I think) but they all need an exhaust probe which I don't have access to
and can't afford to buy. Could I jury rig something with a lambda sensor
shoved up the pipe (it's a kit with straight through exhausts and open
loop system)?
Also, as it's LPG only on a 10.5:1 CR Rover V8, any ideas what the
optimum ignition advance might be?
Many thanks,
Adam.
--
Posted via Mailgate.ORG Server - http://www.Mailgate.ORG
Badger
Adam Coombes <acoo...@totalise.co.uk> wrote in message
news:89bca69bb2daa340c9d...@mygate.mailgate.org...
that you get the same rpm for a given throttle position on both petrol and
gas. Try at approx. 3000rpm first, adjusting big screw in gas feed hose,
then at idle adjusting mixture screw on vapouriser. Not perfect, but it'll
get it going to take it to a garage that has an analyser, as that's the ONLY
way to get an accurate set-up. If you can't get it to run on petrol at all
for whatever reasons, then sorry, you're on your own. You could always buy
one of those Gunson's Gastesters at your local Halfrauds.....? Then you've
always got it for routine checking.
As for timing, try around 9-12 degrees BTDC@idle, vacuum pipe to dizzy
disconnected and plugged whilst setting. Good luck.
Badger.
Stewart Hargrave
Coombes:
>Can anyone give me a good guide to setting up my LPG system without
>recourse to a gas analyser? Is it possible to tune by ear like a carb
>petrol system? I've got some instructions here before (from Mr Shackles
>I think) but they all need an exhaust probe which I don't have access to
>and can't afford to buy. Could I jury rig something with a lambda sensor
>shoved up the pipe (it's a kit with straight through exhausts and open
>loop system)?
I tune mine by ear. Probably not quite as spot-on accurate as it would
be with a meter, but good enough for me (and the emissions test).
With the engine fully warmed up and at idle, twiddle about with the
'mixture' screw on the vapouriser - slowly unscrew it to enrich the
mixture until it begins to run lumpily. Now screw it the other way to
weaken it until the engine starts to lose revs and run unevenly. Don't
rush it, as it may take a few seconds for the engine to respond and
settle down to your adjustments - take your time and find these two
spots carefully. Aim to set it midway between the two. It's much like
setting the mixture on a carburettor. After this, you may need to
adjust the idle speed control at the throttle body to adjust the idle
revs. If you do this, go back and re-check the mixture setting.
Now move to the flow control in the pipe between the vap and the
mixer. Screw this down to about half closed. Now rev the engine to
about half its maximum and fix the throttle there (a screwdriver
wedged in the linkage does it for me). Screw the valve in further
until the revs pick up a little. Screw past this point and then back
off to find it's maximum again. This should now be set at the optimum
for maximum power - too far out and it will be too rich up the rev
range, leading to flat spots. Some people suggest screwing it in a
turn or two for better economy.
>Also, as it's LPG only on a 10.5:1 CR Rover V8, any ideas what the
>optimum ignition advance might be?
Nope. I've experimented with my Volvo up to about 8ยบ more advance than
standard. TBH I can't tell any difference at all (we are talking a fat
old Volvo here, not a light and responsive sports car). LPG is slower
burning so should benefit from an earlier ignition point. I guess a
dyno would give you the answer. Given that over advancing will quickly
lead to inefficiency I tend to err on the conservative side. Also,
LPG's higher octane rating may disguise the point at which you over do
it - petrol will start to detonate sooner which would be an audible
warning of over advancement.
Be aware also that adjusting the timing may require re-setting the
mixture.
--
Stewart Hargrave
A lot faster than public transport
For email, replace 'SpamOnlyToHere' with my name
Austin Shackles
<acoo...@totalise.co.uk> enlightened us thusly:
>Can anyone give me a good guide to setting up my LPG system without
>recourse to a gas analyser? Is it possible to tune by ear like a carb
>petrol system? I've got some instructions here before (from Mr Shackles
>I think) but they all need an exhaust probe which I don't have access to
>and can't afford to buy. Could I jury rig something with a lambda sensor
>shoved up the pipe (it's a kit with straight through exhausts and open
>loop system)?
>
>Also, as it's LPG only on a 10.5:1 CR Rover V8, any ideas what the
>optimum ignition advance might be?
snap!
so's mine.
what system? I've made a little text file about tuning open loop systems
using the big OMVL vapouriser, which you're welcome to if you want -
different vapouriser may have different adjustments though.
Ignition timing is difficult, as I don't know what mine's set to... but 3
degrees more advance than the 4* setting or 6 degrees more than the unleaded
setting is a good starting point. keep advancing it slightly until it
pinks, then back it off a bit :-)
--
Austin Shackles. www.ddol-las.fsnet.co.uk my opinions are just that
"Quos deus vult perdere, prius dementat" Euripedes, quoted in
Boswell's "Johnson".
Adam Coombes
> what system? I've made a little text file about tuning open loop systems
> using the big OMVL vapouriser, which you're welcome to if you want -
> different vapouriser may have different adjustments though.
>
> Ignition timing is difficult, as I don't know what mine's set to... but 3
> degrees more advance than the 4* setting or 6 degrees more than the unleaded
> setting is a good starting point. keep advancing it slightly until it
> pinks, then back it off a bit :-)
Thanks Austin. It's the same big OMVL jobbie I suspect. I've still got
your original instructions on file, but it's proving difficult to get
access to an exhaust analyser and I can't justify buying one just for
the occasional home use so the text file would be dead handy.
Cheers,
Stewart Hargrave
Shackles:
>keep advancing it slightly until it
>pinks, then back it off a bit :-)
This is not necessarily a good guide. I've been thinking about it:-
With LPG's higher octane rating the point at which pinking occurs is
further advanced from the point at which inefficiency occurs. If you
create pinking you will be into inefficiency territory. In a well
designed engine, the two points will be near-coincidental. However we
are adapting an engine that was designed for petrol with a lower
octane rating and a faster burn rate than LPG, so there will be a
discrepancy.
But it does give me the opportunity to expound upon the finer details
of the combustion process:
Pinking as a result of over advancing the ignition occurs when the
huge increase in pressure due to combustion cannot be mitigated
because the piston isn't able to descend fast enough - there's usually
about 15 degrees either side of TDC when the piston hardly moves, so
any thermal energy released during this time cannot do any useful
work.
If max pressure is reached too early the flame front will not have
reached all the fuel before the increasing pressure causes the
remaining unburnt fuel to self ignite (this is detonation, as opposed
to pre-ignition). With LPG's higher octane rating it can withstand
greater pressure before this happens, so you could have lost
significant efficiency before pinking occurs.
Ignition timing is related to the speed at which the fuel burns inside
the combustion chamber, with the aim of developing max pressure at
that critical point around 15 degrees ATDC. The octane rating is
related to the compression ratio an engine can be designed at.
Sometimes people make the wrong link between ignition timing and a
fuel's octane rating. This misconception is created because we all
know that we have to retard the ignition if we use a lower octane
fuel. In fact what we are doing is creating max pressure at a point
where the piston is already descending, so the pressure cannot rise
high enough to cause detonation. This simply wastes the advantage of
compressing the charge as much as we did - we could have compressed it
less and sparked it earlier. Similarly, it does not follow that a
higher octane fuel will benefit from an earlier spark if this simply
results in more thermal energy being released before it can do any
useful work. Both are routes to inefficiency.
We can't really exploit LPG's higher octane without major engine
modifications. We can, however, try to elliminate any inefficiency
that is created by its slower burn rate by adjusting the ignition
timing. I guess a dyno test is the definitive way to establish where
the gain becomes inefficient.
--
Stewart Hargrave
A lot faster than Royal Mail
Adam Coombes
though. I'd always made that link between octance and ignition, but if
the fuel/air mix is detonating I assume you mean that the explosive
wavefron is becoming supersonic, ie flat at the front of the wave, which
would explain the noise. I seem to remember that a high-ex wavefront
like that goes for the course of greatest density rather than least
resistance, which would also explain the effect on the engine if you
don't stop it! What does it mean in real terms though? Does anyone know
just how much slower the flame front of LPG is than petrol and can we
make a pragmatic guess at the optimum additional advance that will yield
greater efficiency given a maximum rev range of say 6krpm? Presumably
you'd need to work out the piston speed/accleration for the given
engine? Is it worth it?
Stewart Hargrave
Coombes:
>OK, that all makes perfect sense. I see what you mean about the pinking
>though. I'd always made that link between octance and ignition, but if
>the fuel/air mix is detonating I assume you mean that the explosive
>wavefron is becoming supersonic, ie flat at the front of the wave, which
>would explain the noise.
I need to scratch my head for a while before I can figure whether the
wave front is the same as the flame front. But the flame front of the
detonating fuel is indeed travelling at supersonic speeds. It is a
true explosion rather than a rapid but progressive burn.
> I seem to remember that a high-ex wavefront
>like that goes for the course of greatest density rather than least
>resistance, which would also explain the effect on the engine if you
>don't stop it!
Partly. But things get really interesting on a molecular scale here.
The thermal energy of the detonated charge is released so rapidly that
the molecules are, briefly, hurling around with an aweful lot of
energy in a confined space. Enough to dislodge the 'boudary layer' of
normally static molecuels trapped on the imperfectly smooth surface of
the combustion chamber. This boundary layer plays an important part in
protecting the engine by deflecting and absorbing the energy imparted
upon it by the thermally active molecules. Once this boundary layer
becomes dislodged, larger amounts of thermal energy can be imparted
directly upon the engine surfaces, rather than used usefully or
chucked down the exhaust pipe. This is one reason why a detonating
engine will rapidly overheat, and eventually burn and blow holes in
the piston top, or anything else that isn't effectively cooled.
>What does it mean in real terms though?
Ah yes; bring me back to reality. Enough of this highfalutin stuff -
I've been reading too many text books.
>Does anyone know
>just how much slower the flame front of LPG is than petrol and can we
>make a pragmatic guess at the optimum additional advance that will yield
>greater efficiency given a maximum rev range of say 6krpm?
This is indeed the kernal of the matter.
>Presumably
>you'd need to work out the piston speed/accleration for the given
>engine?
And there are other factors as well. Like the turbulance of the
compressed mixture, which increases as engine speed increases,
affecting the overall burn speed of the charge. We would probably need
to make assumptions about things like this, which are not necessarily
linear.
>Is it worth it?
Almost certainly not. I set my ignition around 3-5 degrees further
advanced than normal. I can't tell the difference, but it makes me
feel better.
Adam Coombes
message news:r9l8qvoica921q5bh...@4ax.com
>
> >Is it worth it?
>
> Almost certainly not. I set my ignition around 3-5 degrees further
> advanced than normal. I can't tell the difference, but it makes me
> feel better.
Thanks all in this thread.
Adam
(Just enough knowledge to be dangerous)
AndyC the WB
Austin> what system? I've made a little text file about tuning
Austin> open loop systems using the big OMVL vapouriser, which
Austin> you're welcome to if you want - different vapouriser may
Austin> have different adjustments though.
I'd be interested to have a read of that... if you can sent it
along...
AndyC
--
+-------------------------------------------------------------------------+
|Andy Cunningham aka AndyC the WB | andy -at- cunningham.me.uk |
+-------------------------------------------------------------------------+
| http://www.p38a.co.uk - Everything you wanted to know |
| about the P38A Range Rover but were afraid to ask. |
+-------------------------------------------------------------------------+
"And everything we want to get/We download from the Internet
All we hear is/Internet Ga-Ga/Cyberspace Goo-goo"
-- from "Radio Ga Ga"/"We will rock you"
Adam Coombes
http://www.acl.com.au/web/acl00056.nsf/0/359683e8a538a3e64a2566c0007bb33e?OpenDocument
which seems to indicate that more advance is required for low revs, but
actually less advance than petrol is required at higher revs. I guess
this has something to do with the 'extra factors' such as turbulence
mentioned earlier in this thread?
Stewart Hargrave
Coombes:
>On the subject of different behaviour at different revs, I found this>>
>
>
>http://www.acl.com.au/web/acl00056.nsf/0/359683e8a538a3e64a2566c0007bb33e?OpenDocument
>
>which seems to indicate that more advance is required for low revs, but
>actually less advance than petrol is required at higher revs. I guess
>this has something to do with the 'extra factors' such as turbulence
>mentioned earlier in this thread?
That's interesting, and makes assumptions about resetting the ignition
timing even harder. Maybe even pointless.
It always amazes me (and by turns often pleases and frustrates me) to
find that the more stuff we find out about, the more we find there is
more stuff to find out about.
--
Stewart Hargrave
I run on beans - laser beans
Austin Shackles
Hargrave<SpamOnl...@MiserableOldGit.Me.uk> enlightened us thusly:
>From a point at sea, to the circles of your mind, this is Adam
>Coombes:
>
>>On the subject of different behaviour at different revs, I found this>>
>>
>>
>>http://www.acl.com.au/web/acl00056.nsf/0/359683e8a538a3e64a2566c0007bb33e?OpenDocument
>>
>>which seems to indicate that more advance is required for low revs, but
>>actually less advance than petrol is required at higher revs. I guess
>>this has something to do with the 'extra factors' such as turbulence
>>mentioned earlier in this thread?
>
>That's interesting, and makes assumptions about resetting the ignition
>timing even harder. Maybe even pointless.
this ties in with my experimental results - a little bit more advance helps,
but more advance doesn't gain performance overall. setting the Ford to 15
degrees (4* setting is 12) made it more responsive at low revs, but going
further, although it didn't pink, didn;t make it go any faster and made it
more problematic on petrol.
>
>It always amazes me (and by turns often pleases and frustrates me) to
>find that the more stuff we find out about, the more we find there is
>more stuff to find out about.
--
Austin Shackles. www.ddol-las.fsnet.co.uk my opinions are just that
0123456789112345678921234567893123456789412345678951234567896123456789712345
1 weebl: What's this? | in recognition of the fun that is weebl and bob
2 bob: it a SigRuler! | check out the weebl and bob archive:
3 weebl: How Handy! | http://www.weebl.jolt.co.uk/archives.php
aghasee
news:bnrq1m$ndb$1...@newsg2.svr.pol.co.uk...
<snip>
> Halfrauds
ROTFL
--
Pros
-1982 Range Ruster V8-
(E-mail address is forged - reply to group please)
aghasee
in some questions:
"athol" <m...@privacy.net> schreef in bericht
news:br8sfh$j3dh$1...@ID-215292.news.uni-berlin.de
<snip>
> Adjustable vacuum advance is desirable. I typically find that once
> you have the mechanical advance set up right, you don't need as much
> vac advance as with petrol. The Accel brand adjustable diaphram on
> the HEI came with an allen key. To adjust, you simply removed the
> vacuum hose and stuck the allen key in the hole...
On my old ruster (old type points distib.) I got the startup right, low revs
power is very good, but from 3500 rpm onwards it kinda goes to sleep. The
LPG unit is (again old) a Landi-Hartog made to work on the original -very
restrictive- inlet and airfilter combination. Could it be the power at
higher rev's drops away due to generating too much vacuum in the system when
i "give it some welly"? The engine slowly crawls into higher revs on
motorways if it's not being pushed too hard.
Maybe an allen key would be the solution for my system? :-)
aghasee
> aghasee <numbe...@center.court> wrote:
>
>> On my old ruster (old type points distib.) I got the startup right,
>> low revs power is very good, but from 3500 rpm onwards it kinda goes
>> to sleep. The LPG unit is (again old) a Landi-Hartog made to work on
>> the original -very restrictive- inlet and airfilter combination.
>> Could it be the power at higher rev's drops away due to generating
>> too much vacuum in the system when i "give it some welly"? The
>> engine slowly crawls into higher revs on motorways if it's not being
>> pushed too hard.
>
Ah! Swapping the old, worn out, low CR V8 with an as-good-as-new V8 out of a
Rover Vitesse was a bad idea, then? :-)
I had to fit the old timing case and inlet manifold to the new V8 as the
original items were kind of not there when I got it. (Come to think of it,
wasn't the Vitesse V8 a fuel injected engine?) Afaik the Vitesse engine
develops a lot more HP's than the low CR versions? (If it's getting the
right amount of fuel mixture, that is...)
> An undersized mixer will simply not be able to deliver the required
> flow for more than a certain amount of HP. The air cleaner, manifold
> and mixer combination sound like exactly that.
>
> Similarly, an undersized convertor will be unable to vapourise
> sufficient mass of LPG without going unstable. If you pull too much
> LPG from a convertor, it simply won't be able to transfer sufficient
> heat out of the coolant into the LPG and you will start getting cold
> LPG, or, taken further, liquid LPG droplets in the vapour hose. Pull
> enough LPG and the converter can even start to freeze up as if it was
> low on coolant. Once it starts to freeze, there'll be lots of liquid
> droplets and the mixture will be rich enough to stall. Back off and
> the convertor thaws back out as long as it doesn't actually stall!
> :-)
And now that you mention temperature, it _is_ running a bit on the cool
side. A new thermostat was already planned to be fitted, maybe I will go
ahead with the electric fans while I'm at it...
>> Maybe an allen key would be the solution for my system? :-)
>
> The Vacuum advance should be taken from a ported vacuum port on the
> side of the throttlebody. This relies upon the air velocity along
> the wall of the thhrottlebody near the butterfly, giving little at
> closed throttle, maximum at mid throttle and little at WOT.
This is how it's done on my engine. The LPG thinghy is taking it's vacuum
from the center of the manifold.
> If the vacuum advance is connected to the wrong place, put it on the
> throttle body where it belongs.
>
> I doubt that fiddling with vacuum advance will help you much.
>
> A cold air intake duct from outside the engine bay would be a cheap
> way to make some difference...
I'll go and get me some flexible hose then, now the air is being taken from
directly above the nice and shiny tubular SS exhaust manifolds...