Prev: Capcitor selection value
Next: ADC techniques, was Re: Diode-Connected Transistors
From: John Fields on 22 Jul 2010 18:11
On Thu, 22 Jul 2010 19:54:27 +0100, "Nial Stewart"
<nial*REMOVE_THIS*@nialstewartdevelopments.co.uk> wrote:

>I am afflicted with the love and ownership of a classic Mini...
>
>http://pimz.com/site/data/portfoliopix/classic_mini.jpg
>
>These were fuelled with SU carbs (until the last few years when Rover
>developed a reasonable solution to fuel injecting siamesed input ports).
>
>The SU uses a tapered needle to regulate fuelling, the needle has to have
>the correct profile along its length. The needle in my car almost certainly
>isn't correct, it always drives better with the choke out.
>
>I have been thinking of making a rig so I can properly measure & log the A/F ratio
>against the carb piston lift (the needle protrudes from the bottom of the
>piston into the jet). This should allow an optimum profile to be deduced
>and the correct needle produced.
>
>
>I can use one of these to measure the A/F ratio....
>
>http://www.innovatemotorsports.com/products/lc1.php
>
>..but need a method of measuring the piston lift _fairly_ accurately and
>without loading the piston too much (the lift is regulated by a spring
>and too much extra weight on the piston would affect this).
>
>I can mount things so that a vertical probe sits on top of the piston and
>be moved as the piston moves. (I'll try to get a picture tonight).
>
>I can used an LVDT but this is expensive and overly accurate for what I'm
>doing.
>
>Can anyone suggest any simpler/cheaper method of measuring this?

---
Wind a coil such that the probe attached to the piston moves axially
within the coil, without touching it.

That'll give you an inductance change which will vary with piston
position, and you can use that inductance change to vary the frequency
of an oscillator.

JF

From: m II on 22 Jul 2010 18:48
Nial Stewart wrote:

> I am afflicted with the love and ownership of a classic Mini...


I have a couple of 2 inch SU carbs in the garage. They're off a 1974
Jag. AKA DriveWay Queen. They're somewhat larger than the Mini version
but work the same way.

The tops of the free floating pistons have a vacuum applied to them via
a hole to the venturi. That vacuum is what allows the pistons to rise,
as the surface area being acted on is much greater than that of the part
in the carb bore.

I don't know how you could actuate a sensor via this piston without
introducing a leak into the upper chamber.

You should be able to check for correct mixture by manually raising the
piston about 1/8 to 1/4 inch. The engine should die when it is raised.
Many SU models have a small small plunger on the side of the throttle
body for just that purpose.

The carb itself is one of the neatest devices ever invented and when set
up correctly, work very well.

Keep in mind that the weight of oil in the damper rod cylinder affects
mixture also. The heavier the oil, the richer it runs. I think I used
something like 2 or 5 weight oil...like sewing machine lubricant. Flush
all used oil and fill up to the top with new oil. The extra stuff will
be forced out when you replace the damper piston.

The needle jet is usually adjustable up and down the needle. In fact,
the choke (enrichener, actually) works by lowering the jet on the needle
taper.

There may be a variation here, as the 1 inch or so SU is unfamiliar to me.

Later model Strombergs had a rubberized diaphragm sealing the floating
piston. They were cheaper to make, as they didn't need the same precise
machining as the SU.






mike

From: Grant on 22 Jul 2010 22:09
On Thu, 22 Jul 2010 17:11:00 -0500, John Fields <jfields(a)austininstruments.com> wrote:

>On Thu, 22 Jul 2010 19:54:27 +0100, "Nial Stewart"
><nial*REMOVE_THIS*@nialstewartdevelopments.co.uk> wrote:
>
>>I am afflicted with the love and ownership of a classic Mini...
>>
>>http://pimz.com/site/data/portfoliopix/classic_mini.jpg
>>
>>These were fuelled with SU carbs (until the last few years when Rover
>>developed a reasonable solution to fuel injecting siamesed input ports).
>>
>>The SU uses a tapered needle to regulate fuelling, the needle has to have
>>the correct profile along its length. The needle in my car almost certainly
>>isn't correct, it always drives better with the choke out.
>>
>>I have been thinking of making a rig so I can properly measure & log the A/F ratio
>>against the carb piston lift (the needle protrudes from the bottom of the
>>piston into the jet). This should allow an optimum profile to be deduced
>>and the correct needle produced.
>>
>>
>>I can use one of these to measure the A/F ratio....
>>
>>http://www.innovatemotorsports.com/products/lc1.php
>>
>>..but need a method of measuring the piston lift _fairly_ accurately and
>>without loading the piston too much (the lift is regulated by a spring
>>and too much extra weight on the piston would affect this).
>>
>>I can mount things so that a vertical probe sits on top of the piston and
>>be moved as the piston moves. (I'll try to get a picture tonight).
>>
>>I can used an LVDT but this is expensive and overly accurate for what I'm
>>doing.
>>
>>Can anyone suggest any simpler/cheaper method of measuring this?
>
>---
>Wind a coil such that the probe attached to the piston moves axially
>within the coil, without touching it.
>
>That'll give you an inductance change which will vary with piston
>position, and you can use that inductance change to vary the frequency
>of an oscillator.

Wind three coils and you have an LVDT where you measure relative
signal pickup and no longer care about absolute inductance. Drive
coil is in between the two secondaries. Winding all that nice and
evenly is what drives up the cost of an LVDT.

Moving core is ferrite, I think, drive frequency about 1kHz when
I was doing signal conditioners for them a couple decades ago.

Grant.
From: Spehro Pefhany on 22 Jul 2010 23:33
On Fri, 23 Jul 2010 12:09:34 +1000, the renowned Grant
<omg(a)grrr.id.au> wrote:

>On Thu, 22 Jul 2010 17:11:00 -0500, John Fields <jfields(a)austininstruments.com> wrote:
>
>>On Thu, 22 Jul 2010 19:54:27 +0100, "Nial Stewart"
>><nial*REMOVE_THIS*@nialstewartdevelopments.co.uk> wrote:
>>
>>>I am afflicted with the love and ownership of a classic Mini...
>>>
>>>http://pimz.com/site/data/portfoliopix/classic_mini.jpg
>>>
>>>These were fuelled with SU carbs (until the last few years when Rover
>>>developed a reasonable solution to fuel injecting siamesed input ports).
>>>
>>>The SU uses a tapered needle to regulate fuelling, the needle has to have
>>>the correct profile along its length. The needle in my car almost certainly
>>>isn't correct, it always drives better with the choke out.
>>>
>>>I have been thinking of making a rig so I can properly measure & log the A/F ratio
>>>against the carb piston lift (the needle protrudes from the bottom of the
>>>piston into the jet). This should allow an optimum profile to be deduced
>>>and the correct needle produced.
>>>
>>>
>>>I can use one of these to measure the A/F ratio....
>>>
>>>http://www.innovatemotorsports.com/products/lc1.php
>>>
>>>..but need a method of measuring the piston lift _fairly_ accurately and
>>>without loading the piston too much (the lift is regulated by a spring
>>>and too much extra weight on the piston would affect this).
>>>
>>>I can mount things so that a vertical probe sits on top of the piston and
>>>be moved as the piston moves. (I'll try to get a picture tonight).
>>>
>>>I can used an LVDT but this is expensive and overly accurate for what I'm
>>>doing.
>>>
>>>Can anyone suggest any simpler/cheaper method of measuring this?
>>
>>---
>>Wind a coil such that the probe attached to the piston moves axially
>>within the coil, without touching it.
>>
>>That'll give you an inductance change which will vary with piston
>>position, and you can use that inductance change to vary the frequency
>>of an oscillator.
>
>Wind three coils and you have an LVDT where you measure relative
>signal pickup and no longer care about absolute inductance. Drive
>coil is in between the two secondaries. Winding all that nice and
>evenly is what drives up the cost of an LVDT.
>
>Moving core is ferrite, I think, drive frequency about 1kHz when
>I was doing signal conditioners for them a couple decades ago.
>
>Grant.

High mu nickel-iron alloy. They're typically designed for a certain
drive frequency (eg. 2.5kHz) such that the phase shift is zero.

I think you can make a pretty crappy one and it'll still be good to a
percent of so.



Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff(a)interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com
From: Grant on 23 Jul 2010 01:53
On Thu, 22 Jul 2010 23:33:55 -0400, Spehro Pefhany <speffSNIP(a)interlogDOTyou.knowwhat> wrote:

>On Fri, 23 Jul 2010 12:09:34 +1000, the renowned Grant
><omg(a)grrr.id.au> wrote:
>
>>On Thu, 22 Jul 2010 17:11:00 -0500, John Fields <jfields(a)austininstruments.com> wrote:
>>
>>>On Thu, 22 Jul 2010 19:54:27 +0100, "Nial Stewart"
>>><nial*REMOVE_THIS*@nialstewartdevelopments.co.uk> wrote:
>>>
>>>>I am afflicted with the love and ownership of a classic Mini...
>>>>
>>>>http://pimz.com/site/data/portfoliopix/classic_mini.jpg
>>>>
>>>>These were fuelled with SU carbs (until the last few years when Rover
>>>>developed a reasonable solution to fuel injecting siamesed input ports).
>>>>
>>>>The SU uses a tapered needle to regulate fuelling, the needle has to have
>>>>the correct profile along its length. The needle in my car almost certainly
>>>>isn't correct, it always drives better with the choke out.
>>>>
>>>>I have been thinking of making a rig so I can properly measure & log the A/F ratio
>>>>against the carb piston lift (the needle protrudes from the bottom of the
>>>>piston into the jet). This should allow an optimum profile to be deduced
>>>>and the correct needle produced.
>>>>
>>>>
>>>>I can use one of these to measure the A/F ratio....
>>>>
>>>>http://www.innovatemotorsports.com/products/lc1.php
>>>>
>>>>..but need a method of measuring the piston lift _fairly_ accurately and
>>>>without loading the piston too much (the lift is regulated by a spring
>>>>and too much extra weight on the piston would affect this).
>>>>
>>>>I can mount things so that a vertical probe sits on top of the piston and
>>>>be moved as the piston moves. (I'll try to get a picture tonight).
>>>>
>>>>I can used an LVDT but this is expensive and overly accurate for what I'm
>>>>doing.
>>>>
>>>>Can anyone suggest any simpler/cheaper method of measuring this?
>>>
>>>---
>>>Wind a coil such that the probe attached to the piston moves axially
>>>within the coil, without touching it.
>>>
>>>That'll give you an inductance change which will vary with piston
>>>position, and you can use that inductance change to vary the frequency
>>>of an oscillator.
>>
>>Wind three coils and you have an LVDT where you measure relative
>>signal pickup and no longer care about absolute inductance. Drive
>>coil is in between the two secondaries. Winding all that nice and
>>evenly is what drives up the cost of an LVDT.
>>
>>Moving core is ferrite, I think, drive frequency about 1kHz when
>>I was doing signal conditioners for them a couple decades ago.
>>
>>Grant.
>
>High mu nickel-iron alloy. They're typically designed for a certain
>drive frequency (eg. 2.5kHz) such that the phase shift is zero.

That sounds better, 'cos the core was tapped to fit onto non-magnetic
rod that connected to the whatsit being measured. Or spring loaded,
many options.

Though I remember the phase being differently expressed, Think we
use to zero the thing by adjusting phase control of the sync rectifier
with no core inside, then add core to the zero position, then do the
displacement each way to calibrate +/- span.
>
>I think you can make a pretty crappy one and it'll still be good to a
>percent of so.

Not much in them, lots of fine wire though, one would need plenty of
patience to wind one. I suppose with fast opamps one could use less
turns, run at much higher frequency and still get decent signals?

Grant.
First  |  Prev  |  Next  |  Last
Pages: 1 2 3 4
Prev: Capcitor selection value
Next: ADC techniques, was Re: Diode-Connected Transistors