fuel level sender for aircraft

[Franc Zabkar]

I have a fuel level sender and LED gauge from a privately built
aircraft. It uses a strange implementation of an LM3914 dot/bargraph
LED display driver IC. The sender is a standard type and is
constructed from linearly wound resistance wire with a range from 10R
to 100R. I'll post the circuit later, but in the meantime does anyone
know of any design rules covering fuel tank senders? I ask this
because the wiper arm and spring loaded rotating contact are very
intermittent, making me wonder how it is possible for electrical
contacts to safely operate in the presence of petrol vapour.
Presumably there is oxygen in the tank, too.


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.

 

[Tom]

Franc Zabkar wrote:

I have a fuel level sender and LED gauge from a privately built
aircraft. It uses a strange implementation of an LM3914 dot/bargraph
LED display driver IC. The sender is a standard type and is
constructed from linearly wound resistance wire with a range from 10R
to 100R. I'll post the circuit later, but in the meantime does anyone
know of any design rules covering fuel tank senders? I ask this
because the wiper arm and spring loaded rotating contact are very
intermittent, making me wonder how it is possible for electrical
contacts to safely operate in the presence of petrol vapour.
Presumably there is oxygen in the tank, too.


- Franc Zabkar

I guess most cars used that system in the past, long time ago I fixed
one on my VE Valiant (made in 1969) and it was just as you describe. I'm
pretty sure I've seen a lot of other car manufacturers using same or
similar senders - just check some old Gregorys manual in your local
library for details.

Tom

 

[Franc Zabkar]

On Wed, 18 May 2005 13:18:45 +1000, Tom <tom@nospam.net> put finger to
keyboard and composed:

Franc Zabkar wrote:
I have a fuel level sender and LED gauge from a privately built
aircraft. It uses a strange implementation of an LM3914 dot/bargraph
LED display driver IC. The sender is a standard type and is
constructed from linearly wound resistance wire with a range from 10R
to 100R. I'll post the circuit later, but in the meantime does anyone
know of any design rules covering fuel tank senders? I ask this
because the wiper arm and spring loaded rotating contact are very
intermittent, making me wonder how it is possible for electrical
contacts to safely operate in the presence of petrol vapour.
Presumably there is oxygen in the tank, too.


- Franc Zabkar

I guess most cars used that system in the past, long time ago I fixed
one on my VE Valiant (made in 1969) and it was just as you describe.

I'm reasonably familiar with the resistive senders. I believe most
vehicles use these. There was a time when Ford Australia released
capacitive senders in some of their models but these were susceptible
to moisture and were discontinued.

I'm
pretty sure I've seen a lot of other car manufacturers using same or
similar senders - just check some old Gregorys manual in your local
library for details.

Tom

I would have expected aircraft to be fitted with better
instrumentation than the average car, but this home built airplane
continues to surprise me. The pilot tells me that the electronic fuel
gauge was designed and built by a Queenslander, but unfortunately he
is no longer interested in supporting it.

Here is the circuit:
http://members.optusnet.com.au/~fzabkar/Fuel_Gauge2.gif

Notice that the IC's negative supply rail, which is normally connected
to circuit ground, is left to "float" on the sender input. Notice also
where the IC gets its signal input. The fact that pin 4 connects to
pin 8 means that the sender's normal signal range of 0-5V is reduced
to only 1.25V. I can only imagine that this design is attempting to
linearise the sender's input in some novel way, but I can't see how.
If this were my design, I would have used a 50mA current source to
drive the sender, and a standard 0-5V meter circuit straight from
National Semiconductor's datasheet
(http://www.national.com/ds/LM/LM3914.pdf).

BTW, I'm also a Mopar lover from way back.


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.

 

[Jim]

"Mark Harriss" <billy@blartco.co.uk> wrote in message
news:428b24eb$0$253$61c65585@uq-127creek-reader-03.brisbane.pipenetworks.com
..au...

Franc Zabkar wrote:
I have a fuel level sender and LED gauge from a privately built
aircraft. It uses a strange implementation of an LM3914 dot/bargraph
LED display driver IC. The sender is a standard type and is
constructed from linearly wound resistance wire with a range from 10R
to 100R. I'll post the circuit later, but in the meantime does anyone
know of any design rules covering fuel tank senders? I ask this
because the wiper arm and spring loaded rotating contact are very
intermittent, making me wonder how it is possible for electrical
contacts to safely operate in the presence of petrol vapour.
Presumably there is oxygen in the tank, too.


- Franc Zabkar


I used to have a Cessna one and one off a Bell 47 and yes they are
just rheostat types like you'd find in a car, with a gauge needle
driven by a heater winding on a bi-metallic strip. I understand
some of the recent models are of capacitance type too. It's identical
to older car fuel gauges and I don't know why they don't explode
either.

Fascinating. I'm sure that Frank is right, there must be standards for them,
perhaps an aviation ng or similar. Either that or a quick phone call to
the appropriate people.
If there aren't any standards I'm never flying again )

 

[Geoff C]

Franc Zabkar <fzabkar@optussnet.com.au> wrote in
news:g8ll81pbvpdccj263hadrknbq6014jh6pe@4ax.com:

I would have used a 50mA current source to

drive the sender, and a standard 0-5V meter circuit straight from
National Semiconductor's datasheet
(http://www.national.com/ds/LM/LM3914.pdf).

BTW, I'm also a Mopar lover from way back.


- Franc Zabkar

Seems to me a current source is the exact wrong way to go, especially as
high as 50 mA. If the there was a break in the sender with a small air gap,
a current source will try to force 50 mA through the gap until it hits a
voltage limit or a spark jumps. I know the voltage is relatively low, but I
would be surprised if the concept of a current source is used in car or
plane.

 

[Mark Harriss]

Franc Zabkar wrote:

I have a fuel level sender and LED gauge from a privately built
aircraft. It uses a strange implementation of an LM3914 dot/bargraph
LED display driver IC. The sender is a standard type and is
constructed from linearly wound resistance wire with a range from 10R
to 100R. I'll post the circuit later, but in the meantime does anyone
know of any design rules covering fuel tank senders? I ask this
because the wiper arm and spring loaded rotating contact are very
intermittent, making me wonder how it is possible for electrical
contacts to safely operate in the presence of petrol vapour.
Presumably there is oxygen in the tank, too.


- Franc Zabkar

I used to have a Cessna one and one off a Bell 47 and yes they are
just rheostat types like you'd find in a car, with a gauge needle
driven by a heater winding on a bi-metallic strip. I understand
some of the recent models are of capacitance type too. It's identical
to older car fuel gauges and I don't know why they don't explode
either.

 

[Alex Gibson]

"Jim" <nonya_nospammy@spamlesserver.com.au> wrote in message
news:428bbd80_1@news.iprimus.com.au...

"Mark Harriss" <billy@blartco.co.uk> wrote in message
news:428b24eb$0$253$61c65585@uq-127creek-reader-03.brisbane.pipenetworks.com
.au...
Franc Zabkar wrote:
I have a fuel level sender and LED gauge from a privately built
aircraft. It uses a strange implementation of an LM3914 dot/bargraph
LED display driver IC. The sender is a standard type and is
constructed from linearly wound resistance wire with a range from 10R
to 100R. I'll post the circuit later, but in the meantime does anyone
know of any design rules covering fuel tank senders? I ask this
because the wiper arm and spring loaded rotating contact are very
intermittent, making me wonder how it is possible for electrical
contacts to safely operate in the presence of petrol vapour.
Presumably there is oxygen in the tank, too.


- Franc Zabkar


I used to have a Cessna one and one off a Bell 47 and yes they are
just rheostat types like you'd find in a car, with a gauge needle
driven by a heater winding on a bi-metallic strip. I understand
some of the recent models are of capacitance type too. It's identical
to older car fuel gauges and I don't know why they don't explode
either.

Fascinating. I'm sure that Frank is right, there must be standards for
them,
perhaps an aviation ng or similar. Either that or a quick phone call to
the appropriate people.
If there aren't any standards I'm never flying again )

There are standards alright (and tons of manuals(literally) for civil
aircraft).

Homebuilt are a differnt kettle of fish all together.
A LAME - Licensed Aicraft Maintenance Engineer should know
maybe not about homebuilt.

If in doubt try having a look on the CASA website or
try contacting them.http://www.casa.gov.au/

Probably better off contacting them or talking to someone
who knows the rules and regs covering homebuilt.

Some quite tought penalties covering aviation
and airworthiness and safety as well as a shitload of paperwork
regardless of what you are doing.

Alex Gibson

Alex

 

[Franc Zabkar]

On Wed, 18 May 2005 23:17:47 GMT, Geoff C <notinterestedin@spam.com>
put finger to keyboard and composed:

Franc Zabkar <fzabkar@optussnet.com.au> wrote in
news:g8ll81pbvpdccj263hadrknbq6014jh6pe@4ax.com:

I would have used a 50mA current source to
drive the sender, and a standard 0-5V meter circuit straight from
National Semiconductor's datasheet
(http://www.national.com/ds/LM/LM3914.pdf).

BTW, I'm also a Mopar lover from way back.


- Franc Zabkar

Seems to me a current source is the exact wrong way to go, especially as
high as 50 mA. If the there was a break in the sender with a small air gap,
a current source will try to force 50 mA through the gap until it hits a
voltage limit or a spark jumps. I know the voltage is relatively low, but I
would be surprised if the concept of a current source is used in car or
plane.

The existing circuit uses a 180 ohm resistor in series with a 15V
supply. This allows a current of 79mA to flow when the tank is full
and 54mA when it is empty. In the event of a broken sender, the OC
voltage would be 15V, the same as for an OC current source. AFAICS a
50mA current source should be safer.

BTW, my current source would use a series limiting resistor, say 180R.
The sender voltage would range from 0 to 5V, and the series resistor
would drop 9V. This leaves a range of 1V to 6V for the current source.


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.

 

[Franc Zabkar]

On Thu, 19 May 2005 18:40:05 -0800, Mark Harriss <billy@blartco.co.uk>
put finger to keyboard and composed:

You know Franc, for less than $140 USD
your friend could have a new gauge and
sender unit for his plane, thats pretty
good for plane prices. I'd be very wary
of touching anything on planes without
the necessary approvals, even if it is
a homebuilt plane.

Is it a homebuilt with normal
certification or a homebuilt with an
experimental certificate?, the regs
could be very different. A call to an
avionics shop could put you in the
picture.

All I know is that the plane was built from a commercially supplied
kit.


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.

 

[Geoff C]

Franc Zabkar <fzabkar@optussnet.com.au> wrote in
newshdo815q11lopsjknscc4hpa0mi7073d0s@4ax.com:

On Wed, 18 May 2005 23:17:47 GMT, Geoff C <notinterestedin@spam.com
put finger to keyboard and composed:

Franc Zabkar <fzabkar@optussnet.com.au> wrote in
news:g8ll81pbvpdccj263hadrknbq6014jh6pe@4ax.com:

I would have used a 50mA current source to
drive the sender, and a standard 0-5V meter circuit straight from
National Semiconductor's datasheet
(http://www.national.com/ds/LM/LM3914.pdf).

BTW, I'm also a Mopar lover from way back.


- Franc Zabkar

Seems to me a current source is the exact wrong way to go, especially
as high as 50 mA. If the there was a break in the sender with a small
air gap, a current source will try to force 50 mA through the gap
until it hits a voltage limit or a spark jumps. I know the voltage is
relatively low, but I would be surprised if the concept of a current
source is used in car or plane.

The existing circuit uses a 180 ohm resistor in series with a 15V
supply. This allows a current of 79mA to flow when the tank is full
and 54mA when it is empty. In the event of a broken sender, the OC
voltage would be 15V, the same as for an OC current source. AFAICS a
50mA current source should be safer.

BTW, my current source would use a series limiting resistor, say 180R.
The sender voltage would range from 0 to 5V, and the series resistor
would drop 9V. This leaves a range of 1V to 6V for the current source.


- Franc Zabkar

Well that is interesting. I suppose then that must be the standard way to
do it. I did not know that. If that is the way it has been done for the
last 80 years then it must be a proven method.

 

[Geoff C]

Franc Zabkar <fzabkar@optussnet.com.au> wrote in
newshdo815q11lopsjknscc4hpa0mi7073d0s@4ax.com:

On Wed, 18 May 2005 23:17:47 GMT, Geoff C <notinterestedin@spam.com
put finger to keyboard and composed:

Franc Zabkar <fzabkar@optussnet.com.au> wrote in
news:g8ll81pbvpdccj263hadrknbq6014jh6pe@4ax.com:

I would have used a 50mA current source to
drive the sender, and a standard 0-5V meter circuit straight from
National Semiconductor's datasheet
(http://www.national.com/ds/LM/LM3914.pdf).

BTW, I'm also a Mopar lover from way back.


- Franc Zabkar

Seems to me a current source is the exact wrong way to go, especially
as high as 50 mA. If the there was a break in the sender with a small
air gap, a current source will try to force 50 mA through the gap
until it hits a voltage limit or a spark jumps. I know the voltage is
relatively low, but I would be surprised if the concept of a current
source is used in car or plane.

The existing circuit uses a 180 ohm resistor in series with a 15V
supply. This allows a current of 79mA to flow when the tank is full
and 54mA when it is empty. In the event of a broken sender, the OC
voltage would be 15V, the same as for an OC current source. AFAICS a
50mA current source should be safer.

BTW, my current source would use a series limiting resistor, say 180R.
The sender voltage would range from 0 to 5V, and the series resistor
would drop 9V. This leaves a range of 1V to 6V for the current source.


- Franc Zabkar

Well that is interesting. I suppose then that must be the standard way to
do it. I did not know that. If that is the way it has been done for the
last 80 years then it must be a proven method.

 

[Mark Harriss]

You know Franc, for less than $140 USD
your friend could have a new gauge and
sender unit for his plane, thats pretty
good for plane prices. I'd be very wary
of touching anything on planes without
the necessary approvals, even if it is
a homebuilt plane.

Is it a homebuilt with normal
certification or a homebuilt with an
experimental certificate?, the regs
could be very different. A call to an
avionics shop could put you in the
picture.

 

[Ross Herbert]

On Thu, 19 May 2005 17:22:00 +1000, Franc Zabkar
<fzabkar@optussnet.com.au> wrote:

On Wed, 18 May 2005 23:17:47 GMT, Geoff C <notinterestedin@spam.com
put finger to keyboard and composed:

Franc Zabkar <fzabkar@optussnet.com.au> wrote in
news:g8ll81pbvpdccj263hadrknbq6014jh6pe@4ax.com:

I would have used a 50mA current source to
drive the sender, and a standard 0-5V meter circuit straight from
National Semiconductor's datasheet
(http://www.national.com/ds/LM/LM3914.pdf).

BTW, I'm also a Mopar lover from way back.


- Franc Zabkar

Seems to me a current source is the exact wrong way to go, especially as
high as 50 mA. If the there was a break in the sender with a small air gap,
a current source will try to force 50 mA through the gap until it hits a
voltage limit or a spark jumps. I know the voltage is relatively low, but I
would be surprised if the concept of a current source is used in car or
plane.

The existing circuit uses a 180 ohm resistor in series with a 15V
supply. This allows a current of 79mA to flow when the tank is full
and 54mA when it is empty. In the event of a broken sender, the OC
voltage would be 15V, the same as for an OC current source. AFAICS a
50mA current source should be safer.

BTW, my current source would use a series limiting resistor, say 180R.
The sender voltage would range from 0 to 5V, and the series resistor
would drop 9V. This leaves a range of 1V to 6V for the current source.

I am fairly certain that the resistive element in fuel gauge senders
is completely enclosed thus making the risk of explosion from arcing
impossible. The main problem with resistive senders is that they
eventually become unreliable due to wear and tear on the sliding
contact elements. One of the most reliable types is that using a set
of reed relay switches and a float magnet such as shown in these
examples
http://www.avnet.co.uk/lts/pages/fuelgauge.htm
http://www.almegcontrols.com/fuel_level_sender.htm

 

[budgie]

On Fri, 20 May 2005 05:06:02 GMT, Ross Herbert <rherber1@bigpond.net.au> wrote:

On Thu, 19 May 2005 17:22:00 +1000, Franc Zabkar
fzabkar@optussnet.com.au> wrote:

On Wed, 18 May 2005 23:17:47 GMT, Geoff C <notinterestedin@spam.com
put finger to keyboard and composed:

Franc Zabkar <fzabkar@optussnet.com.au> wrote in
news:g8ll81pbvpdccj263hadrknbq6014jh6pe@4ax.com:

I would have used a 50mA current source to
drive the sender, and a standard 0-5V meter circuit straight from
National Semiconductor's datasheet
(http://www.national.com/ds/LM/LM3914.pdf).

BTW, I'm also a Mopar lover from way back.


- Franc Zabkar

Seems to me a current source is the exact wrong way to go, especially as
high as 50 mA. If the there was a break in the sender with a small air gap,
a current source will try to force 50 mA through the gap until it hits a
voltage limit or a spark jumps. I know the voltage is relatively low, but I
would be surprised if the concept of a current source is used in car or
plane.

The existing circuit uses a 180 ohm resistor in series with a 15V
supply. This allows a current of 79mA to flow when the tank is full
and 54mA when it is empty. In the event of a broken sender, the OC
voltage would be 15V, the same as for an OC current source. AFAICS a
50mA current source should be safer.

BTW, my current source would use a series limiting resistor, say 180R.
The sender voltage would range from 0 to 5V, and the series resistor
would drop 9V. This leaves a range of 1V to 6V for the current source.


I am fairly certain that the resistive element in fuel gauge senders
is completely enclosed thus making the risk of explosion from arcing
impossible.

Not on the ones I've attended to, on Nissans and Commodes. They are simply a
wire-wound linear pot with a wiper. Anyway, any exclosure would need to be
vapour-sealed to have any impact on the prospect of pyrotechnic events.

The main problem with resistive senders is that they
eventually become unreliable due to wear and tear on the sliding
contact elements.

which causes intermittent contact and presumably - to the extent permitted by
the 10V reg and meter impedance - some sparking.

One of the most reliable types is that using a set
of reed relay switches and a float magnet such as shown in these
examples
http://www.avnet.co.uk/lts/pages/fuelgauge.htm
http://www.almegcontrols.com/fuel_level_sender.htm

 

[Ross Herbert]

On Fri, 20 May 2005 16:18:13 -0800, Mark Harriss <billy@blartco.co.uk>
wrote:

Ross Herbert wrote:
I am fairly certain that the resistive element in fuel gauge senders
is completely enclosed thus making the risk of explosion from arcing
impossible. The main problem with resistive senders is that they
eventually become unreliable due to wear and tear on the sliding
contact elements. One of the most reliable types is that using a set
of reed relay switches and a float magnet such as shown in these
examples
http://www.avnet.co.uk/lts/pages/fuelgauge.htm
http://www.almegcontrols.com/fuel_level_sender.htm


The ones I've seen are all have the rheostat element exposed to
the fuel. They just have a half can to mount the element on the
inside of

The one I pulled apart was made in the 50's and I don't think it was
from a car.

In order to get any sort of arcing you first need the circuit carrying
the current to go open circuit. Since the wiper is tied to one end of
the resistive element it can never go open circuit so there is no
arcing. The only effect is that the full resistance of the element is
applied.

 

[Franc Zabkar]

On Fri, 20 May 2005 00:58:37 GMT, Geoff C <notinterestedin@spam.com>
put finger to keyboard and composed:

Franc Zabkar <fzabkar@optussnet.com.au> wrote in
newshdo815q11lopsjknscc4hpa0mi7073d0s@4ax.com:

On Wed, 18 May 2005 23:17:47 GMT, Geoff C <notinterestedin@spam.com
put finger to keyboard and composed:

Franc Zabkar <fzabkar@optussnet.com.au> wrote in
news:g8ll81pbvpdccj263hadrknbq6014jh6pe@4ax.com:

I would have used a 50mA current source to
drive the sender, and a standard 0-5V meter circuit straight from
National Semiconductor's datasheet
(http://www.national.com/ds/LM/LM3914.pdf).

BTW, I'm also a Mopar lover from way back.


- Franc Zabkar

Seems to me a current source is the exact wrong way to go, especially
as high as 50 mA. If the there was a break in the sender with a small
air gap, a current source will try to force 50 mA through the gap
until it hits a voltage limit or a spark jumps. I know the voltage is
relatively low, but I would be surprised if the concept of a current
source is used in car or plane.

The existing circuit uses a 180 ohm resistor in series with a 15V
supply. This allows a current of 79mA to flow when the tank is full
and 54mA when it is empty. In the event of a broken sender, the OC
voltage would be 15V, the same as for an OC current source. AFAICS a
50mA current source should be safer.

BTW, my current source would use a series limiting resistor, say 180R.
The sender voltage would range from 0 to 5V, and the series resistor
would drop 9V. This leaves a range of 1V to 6V for the current source.


- Franc Zabkar

Well that is interesting. I suppose then that must be the standard way to
do it. I did not know that. If that is the way it has been done for the
last 80 years then it must be a proven method.

The "standard" way utilises a mechanical gauge with a fixed series
resistance, and a bimetallic strip for damping. The rheostat in the
sender forms the bottom half of a potential divider. This arrangement
is highly nonlinear. For example, the difference between "full" and
half" may be 30L while the difference between "half" and "empty" may
be only 20L. Of course part of the problem may be the uneven cross
section of the tank.


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.

 

[Ross Herbert]

On Fri, 20 May 2005 17:33:55 -0800, Mark Harriss <billy@blartco.co.uk>
wrote:

Ross Herbert wrote:

The one I pulled apart was made in the 50's and I don't think it was
from a car.

In order to get any sort of arcing you first need the circuit carrying
the current to go open circuit. Since the wiper is tied to one end of
the resistive element it can never go open circuit so there is no
arcing. The only effect is that the full resistance of the element is
applied.


What if the wiper wears through the wire while
power is applied and then the fuel level changes
so as to have the wiper pass over the wire break
moving towards the end connected to the wiper.

Well, I suppose anything is possible even if it is remote. Generally,
the resistance wire is of reasonable thickness such that the tension
of the wiper will not wear through it for at least 50 years or more.
Not only that but the wiper is made of softer material than the
resistor wire so it will wear out faster. An open circuit due to wiper
wear will be the most likely event to occur.

 

[Ross Herbert]

On Fri, 20 May 2005 14:09:27 +0800, budgie <me@privacy.net> wrote:

Not on the ones I've attended to, on Nissans and Commodes. They are simply a
wire-wound linear pot with a wiper. Anyway, any exclosure would need to be
vapour-sealed to have any impact on the prospect of pyrotechnic events.

The main problem with resistive senders is that they
eventually become unreliable due to wear and tear on the sliding
contact elements.

which causes intermittent contact and presumably - to the extent permitted by
the 10V reg and meter impedance - some sparking.

It seems that the risk of a petrol tank explosion is next to
impossible according to this document
http://www.fireservicecollege.ac.uk/cssc/fi2/Risk%20of%20Explosion%20in%20vehicle%20fires.PDF

so my theory for having an enclosed sender doesn't come into it where
petrol is concerned.

 

[budgie]

On Fri, 20 May 2005 17:09:38 +1000, Franc Zabkar <fzabkar@optussnet.com.au>
wrote:

The "standard" way utilises a mechanical gauge with a fixed series
resistance, and a bimetallic strip for damping. The rheostat in the
sender forms the bottom half of a potential divider. This arrangement
is highly nonlinear. For example, the difference between "full" and
half" may be 30L while the difference between "half" and "empty" may
be only 20L. Of course part of the problem may be the uneven cross
section of the tank.

Most of the non-linearity comes from two sources:

.. the meter responds to current, the sender is a avariable resistor - partly a
hyperbolic function.

.. the sender arm traverses say 70 degrees as the fuel level goes from full to
empty. The angle at half-tank isn't half-travel or half-resistance.

Obviously these factors can be "aligned" to partially compensate, but the result
is always going to be non-linear with such a configuration.

 

[Mark Harriss]

Ross Herbert wrote:

I am fairly certain that the resistive element in fuel gauge senders
is completely enclosed thus making the risk of explosion from arcing
impossible. The main problem with resistive senders is that they
eventually become unreliable due to wear and tear on the sliding
contact elements. One of the most reliable types is that using a set
of reed relay switches and a float magnet such as shown in these
examples
http://www.avnet.co.uk/lts/pages/fuelgauge.htm
http://www.almegcontrols.com/fuel_level_sender.htm

The ones I've seen are all have the rheostat element exposed to
the fuel. They just have a half can to mount the element on the
inside of