scudderfish
|
| posted on 29/6/19 at 11:27 AM |
|
|
Comedy fuel gauge
I had to replace my old fuel gauge with a cheap Chinese ebay special. It works, but it has no dampening so I get to see every slosh of the fuel in
the tank. I think if I put a capacitor between signal & ground I should be able to smooth it out but I have no idea how to calculate (or even
educated guess) what I need. Has anyone done this or have any idea on what I need to do?
Regards,
David
|
|
|
|
|
gremlin1234
|
| posted on 29/6/19 at 11:45 AM |
|
|
a capacitor should work,
I would use an electrolytic one, rated at least 16Volt
as per its value, not sure...
but I think I would try about 10uF, to see if it had any effect and work from there
edit: or start with 100uF, if thats what you have to hand.
[Edited on 29/6/19 by gremlin1234]
|
|
|
PAUL FISHER
|
| posted on 30/6/19 at 05:00 PM |
|
|
I've got a Westfield fuel tank gauge if it's any good to you, £20 delivered uk
|
|
|
G13BLocost
|
| posted on 30/6/19 at 07:12 PM |
|
|
Bigger the cap the greater the effect. This is know as a low pass filter and the cut off frequency is proportional to 1/(R*C). R being the resistance
between the gauge and tank.
If you can give us that resistance value I can calculate an exact number (say 1hz filter) or you can just keep adding capacitance as the post says
above until you are happy with it.
My Locost blog: ogilvietacing.com
|
|
|
scudderfish
|
| posted on 30/6/19 at 08:23 PM |
|
|
I've ordered a pack of different sized capacitors from ebay so next weekend I'll just have a go at working my way through them until it
behaves itself a bit more  Thanks for the pointers.
|
|
|
gremlin1234
|
| posted on 30/6/19 at 08:31 PM |
|
|
remember electrolytic caps are polarity sensitive
long lead is plus
short lead minus,
and usually a stripe on the side of the cap to mark minus/neg too.
|
|
|
nick205
|
| posted on 1/7/19 at 08:59 AM |
|
|
quote:
Originally posted by gremlin1234
remember electrolytic caps are polarity sensitive
long lead is plus
short lead minus,
and usually a stripe on the side of the cap to mark minus/neg too.
Important to remember this!
As a junior PCB engineer I got this wrong some years ago in the PCB assembly instructions. The PCBs were assembled and seemed to work long enough to
pass the test procedure and then fail rather quickly at the customer site.
An expensive mistake I brought about upon my employer since we had to do a product recall, rebuild, longer test and re-ship at my employers expense.
|
|
|
nick205
|
| posted on 1/7/19 at 10:37 AM |
|
|
More expensive and time consuming than capacitors, but is fitting a baffle around the fuel level sensor an option?
If you can slow the fuel from sloshing around in the tank it should get you a steadier read on the gauge.
|
|
|
907
|
| posted on 1/7/19 at 11:32 AM |
|
|
The Suttol has a damped fuel gauge,
Trouble is the Lotus engine can use fuel faster than the gauge goes down. I have to pull in a lay-by to let the gauge catch up.
Paul G
|
|
|
nick205
|
| posted on 1/7/19 at 04:14 PM |
|
|
Just a thought - do you still have the old gauge?
If you do can you dismantle it to see if there's a capacitor built into it for an idea of capacitor type and rating?
|
|
|
scudderfish
|
| posted on 2/7/19 at 04:05 PM |
|
|
I think I'm going to go for a ridiculous solution because :-
1) I can
2) see 1.
My plan is to use an ESP32 microcontroller to measure the resistance from the sender and then use PWM to drive the gauge with code to suitably damp
the motion. I can then set it up with a webpage so that I can configure the gauge range and speed of response. The idea of having a webserver on a
powerful computer that cost me less than £10 just to fiddle a gauge in a car that tries to give off a 60's vibe tickles me.
Regards,
David
|
|
|
scudderfish
|
| posted on 2/7/19 at 04:06 PM |
|
|
Incidentally, I believe the original gauge used a bi-metallic strip to move the needle (and damp the motion)
|
|
|
gremlin1234
|
| posted on 2/7/19 at 05:08 PM |
|
|
quote:
Originally posted by scudderfish
Incidentally, I believe the original gauge used a bi-metallic strip to move the needle (and damp the motion)
no, the bi-metallic strip was in
the 'voltage regulator' nominally 10volt, but quite a slow cycle time between 0 and 12+v
so that was a mechanical pulse width modulated voltage control.
the gauges went up and down, even without the fuel sloshing around.
engine temp looks high, - oh but we have plenty of fuel... oh, engine temp looks a lot cooler now, but we need to fill up!
|
|
|
scudderfish
|
| posted on 2/7/19 at 05:12 PM |
|
|
quote:
Originally posted by gremlin1234
quote:
Originally posted by scudderfish
Incidentally, I believe the original gauge used a bi-metallic strip to move the needle (and damp the motion)
no, the bi-metallic strip was in
the 'voltage regulator' nominally 10volt, but quite a slow cycle time between 0 and 12+v
so that was a mechanical pulse width modulated voltage control.
the gauges went up and down, even without the fuel sloshing around.
engine temp looks high, - oh but we have plenty of fuel... oh, engine temp looks a lot cooler now, but we need to fill up!
Gauges could use a BM strip as well http://www.speedyjim.net/htm/fuel_ga.htm
|
|
|
