* * * The Minimal Ammeter? * * *

[safe]

From Wikipedia:

Use in current measuring

A shunt can also be used to measure current. In this case a resistor of accurately-known resistance, the shunt, is placed in series so that all the current to be measured will flow through it. Since the resistance is known, by measuring the voltage drop across it, one can calculate the current flowing.

In order not to disrupt the circuit, the resistance of the shunt is normally very small. Shunts are rated by maximum current and voltage drop at that current, for example, a 500 A/50 mV shunt would have a maximum allowable current of 500 amps and at that current the voltage drop would be 50 millivolts. By convention, most shunts are designed to drop 50 mV when operating at their full rated current and most "ammeters" are actually designed as voltmeters that reach full-scale deflection at 50 mV.

If the current being measured is also at a high voltage potential this voltage will be present in the enclosure containing the reading instrument. Sometimes, the shunt is inserted in the return leg (low voltage side) to avoid this problem. Another solution is to use a Hall effect (non-contact) current sensor instead of a shunt.

So the idea would be to attach a simple LED to the shunt and you would know while riding that if it glowed really bright that you are using a lot of current and need to either back off the throttle or downshift. (depending on your machines abilities) You might also decide to just allow the high current usage in cases like climbing a steep hill. After all, the actual number of amps doesn't really matter if you already know what your controllers current limit is, so all you care about is what percentage of the limit that is being used. Full brightness means the full limit, partial brightness means less than the limit.

Any reason this wouldn't work?

Out of the corner of your eye you would know exactly what was going on..

 

[safe]

As a followup question:

Since the LED has it's own resistance could the LED "itself" somehow be the shunt and eliminate the need for a separate resistor?

How simply can this be done?

What kind of LED might do the job without breaking?

 

[Mathurin]

Well, use of a lamp sounds good, the one thing I didn't like of the Drain Brain is that one must actually read the numbers on it to acquire info... But wouldn't an analog needle meter give the best of both worlds - An at-a-glance guesstimate, and the ability to read it more precisely?

 

[Lowell]

An LM3914 (or several cascaded) would make a nice bar graph ammeter that's easily visible at night. Multi color LEDs would make for easy current range estimation without actually looking at the display.

 

[safe]

But wouldn't an analog needle meter give the best of both worlds - An at-a-glance guesstimate, and the ability to read it more precisely?

If they could make the needles a little bigger this would be true. A light is easy because your eye is attracted to it and the mind doesn't have to process visual information like "where is the needle within the spectrum?".

The Indy Cars are using LED's to show the rpms of the motor. When the rpms drop too low then three LED's light up on the left side of the steering wheel. When the rpm's go too high then three LED's light up on the right side of the steering column. Out of the corner of your eye you can still read them and never have to look down.

Maybe the LED light ONLY comes on when full current is being used?

The idea is to create an instrument that can be read without really looking directly at it...

 

[safe]

An LM3914 (or several cascaded) would make a nice bar graph ammeter that's easily visible at night. Multi color LEDs would make for easy current range estimation without actually looking at the display.

Sounds good to me... sounds like the Indy Cars... 8)

Could you give more detail about this idea?

 

[Mathurin]

Bar graph integrated into the DB's case... Bad ass.

 

[safe]

Since I'm no electrical expert let me see if I understand the shunt. The theory is that electricity follows the path of least resistance. If there are two paths that current can follow then electricity prefers the easy way, but it will also send a little current on the harder path.

This image I made sort of places the LED in the role of the shunt itself and I'm not totally sure if this idea is right. Basically you use a little extra wire to go around the LED which is the shorter path, but since the LED produces a small resistance then it somehow balances out. Am I even remotely on the right track? :?

The goal would be to know the resistance of the LED and then calculate the length of wire you need to match that resistance.... Right?

 

[xyster]

But wouldn't an analog needle meter give the best of both worlds - An at-a-glance guesstimate, and the ability to read it more precisely?

I find the big analog meters like allelectronics carries are very easy to read out of the corner of my eye. Fechter I noticed has the same kind on his scooter. My only wish is that they were lit for nightime use.

 

[safe]

I find the big analog meters like allelectronics carries are very easy to read out of the corner of my eye.

For years people used mechanical connections to their motors to sense things like rpms. But in recent years people have been reexamining those ancient traditions and the racers are now using lights.

You have history on your side, but that doesn't necessarily mean that it's the better solution... so let's not disrupt this <<< NEW >>> idea and go right back to the status quo of the needle meters.

I'd personally prefer an LED over a needle meter... why it's like trying to find a needle in a haystack!

P.S: I've heard that LED lights are good for night time use too. :wink:

 

[safe]

There's also the question of mounting the needle meters and crash damage potential that should be mentioned too. A single LED could be easily placed in the handlebar stem and would be virtually impossible to damage. The needle meter is a very vulnerable item that needs to be placed out in the open to be read. Since you can't really get a peripheral vision look at it, you have to actually LOOK with your eyes right at it, so it must be propped up into your face.

So the LED sort of frees up the rider...

Then there's glare bouncing off the needle meters surface that makes it harder to read... while glare would also effect an LED you wouldn't try to squint to look at because you would either be able to see it or you wouldn't. So that's another issue...

 

[Lowell]

Bar graph integrated into the DB's case... Bad ass.

Now there's a good idea, gotta add that to the project list.

Sample schematic for the LM3914:

http://www.electronics-tutorials.com/devices/lm3914.htm

 

[safe]

So this is an actual circuit that you can buy?

Ten LEDS that cascade would make an awesome display and it doesn't sound that hard to do. More than the "Minimal" idea I had of just "one" this would be really top of the line to pull off.

REALLY COOL! 8)

 

[Stevil_Knevil]

I don't know it's part number, but I have this really cool LED that ranges through three colors; Red Green Blue.

It's mounted in the mouth of a really cool blue gecko carabiner key chain, so I just can't bring myself to break it open.

One heads-up RGB LED indicator (coupled to an analog meter) would offer totally acceptable resolution at speed.

Ten LEDs ranging from green to yellow to red would be even better.


-S

 

[Lowell]

So this is an actual circuit that you can buy?

Ten LEDS that cascade would make an awesome display and it doesn't sound that hard to do. More than the "Minimal" idea I had of just "one" this would be really top of the line to pull off.

REALLY COOL! 8)

You should be able to buy ready made circuits, or experimenter kits with everything you need. The only thing you have to do is adjust resistors to scale the meter reading properly.

http://www.vakits.com/index.php?cPath=110_46_139

If you can wire an e-bike, you can solder this kit together.

 

[Stevil_Knevil]

Ten LEDs ranging from green to yellow to red would be even better.

..IF more resolution would be better for you. -S

 

[xyster]

Most auto racers/enthusiasts seem to prefer analog gauges, as do I. Analog gauges provide immediately visualize-able feedback about where in the spectrum from low to high a measurement lies. The same kind of "whole spectrum" meter can be constructed digitally, but it takes a lot more lights, and a lot more circuitry.

 

[safe]

One way is to incorporate the advantages of the lights in a circular shape that mimicks the old fashioned needle meters.

https://www.rbracing-rsr.com/tachometer.htm

And the newer steering wheels with built in LED's...

http://www.grandprix.com/ft/ft00347.html

"Displays. Gone are the days when drivers were presented with a rev-counter and a variety or other gauges telling him oil and water temperatures and maybe pressures. The RPM of today's 18,000rpm engines changes too fast to be of any use to the driver as an aid to changing gear. Instead, a series of LED's flash on in sequence to tell him when to change up a gear, automatically adjusting to different rev-limits. Intelligent software monitors all the significant parameters of the engine, gearbox and hydraulic systems, and any deviation from normality is warned to the driver by a message on an LCD display"

 

[Matt Gruber]

Except an ammeter is not a tach.
Normally, U should not be looking at an ammeter when racing!
I could see looking at it on a long hill of course, to see if the wrong gear was selected.

It should be used to TRAIN for a race. NOT during the race.

 

[Toorbough ULL-Zeveigh]

This same style of tach as well as LM3914 based ESV's with the tri-color arc were popular school projects back in the 80's. No one ever built an ammeter tho. I'd be surprised that NatSemi even still produces these dot/bar drivers in this day & age of MCU based everything.


As a possible alternative to a sense shunt, does anyone know if an inductive pickup that's used in a clamp type ammeter would work accurately & what's the lower current limit they can detect. Anybody try this on their bike yet?

 

[safe]

It should be used to TRAIN for a race. NOT during the race.

Well that's why my argument is that "Minimal Is Best".

By making it a single LED then it's either fully bright or it's not. Out of the corner of your eye you can spot the degree of "brightness" and that's all you might need to know.

In reality you ride as fast as you can and usually don't pay any attention to any meters unless you might be reading your speedometer to know how fast you are going. (just for grins )

The idea is simplicity... a lone little LED that has no circuit other than the wires that connect it...

But getting back to the "technical parts" what about the simple shunt design I was referring to? Since wires have resistance and an LED has resistance shouldn't you be able to select an LED of known resistance and then calculate the amount of wire that would equal that resistance. There need be no "shunt" at all... just an extra wire... (can LED's handle this situation?)

 

[safe]

Or is this design better?

Attach a wire at two points along the main wire and just assume a "trickle" of current will manage to make it through. So you would simply calculate the lengths of all the wires involved to make it appropriate for the LED resistance you are using. (assuming a much weaker LED than the last one)

 

[fechter]

The analog meter shunt has something like a .050v drop at full scale. That's as simple as it gets.
LEDs need about 2v to light up depending on color. You wouldn't want a shunt with a 2v drop!

You really need a sensing / amplification circuit to drive it.

An indicator that changes color is good since you can detect a change without looking directly at it. That LED tach looks cool. I'm sure something like that could be modified into an ammeter.

 

[safe]

The analog meter shunt has something like a .050v drop at full scale. That's as simple as it gets.
LEDs need about 2v to light up depending on color. You wouldn't want a shunt with a 2v drop!

You really need a sensing / amplification circuit to drive it.

Ah ha! Wisdom Speaks!

So there is a problem, you can't easily light up an LED using a shunt alone because it drains too much current.

What about that alternative idea of simply attaching an extra wire in parallel? It would not create a voltage drop at all (since the path of least resistance remains) but wouldn't some "excess" current "trickle" out along that extra path?

Couldn't you still use the LED, but with a slightly different design?

 

[Matt Gruber]

imo a racer would prefer a buzzer that warns of HIGH MOTOR TEMP