Subject:
|
Re: To all sensor developers, why use voltage drive ?
|
Newsgroups:
|
lugnet.robotics.rcx
|
Date:
|
Sat, 3 Mar 2001 09:55:07 GMT
|
Viewed:
|
1533 times
|
| |
|
|
hi Michael, thanks for you reaction.
Michael Gasperi wrote:
> Stef Mientki <s.mientki@ids.kun.nl> wrote in message
> news:3AA01C30.6FFBA740@ids.kun.nl...
> > Hi,
> >
> > Why is everyone using a voltage driven cirquit when devolping a RCX
> sensor-cirquit, while the RCX sense input is in fact a current sensing input
> ?
>
> What the RCX measures is voltage. It has a 10k pull up to 5V, but it
> measures voltage.
So here I disagree with you , you are correct if you'r saying the inside AD-converter is measuring voltage, but we're not dealing with the AD-converter alone, but with a combination of the AD-converter, the pull up resistor, the diodes and may be more, in short the outside RCX terminals.
In fact you can not drive the sense input with a real voltage source, because then the driver is killing his own power supply, that's why you have to use the 1 kOhm resistor. And in that case you haven't a voltage source anymore.
But there is more, a current coupling between sensor and RCX input is totally insensitive to (voltage) noise, that's why the diodes, or what so ever is in series with the sensor (long wires), doesn't play any role anymore.
> If you force a voltage on the input using a low enough
> impedance you will see that it measures that voltage. Your current based
> circuits are compensating for the diodes non linear behavior. Also the
> LM324 is a pretty poor opamp.
>
> The reason for the opamp with a 1k resistor circuit is that it is simple.
> Just like the diodes in place of FETS.
Since I discovered that the sense input is a current input, my opinion about the FETs has changed (they are in the wrong direction, there is a (theoretical) problem in sensing low voltages, and of course there's a better way to gain linearity), but I've to adapt my webpage.
> Great linearity is often not that
> necessary. My projects are driven mostly by solutions that can come from
> the Radio Shack retail store and can be built without surface mount
> technology. It is a compromise.
Ok, that's a good argument.
I've looked at the prices at Radio Shack, the TLC2274 is about 10 times as expensive as the LM324.
In the Netherlands (Conrad,Display) the price difference between these is only about a factor 2.5 .. 3
Maybe I've used a too expensive opamp, the most important spec of the opamp, in the suggested current drive cirquit, is that it has to perform rail-to-rail input characteristics.
If I find another (cheaper) opamp that performs equally, I'll inform you.
Stef Mientki
|
|
Message has 1 Reply:
Message is in Reply To:
6 Messages in This Thread:
  
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
