Hey All -
Need some advice/help here...I have a standard sine wave analog signal that I need to split into two signals...It doesn't really matter how I get there, as long as it's reproduced into to separate signals. I'm sure the Arduino will read the signal via input, but are there output pins assignable to replicate it? Or perhaps a shield that will do this?
Thanks in advance for the help.
Why not just buffer it with an op amp or two?
http://www.eecs.tufts.edu/~dsculley/tutorial/opamps/opamps5.html
Are these "separate" signals both referenced to the same ground? If so, then as Crossroads says, why not buffer with an opamp. If they need to be electrically isolated from each other then it becomes more complex - there are amplifiers / buffers out there designed to replicate the input signal on the output but isolated from it (floating), but those are much more complex (and expensive). You need a clear definition of just what is meant by "separate" to be able to solve the problem. If it is at a fixed frequency, then the solution may be as simple as a couple of transformers to provide the isolation (if we are talking of the floating outputs).
Thanks guys. The signals are coming from two separate monopole RPM sensors. One sensor produces a sinewave that does not go much above 50mV, at around 5-8kHz. The other produces a sinewave at peak somewhere around 1.5-2.0V, 8-10kHz. I need to split each one of these signals, as I have 2 indicators that need to read the same for each signal. An Op-Amp gain/buffer circuit would work I suppose as long as the impedance on the indicators is not higher than the buffered/amplified signal, correct? Would it be as simple as amplifying the signal by exactly 2x then splitting the output? I believe they use the same ground, as the rest of the equipment is all grounded through the same framework...
I attached printouts from my Fluke O-scope just for your reference...
IMG_20160321_0001.pdf (456 KB)
If the indicators do not load down the input signal (i.e. cause a significant voltage drop) then all you need to do is run a wire from the signal output to each of the two indicator inputs (a "Y" connection). The grounds must all be common, of course.
One sensor produces a sinewave that does not go much above 50mV, at around 5-8kHz. The other produces a sinewave...
The Arduino cannot accept or read the negative half of an AC waveform. The Arduino can be damaged by negative voltages and if the Arduino survives, the negative half of the signal will be clipped-off. (Minus 50mV won't hurt the Arduino, but you'd only be able to read the positive half of the waveform.)
The simplest way to handle the an AC waveform is to [u]bias the input[/u]. With the input biased at 2.5V, you'll get an ADC reading of (about) 512 with no signal. You can subtract that out in software if necessary.
You won't get much resolution at 50mV.... You should be able to "detect" the signal, but with the 10-bit ADC and the default 5V reference, you get a resolution of about 5mV per count. i.e. If no signal reads 512, 50mV will read 522 (on the positive peaks). You can use the optional 1.1V reference for about 5 times the resolution, but you'll have to lower the bias voltage.
Need some advice/help here...I have a standard sine wave analog signal that I need to split into two signals...
Since the Arduino has almost infinite impedance you could probably connect one signal to 100 Arduino inputs!!! (There is some capacitance, but at 8kHz or less, that shouldn't be an issue.)
There's a general rule that you can connect multiple inputs together, but you cannot connect outputs together.
It doesn't really matter how I get there, as long as it's reproduced into to separate signals.
It's not clear what you mean by "separate".
...but are there output pins assignable to replicate it?
No... The standard Arduino doesn't have an analog output. (It has PWM which can dim an LED or otherwise sometimes approximate analog.)
Are you dealing with 1 or two signals you want to work with? Your initial post indicated "I have a standard sine wave analog signal that I need to split into two signals" which is entirely different from 2 distinct signals you want to process separately. You later indicate you are talking about 2 signals you want to process. It is confusing as to just what the configuration really is.
glturbines:
Thanks guys. The signals are coming from two separate monopole RPM sensors. One sensor produces a sinewave that does not go much above 50mV, at around 5-8kHz. The other produces a sinewave at peak somewhere around 1.5-2.0V, 8-10kHz.
I assume you want to measure RPM.
A zerocrossing detector then seems logical.
Use a 100k resistor between your source to the Arduino analogue input pin.
Anything with an A/D value of 0 is negative sinewave, anything >=1 is positive.
If you set the Aref to INTERNAL, you have ~1mV switchpoints.
The source is loaded with the resistor value you use.
Use the resistor close to the Arduino to avoid hum/hash pickup.
Don't know if the Arduino is fast enough to process one or even two of those signals.
Leo..
It beginning to seem as though the Arduino may not be able to replicate what I need...Without an analog output replication signal, I would have to convert to digital then back to analog...with the same voltages...Sounds like I'd be in for quite the ride...
CrossRoads & gpsmikey -
The sensors I have are two wire...If I attempt the buffer route, do both wires go to one op amp buffer circuit? Seems like if I connected one of the wires to ground I would lose the signal...Or do I need to buffer each wire with a separate op-amp? Or can I split the signal on the output side? Sorry, I've played with these circuits a little before, but it was all for a single circuit...Never had to buffer peak-to-peak AC sinewaves for duplication...Also, you mentioned options for replicating the input signal on the output side...I don't think they would have to be isolated as long as I have 2 distinct sinewaves that are the same amplitudes that I can carry to each indicator. The Y connection makes sense if it will not ground out the waveform and amplitude...But I'm assuming that's what the buffer circuit is for...
Maybe if you drew a block diagram of what the configuration looks like and what you want to accomplish we can figure it out. You keep going between doing something with one signal and splitting/buffering it and two different signals you want to do something with. It still is not clear just what we are talking about here (what do you want to do with the arduino, where do the two displays fit in and what else happens to the signals? The toughest part of a project is often defining the requirements and what you are trying to accomplish. I think we are still in that phase - we have not made it to the "lets stick the hardware together" yet.
Why do you want to replicate. It might not be needed.
If the output impedance of the sensor is low enough (powerfull enough), it can connect to multiple inputs.
Leo..
I drew up a very crude block diagram of what I need to do. You will note that it is a 2-wire signal coming from each sensor going to a powered indicator. I need to split each signal and run them to additional indicators, respectively. I have already tried simply splicing in and running a second indicator and it did not work.
IMG_20160330_0001.pdf (130 KB)
How about a link to what the sensors and indicators are?
CrossRoads:
How about a link to what the sensors and indicators are?
Can't. The sensors are proprietary to Honeywell turbine engines and the indicators are out of a Hawker HS125-700A jet. The engine is a Honeywell TFE731-3R-1H high bypass turbofan. This is why I had to get some readings from the signals, so I would at least have an idea of what I'm dealing with. It's really anyone's guess as to what the indicators impedance is at the signal pins or how it processes it. Unless you can tell me what to look for with a Fluke at the input signal pins on the meter...This is why I am trying to duplicate the signal right from the sensor. Seems to me with all the technology in the world, someone would have found a way to split an analog signal...lol
I doubt it. Those are going to be commonly-available sensors although perhaps the housing is custom.
With a high impedance input such as an opamp or an Arduino analog input, you can just wire it exactly like your diagram. You obviously know something about the signal, do you have an oscilloscope?
MorganS:
I doubt it. Those are going to be commonly-available sensors although perhaps the housing is custom.With a high impedance input such as an opamp or an Arduino analog input, you can just wire it exactly like your diagram. You obviously know something about the signal, do you have an oscilloscope?
True...However there are no numbers or markings on the pickups themselves. Yes I have an oscilloscope. Attached a print out of the waveforms and amplitudes earlier in this thread.
So which side would of the split would the op amp go on?
OK, I looked at the waveforms. I am not familiar with that scope to see if you're in AC or DC mode and what the centre voltage of the waveform is. We also need to know how both of the sensor wires are biased with respect to system ground. As one of the regulars already posted, you can't have anything go below zero for the Arduino. An opamp is a perfect way to buffer and shift the voltage.
This "split" you seem so worried about could be just three wires crimped into a single crimp terminal. There is no "side".
5KHz is a little high to measure with the Arduino's analog input. The best you can do without special effort on an AVR-based Arduino is 10KHz sampling, so you won't get much of a waveform. What is it you want to do with this signal again?