Facebook
Google
GitHub
Linkedin
interesting conceptual question
- Thread starter durable126
- Start date
Wouldn't it be closer to 10 feet (I'm assuming a velocity factor of 2/3, which may be too low for loose wire).
That would be a VOP of 93%
In looking around the past couple minutes I've run across that number in a few places. Though I've seen other sources that give ranges between 42% and 98%. It really depends on the dielectric more than the wire, so it's hard to find firm numbers quickly. But it looks like your 11 ft/ns is at least a pretty common rule of thumb for wire (as opposed to actual transmission lines). My number comes from coax (many, not all, seem to be in the 66% ballpark).
Those are only about 1/6 of a decade apart, so filtering them might me challenging, particularly if you are doing it in analog. If you sample it fast enough you can implement a pretty high order filter digitally. So you might be better off using an RX module that has the high-order built in but that doesn't process the signal beyond that. I think such modules exist (I think some of the previous recommendations are probably of that type).
as far as I'm aware the filtering is built in. Also the traditional RX such as the TSSP4038 is only sensitive to about +- 50 hz or so.
So as far as i know they should not effect the other one.
I just need to figure out how to add a small delay somehow. to break up the continuous signal
I'm still not sure how you are using this delay.
Could you sketch something shows what you want the waveforms to look like, including this delay?
At 38 kHz, a 15 ns delay represents a shift of about 0.05% of the period. Your waveform jitter probably swamps that, so how is this going to help you?
Could you sketch something shows what you want the waveforms to look like, including this delay?
At 38 kHz, a 15 ns delay represents a shift of about 0.05% of the period. Your waveform jitter probably swamps that, so how is this going to help you?
Something like this perhaps
So i get a delay in the middle of the 38 kHz signal to break it up.
on 38 khz
small delay
on 38 khz
small delay
The RX sees the delay as a zero.
the 38 kHz signal by the IR LED is interpreted as a high or "1"
Right now the problem is that i just have
on 38 kHz
on 38 kHz
on 38 kHz
on 38 kHz
on 38 kHz
The RX is designed to filter that kind of constant signal
It sounds like you are gating the signal.
What is the "delay" -- a delay means to make something appear later than is otherwise would. What you show above isn't delaying anything -- it is outright suppressing some of the pulses.
And you talked about a 15 ns delay. Where did that 15 ns number come from?
Are you using this IR signal to transfer data, or is the goal merely to detect that it is present?
If the latter, then all you need to do is gate it (using the circuit you show above) with a suitable squarewave signal so that the receiver is seeing bursts -- as long as the receiver is designed to respond to such bursts instead of ignoring them as noise.
What is the "delay" -- a delay means to make something appear later than is otherwise would. What you show above isn't delaying anything -- it is outright suppressing some of the pulses.
And you talked about a 15 ns delay. Where did that 15 ns number come from?
Are you using this IR signal to transfer data, or is the goal merely to detect that it is present?
If the latter, then all you need to do is gate it (using the circuit you show above) with a suitable squarewave signal so that the receiver is seeing bursts -- as long as the receiver is designed to respond to such bursts instead of ignoring them as noise.
Yes my apologies
Not a delay it needs to be suppressed for a certain time to break the signal
The goal is just to detect its present. No data transfer required.
I have heard the term "gate it" before but don't know what it means
I thought of using an and gate but i don't know how to drive the other side of the and gate. I have nothing to trigger the other side of the and gate which is why i just wrote trigger in the picture because I'm not really sure how to trigger it
Any ideas on a simple way to do that?
"Gate" basically means to turn on and turn off under the control of a signal.
You can use a 555 timer circuit (or any of a host of other timing circuits) to provide a square wave into one input of the AND gate. When the signal is HI the 38 kHz pulses are passed through to the output and when the signal is LO they are blocked and the output is LO.
Another alternative is to use a simple, cheap microcontroller to generate the signal going to the LED driver in software -- but there IS a cost to get set up to program the microcontroller (MCU) and a learning curve to be able to write the program. So if you aren't going to be doing things that would benefit from an MCU at least time to time, I wouldn't go that route or I would pay someone else to do it for you.
Yes i understand that
Recall many posts ago i said i could do this with an tiny 85 MCU very easy
I am just trying to keep it discrete parts
So say i wanted to gate it
How would i calculate the frequency i need from the 555 timer to suppress for a certain time?
Recall many posts ago i said i could do this with an tiny 85 MCU very easy
I am just trying to keep it discrete parts
So say i wanted to gate it
How would i calculate the frequency i need from the 555 timer to suppress for a certain time?
That will depend on the receiver you are using. You need to craft a waveform that it will respond adequately to. Ideally, you can find one that will respond suitably to a string of pulse bursts.
I have one
https://www.sparkfun.com/datasheets/Sensors/Infrared/tsop382.pdf
Now exactly sure how to interpret the data sheet as far as the burst specs go
https://www.sparkfun.com/datasheets/Sensors/Infrared/tsop382.pdf
Now exactly sure how to interpret the data sheet as far as the burst specs go
A quick read would indicate that a burst needs to contain between 10 and 70 cycles and the gaps between bursts need to last for an amount of time equivalent to 10 cycles. So you might set the 555 so that it allows 50 cycles to get through and then blocks 20 cycles -- the exact numbers shouldn't be critical.
This result in a pulsing output from the receiver, but you can turn that into a solid signal using a retriggerable one-shot that lasts for about 100 cycles. That will mean that your output will be high for a minimum of about 2.6 ms anytime it sees a burst of IR and that it will remain high for that long after the last burst is over.
This result in a pulsing output from the receiver, but you can turn that into a solid signal using a retriggerable one-shot that lasts for about 100 cycles. That will mean that your output will be high for a minimum of about 2.6 ms anytime it sees a burst of IR and that it will remain high for that long after the last burst is over.
hmmmmm...............
