JP1 Remotes

[Tommy Tyler]

John is correct that LED duty cycle plays almost no part in the demodulation of a remote's transmissions. It's simply a means to help control the power to an LED so it doesn't burn out. To get very bright light output pulses that can be seen at great distances, the LED is pulsed at a current level that would destroy it quickly if left on continuously. To squeeze more transmission distance out of a remote the designers usually use less than a 50 percent duty cycle, which allows them to make the pulses a little brighter. But the receiver doesn't care about the width of a pulse, just the time between pulses.

The device in my article is 100 percent reliable at capturing all the IR pulses in a burst. You can easily tell by visually comparing successive records. They fall right on top of each other because they're all triggered by the same event, the start of a message. They may not have the same duty cycle as the actual LED, but they are all there. Measurement of carrier frequency is always possible by finding the longest burst you can, visually determining the number of pulses in that burst, and using the program to measure the time. So it's not a matter of whether it can be done, it's how accurately can it be done. A nice fat lead-in burst with 40 cycles or more of carrier will give pretty good accuracy. A protocol that skimps on the lead-in burst size will obviously not be as good. When you measure time from a point on one cycle of carrier (say the rising edge of the waveform) to the same point on another cycle, you essentially eliminate any effect the shape of the waveform has on frequency calculation.

Alas, Jon, I was unable to find a solution under about $500 for determining RF frequency. But you know the FCC ID number is usually conspicuously marked on any RF device, and you can always obtain the frequency if you know that number. Most FCC files these days seem to include the confidentiality option that excludes schematics and parts lists from the public eye, but the frequency is never confidential.

I'm sorry to throw a wet blanket on the ideas to add RF capability to IR remotes by just adding a transmitter that is gated on and off by a sample of the LED signal. I used to think that's the way it was done. Examination of the Dish remotes shows the IR modulation signal originates on one pin of the processor, and the RF modulation signal on another. That's because the signals are NOT the same. The RF signal looks like a demodulated copy of the IR signal. To do it otherwise would add unneccessary expense and complexity to the RF receiver. It would have to first demodulate the RF carrier, then demodulate the IR carrier. That redundancy would add nothing to performance.

Of course, the Dish remote is a sample of only one. I would like to take a look at the schematics wkearney99 wrote about (X10 UR47A and 15-2117) that use the same signal for IR and RF. Where can I see those?

[The Robman]

What about the URC-9910 and 15-2117 that use an RF puck to send a copy of the IR signal via RF? It seems to a non-electronics-expert like me that someone could fashion up a custom RF puck that could be added to URC-8910 and 15-2116 remotes to send an RF signal simulaniously with the IR signal.
_________________
Rob
www.hifi-remote.com
Please don't PM me with remote questions, post them in the forums so all the experts can help!

[classicsat]

You can use an L/C filter to simply filter the carrier from the data signal. My old Startrak UHF remotes (which are the source of the transmitters for my remote extenders), which use tpanard NEC uPD6020g IR remote chip, and an L/C filter on its IR output to the transmitter gates (which is a 433 Mhz SAW based transmitter). The trouble there is interfacing the receiver module to the equipment.

For the UEIC RF remotes, they modulate the carrier on the RF, so the base unit simply receives the rf singal and uses it to drive and LED (with drive transistors).

[skryd]

Before I order parts and build this.
Anyone actually using this and finding it useful?
I have 2 OFA 9910's I'd love to use with my X10 system without the IR543 boxes I'm using now.

[The Robman]

I have ordered parts to build it, nobody else actually has the device yet (other than Tommy). I should point out though that the purpose of this device is to be able to record and decode RF signals, it's not designed to be used to control anything.
_________________
Rob
www.hifi-remote.com
Please don't PM me with remote questions, post them in the forums so all the experts can help!

[jon_armstrong]

Tommy,

The UR47A schematic is here:
https://gullfoss2.fcc.gov/prod/oet/forms/blobs/retrieve.cgi?attachment_id=45655&native_or_pdf=pdf

There are a couple of other links about x10 hacks for the TM751 that is the receiver for the transmitter in the UR47A:
http://www.geocities.com/ido_bartana/TM751_RF_to_RS232_Converter.htm
http://www.geocities.com/ido_bartana/TM751_RF_to_IR_Converter.htm

and the home page for all x10 hacks:
http://www.geocities.com/ido_bartana/

I agree that the FCC is a very straight forward source of RF frequency information. I'm just wondering how my car can learn my garage door opener command so well
_________________
-Jon

[Tommy Tyler]

I may have to retract my previous comment about dual (IR/RF) remotes and what they transmit. After Jon pointed me to the schematic of the UR47A that, sure enough, gates the RF transmitter on and off with the same signal that drives the IR LED, I dug a little deeper. Specifically, I was mystified as to how a pair of "cone" or "pyramid" type IR extenders were seemingly demodulating an IR signal, transmitting it as RF, then converting the received RF signal back into the correct IR carrier frequency. A 40KHz TV signal was reproduced at the RF receiver as a 40KHz IR signal, and a 58KHz VCR signal as a 58KHz IR signal. "Seemingly" is the key word. In actuality, they weren't demodulating the IR at all. In each case the RF transmission was being modulated at the IR carrier frequency. What fooled me was the fact that the RF Sensor described in my article was integrating the IR pulses because they were so close together, and so the resulting record on DigiTrace looked like demodulated IR.

Here's the experiment for anyone who wants to see this for themselves. When you remove the top of one of those cones, you see a small tin box, which is the shield surrounding the highly sensitive optical receiver. A few holes at one end are where the light goes in. Sitting on top of that is a triangular-shaped circuit board, which is the UHF transmitter. The little "C"-shaped loop of wire on the back side is the transmitting antenna. If you hold the RF Sensor close to the UHF transmitter and fire an IR remote at the tin box, the resulting record on DigiTrace looks like the demodulated IR signal. But if you carefully back the RF sensor away from the transmitter a couple of inches as you keep trying, suddenly the resulting record will show lots of IR pulses. That's the point where the detector is not being overloaded by signal strength, and so is able to discriminate individual RF bursts on the order of 10 microseconds wide.

It is possible that my RF sensor design could be modified so that it would detect those pulses more readily. The obvious thing that comes to mind is to reduce the detector capacitor from 100pF to maybe 10pF. After all, the RF carrier is usually several thousand times the IR carrier frequency, so there's plenty of room to work with.

I don't plan to do any further work toward that end unless someone convinces me there's a need for it. The RF Sensor was designed primarily to receive signals from RF-only remote controls, not dual IR/RF remotes. You can usually tell if a dual remote is modulating the RF at the IR carrier frequency by visual inspection. If the UHF transmitter and IR LED signals come from the same place, it is. If they come from separate pins of the processor, it isn't.

Rob, all this confirms that a UHF puck could be stuck inside a remote and connected appropriately to the LED signal, and it would send out an RF signal OK. But what would it be sending it to? You would have to have a puck whose frequency matches the receiving device you want to control. Do you have such a combination? As far as designing and/or building UHF transmitters for a specific frequency, that is probably hopelessly beyond the skills of anyone in our group, not to mention the legal implications FCC). My crude little RF Sensor works only because you can stick it right up the kazoo of the transmitter you are working with, to get a strong signal. Trying to make something that picks up a UHF signal from 50 or 100 feet away is light-years more complicated, certainly beyond my ability.

Tommy

[Wheelie4]

Tommy,

I'm guessing one reason for Rob's idea about building a puck for the 8910 is this. The URC9910 comes with a RF to IR transceiver. If the 9910 remote breaks or dies or it's puck becomes inoperatable you have to pay for another 9910 and a non-needed transceiver.

If an inexpensive puck could be built then someone could just buy a 8910 and add a puck without having to pay the higher cost for the 9910 and a transceiver (another transceiver thats not needed).
_________________
Daryl

[The Robman]

[mr_d_p_gumby]

[mr_d_p_gumby]

Rob, I have the feeling that your next question will be, "Can I re-tune the RF puck to control an RF device that operates on another frequency?"

The answer to that is not simple. Unless you have the proper equipment, you'd be trying to re-tune the transmitter blind, and that can be a hit-or-miss proposition. If it were a Survivor challenge, I'm sure you'd find a way, but in reality, it's much harder to do. First, the means of tuning the transmitter has to be identified. This can be a variable capacitor (rare), variable inductor (with tuning slug), or it may be done by bending a wire-wound inductor. In cases where the transmitter uses a SAW device, there may be no tuning means at all.

Assuming you can properly re-tune the transmitter, you will probably find that does not have very good range. While these circuits tend to look very simple, what is not obvious is that a great deal of effort has gone into making the transmitter FCC compliant. This means suppressing harmonics of the fundamental carrier frequency. Translation: it won't be very good at transmitting on other frequencies because the pcb layout and other components have been optimized for operation on one frequency.

I'm not trying to say it's impossible. Certainly, given enough time and patience, you could do it. But I'd classify this well outside of the easy-to-do mods range. You'd be much better off finding an RF transmitter that is already set for the correct frequency.
_________________
Mike England

• JP1 Group at Yahoo!, JP1 Files Area

[Tommy Tyler]

I second that. Even replicating an existing puck design could be tough without RF test equipment. That's what I hate about RF. You can't taste it, hear it. see it, or feel it. I'm not even sure it's not all a hoax.

Tommy

[The Robman]

[jon_armstrong]

A couple more questions for the group. Tommy, (assuming away physical constraints and battery life iisues) do you think that you could drop in the UR47A's RF daughter board into a JP1 remote and with a few parts make it work? Now back to power, is it possible that one of the control bits in the executor can turn off/on the RF?

If so, the TM751 mods might be interesting -- the links are in my last post. The RS-232 mod is really just a concept but going from a JP1 remote to RS-232 but could be of real interest. A lot of AV gear, particularly projectors, have an RS-232 interface and in most (and maybe all) cases the RS-232 command set is much larger than that of the IR commands. There are some companies that specialize in creating an IR to RS-232 interface to allow IR control of some Plasma displays and projectors since the built in IR command set doesn't have discrete power and/or inputs. (Personally, I'd be pretty mad at spending >$10K for a display and having only toggle commands.)

RTI Theater Touch has a very expensive IR/RF remote that does RF to RS-232 that sells to that market.

John Fine figured out a couple of years ago that some IR commands are really Async serial commands. Most are run at 1200 baud and have a start bit, eight bits of data and one or two stop bits and maybe a parity bit. The IR keyboards are usually like that. So it seems like the JP1/OFA protocol side would be easy -- but like the old saying goes: "Everything is easy when you don't know what you're talking about." Which sums up my real knowledge about this stuff.
_________________
-Jon

[Tommy Tyler]

Jon,

The daughter board of the UR47A has the following characteristics:
1. It needs a 6VDC supply, a complication for 2-cell remotes.
2. It is enabled when the ENABLE input is high or disconnected.
3. When enabled, it will transmit when the SIGNAL input is high.

My guess is that the remote uses about 10 percent more energy when transmitting an RF signal along with an IR signal. So the ENABLE signal may be only for extending battery life.

About half of all remotes use positive-going signals to turn on the IR LED, but the other half use negative-going signals, and you would need a few more parts to make that compatible with the daughter board.

So the bottom line is that (notwithstanding physical constraints) you could PROBABLY connect the daughter board to most remotes, but there are no generalizations. Each remote would require individual considerations, and satisfactory battery life may or may not be achievable.