Getting to grips with IR


We have put this first so you can understand the text when you get to it, no point in struggling through the text only to find this lot at the back! There isn’t much because it is not rocket science.

IR = Infra red

STB = set top box but is actually anything that can be IR controlled, VCR, amplifier, TV etc.

1 KHZ = 1 Kilohertz = 1000 Hz

1 GHz = 1 Gigahertz = 1000,000,000 Hz

1 mS = 1 millisecond = 1/1000th of a second (1 thousandth)

1 uS = 1 microsecond = 1/1000, 000 of a second (1 millionth)

So 0.25S would be 250mS (1000mS is one second)

0.025S would be 25mS

0.0025S would be 2.5mS

0.00025S would be 250uS (1000uS being 1mS)

and finally 0.000025S is 25uS  if you persevere with the text you'll see where you need this.

Why on earth would I want to get an IR distribution amplifier?

I was hoping you would ask that. These days with the number of set top boxes and switchers, recorders etc increasing all the time it is often convenient, in fact sometimes essential for domestic harmony, to be able to locate much of the equipment in a small room or cupboard out of sight.

Now unless you want to get up and go to the cupboard / room / loft or wherever every time you want to change the channel or volume, you will need some means of controlling the gear from you armchair.

Also as we are getting more Audio / Video signals distributed around the house by Cat5 and similar systems, being able to control the system from the bedroom or kitchen is a big bonus.

Installing an IR distribution system can allow you to do just that and is very cost effective.  It makes use of the equipments existing infrared control mechanism, so there is no massive outlay in hardware, and all of the function that were controllable from your armchair are now controllable from your bedroom lounge etc.  IR control is very universal and most other control systems such as X10 can accommodate IR control, if you already have a home automation system you can extend that so you can control the home automation devices from your bedroom or kitchen just by adding an IR distribution system.

Any decent Cat5 based AV distribution (well ours does) system or RF distribution system, will have some form of IR back from the display end to the source. The distribution amplifiers can mix all that lot together, add in IR receivers from other rooms and amplify the lot so that it can control a bank – or even a number of banks – of AV gear. Sorry that’s the daily exercise out the window then!

Input and output connections are typically via 3.5mm stereo jack sockets and for shorter distances 3.5mm headphone extension cables can be used, for longer distances cat5 is readily available and easily handled. So installation is usually pretty simple.

How does infrared remote control work?

I think we are all aware that infrared is just light that we cannot normally see. It is light at the low (red) end of the visible spectrum, and shares many of the properties of visible light.

Like visible light infrared travels in straight lines and will not go around a corner unless there is something about that will reflect it or refract it. Also like visible light it is reflected by white surfaces and usually absorbed by dark surfaces.

It is worth noting that IR emitter LEDs, the sort of thing that is used in remote controls etc, are like a torch and they have a narrow beam, and will only "shine" onto a narrow area.  A 10 or 15-degree beam width is not unusual.

It is not necessary to understand – or for that matter even read - the following detail on IR codes, but we have produced it here for those that are interested, would like to get the best from there system, or are just so bored they’ll read anything!

I suppose the first thing to understand is that the remote control is “talking” to a small microcontroller inside the TV or VCR etc, so the instruction (play, pause...) is in a form that computers can understand, made up of a sequence of  '0' and '1'. This stream or sequence can contain up to as many as 50 digits.  Often the code will contain simple error checking mechanisms such as sending the data first in the right order and then reversing the order and sending it again. The microcontroller can then compare both halves of the data to see if there is any receive errors

Each button pressed will result in the remote control sending a different instruction code to the receiving microcontroller. Usually there will be 2 parts to the code. The first part identifies the general device type that the code is intended for. This “DEVICE” code could be for example “VCR”. The device code is sometimes referred to as “ADDRESS” code. The second part would be the “COMMAND” codes i.e. “PLAY”. Some IR code schemes are adopted by a number of manufacturers and in this case the code may well contain a third “CUSTOMER” section to identify different the manufacturers. It soon becomes easy to see why as many as 50 digits can be needed. A complete code might go something like... 011010 11100011 10101101 being CUSTOMER then DEVICE then COMMAND codes The whole lot could then be sent in reverse order or inverted (“1” replaced by “0” and “0” replaced by “1”) giving a total length of 44 bits to this hypothetical code (each digit, 1 or 0 is called a “bit” in computer speak)

Most infrared remote controls use a carrier-based system. In a carrier based system the infrared light beam is switched on and off at the carrier frequency. (For example 40 KHz or 40000 times a second) This is much quicker that the human eye can follow and if you could see it, it would appear to glow continuously, rather like the visible red LED that is sometimes also in the remote control.

Carrier = rapid flashing of the LED, the LED is first on for a short time and then off for a short time - repeated at the carrier frequency that’s 40000 times a second for 40 KHz carrier.

Space = no-carrier = nothing = the LED is just off continuously for this period of time.

A '1' might be sent by infrared as a burst of no-carrier for say 600uS followed by a burst of carrier for say 1200uS. The '0' might be a burst of no-carrier for 600uS followed by carrier for 600uS. This type system is called Pulse width modulation or PWM. It is probably the most common system and is used by Sony® Toshiba® and many others. The length of the carrier portion changes either 1200 or 600 while the no-carrier portion remain the same. Another system “REC-80” is space coded where the length of the no-carrier part changes while the carrier part remains the same length.

Example PWM data is “101”

No-carrier 600uS carrier 1200uS, no-carrier 600uS carrier 600uS, no-carrier 600uS carrier 1200uS.


Example SPACE coded data is “101”

No-carrier 1200uS carrier 600uS, no-carrier 600uS carrier 600uS, no-carrier 1200uS carrier 600uS.



There is another common system where sometimes the “1” is sent as carrier burst followed by an equal length no-carrier burst. In this system the “0” is the opposite of this being a burst of no-carrier followed by a burst of carrier. This system is called bi-phase modulation and is used in Philips® RC5 based IR code as well as some others.  Here we might have 600uS of carrier followed by 600uS of no-carrier for a digital “1” while the digital “0” would be sent as no-carrier for 600uS followed by carrier for 600uS. Bi-phase code is sometimes referred to as “Manchester code” as one of its first recorded uses was by Dr G. E. Thomas on the “Manchester Mark 1” computer in 1949


Example Bi-Phase data  “101”

 Carrier 600uS no-carrier1200uS carrier 1200us no-carrier 600uS.



As the “1” ends with 600uS of no-carrier and the following “0” starts with 600uS of no-carrier you get a 1200uS no-carrier burst when the two are next to each other.

In any of the above systems there are usually extra long bursts of carrier at the beginning and a long space at the end of the code so that the microcontroller knows when it at the beginning or has finished etc. If the button is held down the code will repeat after a relatively large gap. You might have to wait a whole 40mS or 0.04 Seconds for the code to repeat!

Why is a carrier used?

Good question, a carrier is used because it allows the receiver system to be designed so that it can filter out noise as well as IR codes based at other frequencies, resulting in a more robust and reliable system. In the above system if the carrier was 40KHz then the LED would flash on then off exactly 24 times in the 600uS burst of carrier.

Some plastics can be made that are seen as very dark almost black by a human eye but will pass infrared light through relatively unimpeded. These are used as daylight filters in IR receivers and the Keene Standard receiver housing is made from such a material.  If you’ve not seen one they look like a strangely shaped lump of black plastic

Scope grab of a good 40 KHz based code, top window is the complete code while the bottom window is zoomed onto one 600uS burst of carrier. If you look carefully you should be able to count the 24 pulses of the carrier. All grabs done with a Keene Wideband receiver


The first job the STB has to do is demodulate or remove the carrier from the signal, leaving just the data. Because the code is often learnt by learning remotes and generally played about with, there is quite a wide tolerance in the IR system and typically errors in burst / space times of about 10% can be accommodated by the decoding microcontroller. So if it were expecting a pulse of 600uS it would usually be happy with anything from 550uS to 650uS

Demodulated IR code, the upper window is the complete code while the lower window is zoomed in and shows just the bracketed section. Note absence of carrier. In this picture the output is inverted, the trace is normally high (at the top) and goes low when the signal is present. Most IR demodulation ICs operate this way.


Whether the signal includes the carrier or not can lead to difficulties when trying to make use of the IR ports on many of the better Surround Sound amplifiers.  Some expect the signal to include the carrier while others expect it to be just data. And if it is just data should it be normal or inverted?

Get info from Yamaha / Denon/ Xantech

Different companies all have there own variations on the above theme.

Sky use a 36 KHz based mix of PWM and Bi-phase code

Sony use a 40 KHz based

Most equipment manufactured by Toshiba, Panasonic, Samsung, Philips, Marantz and the majority of box suppliers use IR codes in the 36 KHz – 40 KHz band

More info on remote controls and IR signal systems can be found on the net at places like  Don’t know why we should be telling you this as they are a competitor of ours – but never the less the forum is very good particularly if you have one of the Philips® pronto based remotes

Are there any other types of IR codes?

There are some companies that use code that is not carrier based. It is just a sequence of IR pulses, the duration and spacing of the pulses defines the IR code much as the demodulated carrier based code does. The reason for the pulse-based code is usually that it is possible to generate a complete code in about 1mS, that’s about 20 times quicker than carrier based code. The extra speed from pulse code is usually done so that the system can be used for multi player games without blocking each other and still getting a quick response.

Even carrier based code is quick compared to the human operating the remote, most codes will repeat at least 10 times in one second even allowing for the relatively large gap that the remote leaves between codes blocks. In some codes the actual data is different if it is a repeat code. (Button held down continuously)

The downside to pulse based IR systems is that they are more likely to be affected by noise in the environment and just one pulse of noise arriving at the wrong moment can render the code unreadable, although because the code is so short there is less likelihood of that happening.

Pulse code does add extra difficulty to IR distribution. IR amplifiers use the first few microseconds of a code to set up its own sensitivity (another reason for the long carrier burst in traditional code) so that it can receive a clean undistorted signal; this is difficult with the very fast codes. But because the code is very quick the amplifier can use the first block of code to setup its own gain and then hold that level for a few mS to receive the next block cleanly. Secondly the signal being very quick means that a wide bandwidth receiver needs to be used, this, of course, means that more thought must be given to positioning the receiver. Finally the rest IR distribution chain must be designed to handle the wide bandwidth signals.

Some of the Pace manufactured cable boxes use this type of code, it is based around a code system at 115 KHz, and the pulses are particularly narrow and very strong. They can easily overload an IR receiver amplifier, particularly in the first few microseconds until the amplifiers AGC, or automatic gain control, has had the time to adjust to the input

IRDA and 2 way communication

IRDA is an infrared communication system. It is a 2 way system and is employed by computers, phones and other PC peripherals. It is very loosely based around the old RS232 serial computer system and the supported speeds tend to follow this standard. (Sorry no room here to go into detail on RS232) There are a lot of different types and speeds of IRDA and anyone who is interested can get more information from . (Unfortunately you have to pay for most of it!)

The fact that it is 2 way makes it very difficult to handle with an IR distribution system as you would require a minimum of 2 systems. One from point A to point B and the other back from point B to point A There is a considerable risk of a type of IR feedback where the output of one amp is then amplified by the other only to get amplified again by the original amp ad infinitum. It has been done with careful receiver positioning but I wouldn’t want to try it

Infrared keyboards and mice are often IRDA based and as such are not easy to handle with IR distribution.

One reason for mentioning IRDA is that the cable box code mentioned above is often referred to as IRDA even though it is only one way.  115.2 KHz is one of the RS232 based speeds.

IRDA based code displayed at a much faster time base than the carrier based ones above  - 100uS per div


Another view of the Pace code code is repeating with button held down


What do I need to know to get the best from my system?

Guidance on positioning the receivers etc

There are many other sources of infrared, some are fixed like sunlight and some have a carrier frequency like the infrared off LCD and plasma panels and fluorescent lights etc. Our standard receiver and most of the receivers built into TVs and STBs etc are filtered to remove the effect of sunlight and reduce the effects of interference. But sometimes you need to receive a larger range of signal frequencies. See the scope grabs below, LCD tend to generate more interference but at a lower intensity while Plasma screens generate less but the pulses can be greater.

Not all carrier based IR controlled boxes work at in the 36 KHz to 40 KHz band, some work at a lower frequency and some higher. Then there are the pulse-based codes, a range of about 20 KHz to 120 KHz will accept most. BUT into this range also fall some of the interference sources. So if you opt for a wideband receiver you would need to position it more carefully than a standard one.

Bear in mind that Infrared is line of sight and is reflected by white surfaces. Position the IR receiver so that the infrared from the interference sources does not fall directly onto the IR receiver’ Try to position the receiver directly beneath or to one side of the screen and slightly behind the face of the screen, do not have something reflective directly opposite and try to find a spot that will not be in direct sunlight.

If you are using an IR distribution amplifier it is imperative to make sure that the equipment you are trying to control can only see one source of IR. If the original IR from the remote handset as well as the distributed signal can both find a path to the equipment’s eye then the signal will be garbled and the equipment will not respond.

This can occur even if one of the signals is very feint. The majority of the problem calls we get are down to this!

This is also true if 2 or more of the distribution amplifier receivers can receive a version of the IR code, again the signals get mixed together and because of the time and path differences you get a garbled code.  This can be one via say an RF link like a Powermid and another Cat5 based systems. The possibilities are almost endless.

The very high output learning remotes like the excellent Phillips Pronto range need particular attention because they seem to find a route from some amazing distances.

The top window here appears to show an almost normal 40KHz based code but a look at the zoomed in lower window shows that the carrier is seriously garbled. This was obtained by using 2 almost adjacent receivers into an IRBKIT


Interference from an LCD screen at about 0.5 meters The IR receiver was looking directly into the screen


Interference from a Plasma screen at about 4 meters The IR receiver was looking directly into the screen


OK so what does an IR distribution system need to do?

An IR distribution system is basically expected to receive the IR signal and then send it out or “emit it” again in another location. Over the years they have evolved, and now the IR receiver, the distribution amplifier and the emitter are usually separate components.

The receiver extracts the required IR signal from any interference and other infrared signals and converts it into an electrical signal that can be amplified by the distribution amplifier. 

The amplifier can accept one or many of the signals from the receivers and provide a number of outputs, each of which is of a sufficient level to power an IR emitter.

The IR emitters accept the electrical signal from the distribution amplifier and convert it into Infrared light. They are usually constructed from an infrared LED in a small housing.

To drive a typical IR emitter to a level to operate, as a remote control requires much more output than it does to illuminate a visible LED. With modern LEDs you can easily see a visible red LED at 10m with just 1mA (1/1000th of an amp) of current flowing through the LED. In contrast it is not uncommon to pulse IR LEDs with well over 100mA or even up to 1 Amp or more if you want them to operate at 10m or so. A note here, visible red LEDs also emit a small amount of infrared and if you are only going a very short distance, like the emitters that go directly over the receiver eye, you can use a red led – they can be cheaper and give you visible feedback.

In the original remote the IR diode is typically connected directly to the battery when the IR pulse is required, a fast transistor switch does this. The pulse half of each cycle of carrier is often kept shorter that the space half in order to get the best range and battery life out of the remote control. In a 40 KHz signal the time for a complete cycle is 25uS to get better distance and battery life this could be an “on” pulse of 10uS and an “off” pulse of 15uS.

The maths! For those who wandered where we got the 25uS from. If something is happening 40,000 times a second then the time required for just 1 to happen is going to be 1 second / 40,000 which my Windows calculator tells me is 0.000025 of a second  or 25uS

So with a carrier frequency of 40KHz the IR emitter is flashing “on” for a bit and then “off” for a bit, the “on” and “off” times add up to make the cycle time and that cycle time is 25uS.  Pretty damn quick eh.

Another very useful function on top end IR distribution amplifiers is the IR matrix. Using the matrix you can route the electrical signals from different IR receivers to different outputs. This way it is possible to have for example 2 Sky boxes in the same equipment rack and still control one box from the lounge and the other from the bedrooms etc You can still have both areas control all (or just some) of the other equipment if you like! Flexible or what?

WOW I’m think I need to install an IR distribution system what decisions do I need to make?

Firstly you need to decide on a system – that bit is easy – a Keene Electronics one!

Next you have to work out what you will need to control and decide on its location.

Once you know this you can start to work out the emitters you will require.

Next decision is the number and type of IR receivers you need.

If you are making use of any of our Cat5 AV distribution systems then they you can feed the IR outputs from these into one of the distribution amplifier inputs. If not you will need an IR receiver for each area in which you need IR control.

The type of receivers is easy. Always start with the standard receiver! Only go for the wideband if you need it. No point in making things any more difficult than they need to be. We always let people send the standard receiver back and pay just the difference to get a wideband if required. You can do this through us even if you bought the original system through one of our distributors.

Finally once you know how many IR receiver sources you require you can decide on the actual amplifier. If you only have a couple of sources than the standard amplifier is probably going to be the one for you if however you have 3 or more IR sources then it would be better to think in terms of the IR commander. Fully loaded this unit can handle up to at least 8 IR receiver sources and a wealth of outputs. Similarly if you need to be able to route IR signals from different areas differently then you would definitely require the IR commander matrix unit.

 If you are installing into a cabinet a useful trick can be to pay attention to the colour etc of the interior walls of the cabinet. If you make them white and the inside of the doors white it is possible to get away with fewer emitters as a high power emitter blasting around inside the cabinet will probably operate most devices. Note though that the doors will need to be closed for this to work open the doors and suddenly all stops working.

Click on the menu on the left to see the Keene Electronics family of IR distribution Products