IR carrier of the AXE040/TVR010 -- is 41.6KHz or 40kHz or 38KHz or ... ?

I'm about to start playing with the Picaxe IR remote control, and have been looking into the correct carrier frequency to use for the receiver so I can order some extra receivers.

I found a thread "Question on Infrared Modulation Frequency" from 2012 that suggests that the correct Sony SIRC carrier frequency is 40KHz, the Picaxe output is 41.6KHz, and the industry standard has migrated to 38KHz.

So, my first question is: What is the carrier frequency used by the AXE 40? Is it the Picaxe 41.6KHz? If yes, what is the frequency of the IR receiver that is shipped in the kit?

Nick12ab, in the above thread, found a chart that suggests that "responsivity" is down by 50% when the frequency is mis-matched between 38KHz and 40KHz, and I would like to avoid that and get as much out of this setup as I can.

In order to join the industry, should I buy 38KHz receivers and accept the reduced range from the AXE040?

Any clarification would be appreciated.

I never got this answered as well.

My only suggestion is to get an AXE40 , take the cover off and measure the frequency at the IR LED with your scope or with the frequency counter feature of your DVM. Then knowing the frequency you can select the correct receivers.

I've been using Vishay 38KHz receivers at a distance of ~15 feet in direct sunligh with accurate results. Haven't tried any farther.

What distance are you trying to achieve? If it is just across a room it really doesn't matter, in reality there will be very little difference in performance between a 38KHz or 40kHz rated receiver and so just buy whatever is locally available.

What is far more important to range is the infra-red LED current in the transmitter. That is obviously a fixed circuit in a TV style remote, but if building your own circuit you can buffer the output with a FET to get an increased current through the LED.

Technical said:

What distance are you trying to achieve? If it is just across a room it really doesn't matter, in reality there will be very little difference in performance between a 38KHz or 40kHz rated receiver and so just buy whatever is locally available.

What is far more important to range is the infra-red LED current in the transmitter. That is obviously a fixed circuit in a TV style remote, but if building your own circuit you can buffer the output with a FET to get an increased current through the LED.

Click to expand...

i am using a TV-style remote, initially the AXE040. This is for use with high end hifi systems and the rooms can get very large so I want to get as much distance as I can without resorting to increasing the current through the transmitter. Anyway, I don't want to throw away any range because the transmitter and receiver are mis-matched when a little work could result in a proper match. So, I'm still looking to optimize the receiver. It's curious this issue cannot be resolved, but the AXE kit does not include any information on frequencies used.

So the question is really 'what is the exact frequency of the TVR010A TV style remote control (as supplied in AXE040) whilst in Sony emulation mode' ?
The answer is we have never measured it as it works just fine. The remote is a mass produced device simply bought in from China.
But we might get the Logic analyser out and take a look!

My experiments a while back showed that the RevEd supplied IR receiver worked fine upto 70KHz, so I don't think a difference of +/- 2KHz is going to make any significant difference in performance, no matter how big your room is. The cheapo Chinese remotes probably have a spread of frequencies around that figure anyway, I've never seen a crystal in one.

The phenomenal success of IR remote control is testament to its tolerance of batteries that have been in there for ten years, and a spectacularly non-optimum optical system.

I wouldn't bother getting hung up on this frequency malarkey, just build one and it will do the job !.

buzby said:

i wouldn't bother getting hung up on this frequency malarkey, just build one and it will do the job !.

Click to expand...

ok! Lol

I don't believe that frequency matching for optimal performance is "malarkey". For optimum performance, which is what Wapo was looking for, the TX and RX frequencies should be matched as closely as possible. If you KNOW that the TX carrier is 40Khz why would anyone intentionally select a 38KHz receiver? If working with Picaxe generated IR it would be prudent to use a 40KHz receiver, vs a 38KHz receiver.

However, I do agree that many times "good enough" is good enough. But to call designing for optimal performance "malarkey" is, well .... malarkey.

I have an old Heathkit frequency counter I built in another lifetime. When my AXE040 arrives, I will take it off the shelf and see if it still works. If I see anything useful, I'll post it to this thread.

Goeytex said:

I don't believe that frequency matching for optimal performance is "malarkey". For optimum performance, which is what Wapo was looking for, the TX and RX frequencies should be matched as closely as possible. If you KNOW that the TX carrier is 40Khz why would anyone intentionally select a 38KHz receiver? If working with Picaxe generated IR it would be prudent to use a 40KHz receiver, vs a 38KHz receiver.

However, I do agree that many times "good enough" is good enough. But to call designing for optimal performance "malarkey" is, well .... malarkey.

Click to expand...

I don't know why you would use a 40KHz receiver for PICAXE generated IR, when manual 2 specifically states the IROUT command uses a 38KHz modulated signal. I intentionally picked a 38KHz receiver because I figured if the IROUT command used the SIRC protocol at 38KHz, then more than likely the IRIN command would as well?

Hemi345 said:

I don't know why you would use a 40KHz receiver for PICAXE generated IR, when manual 2 specifically states the IROUT command uses a 38KHz modulated signal. I intentionally picked a 38KHz receiver because I figured if the IROUT command used the SIRC protocol at 38KHz, then more than likely the IRIN command would as well?

Click to expand...

IF you take the time to actually measure the Picaxe generated carrier, you will find that it is NOT 38KHz, regardless of what the datasheet says. The manual is simply wrong, providing an erroneous figure.

On 20X2 it is 41.67Khz with a duty cycle of 33.3%. Other Picaxe Chips I have tested are the same. This represents about a 9% error. So with this information you can pull up the datasheet for a Vishay receiver and look at the chart of responsivity vs frequency and calculate the loss of receiver responsivity. It is quite significant at about 50%.

I just looked at a Vishay TSOP22xx (xx is Fq in kHz) datasheet.
They provide a plot of Relative Responsivity vs f/f0 Relative Frequency.

That will give a good starter on 'Best' versus Your Fq.

I would imagine for most hobbyists who just want to blast across their kitchen the Fq deviation is neither here nor there.
In many cases 'good enough' is fine.
But if you want optimum performance then it's up you to measure things and select the best component for the job.

Farnell, and many others, sell a range of modules with various centre Fqs.
http://uk.farnell.com/jsp/search/browse.jsp?N=2031+204197&Ntk=gensearch&Ntt=ir+receiver&Ntx=mode+matchallpartial
But I realise that's only any use when you get the actual timing info.
So, out with the 'scope

Edit: I see some info has appeared... but I guess you want the equivalent for the Chinese remote.

Interesting. I don't have the tools to test it so I was going by the manual. Thanks for testing that.

Dippy said:

Edit: I see some info has appeared... but I guess you want the equivalent for the Chinese remote.

Click to expand...

Exactly, I am not building a Picaxe transmitter, I want to use commercially available transmitters and I have two AXE040 kits on order for experimentation. The ultimate goal is to build Picaxe-based receivers to work with commercial remotes.

There seems to be no information published on the AXE40 remote; maybe what ships is whatever is available at the time and the specs are not specified because they cannot be specified because suppliers are given only loose specs to work with.

I guess if the industry is headed toward 38KHz the AXE040 probably is going to turn out to be 38KHz which is fine with me, I just need to know; I'll check them out when I get them.

This thread is getting a bit muddled again - focusing on the carrier frequency of the irout command when actually the first post was actually about the TVR010 remote control, which is something completely different. The vast majority of people using IR control with PICAXE do actually use this second method - a separate TV style remote.

1) TVR010 frequency - don't know. We will take one apart and test it when we have a minute.

2) IROUT command
Last time this was discussed several months the online irout command description was corrected to state the nominal value 40kHz instead of 38KHz (40KHz is actually correct for the 'official' Sony SIRC spec, whilst other protocols such as Philips use 38kHz). The manual has also been corrected for future versions. You can see this correction at

www.picaxe.com/commands/irout

Each particular chip e.g. M2 or X2 may vary a bit from this nominal value, up to around 5% or so. This is due to other processing that may go on in the PICAXE firmware. However this makes almost no difference to the effectiveness of the irout command, in fact the current through the IR LED is what most affects the range, not the carrier frequency.

In reality it makes very little difference to your project if you use a 38kHz or 40kHz spec'ed receiver. Both will work fine over a distance of several metres.

If it helps, I've used a bunch of different remotes with the 38KHz receivers. One was a Sony remote that came with my TV (non-universal) so Sony devices only but it worked fine. The rest are Sony branded universals (I think they're a decent inexpensive remote) and a very old One-For-All. I don't have the time to see if they list the frequencies used or even if that information is available, but when programmed to use Sony equipment (using the universal device codes), they have all worked reliably with my PICAXE projects. Gotta love the purchase history feature of websites nowadays from my amazon history:

Sony RM-VLZ620
Sony RM-VL600
Sony RM-V302 (cheap remote I keep in my workshop to test with - replaced the one below)

One-For-All (don't remember the model number, but it looks like this and it worked fine as well when programmed for a Sony TV.

In reality it makes very little difference to your project if you use a 38kHz or 40kHz spec'ed receiver. Both will work fine over a distance of several metres.

Click to expand...

But are we talking about "several meters" or optimal performance?

The receiver datasheets are clear. Responsivity significantly, if not drastically affects range. I do not understand why you pooh-pooh on frequency matching/optimization while shifting the conversation to LED current. Both are important and with proper frequency matching less LED current will be required to get the same range.

The notion of "the heck with frequency let just blast out way out", does not work for me.

Goeytex said:

I do not understand why you pooh-pooh on frequency matching/optimization while shifting the conversation to LED current.

Click to expand...

Because, several years ago, we tried out several brands and makes of IR receiver to see which would be the best for us to sell as part LED020. We had about 12 in a line, each connected to its own 08M, with receivers ranging from 36 to 42KHz and from about 4 different manufacturers, with a purchase price difference of up to £1 between some of those parts.

We then simply walked backwards down the warehouse with 3 transmitters, which had increased range in this order

1) a IR LED with 330 ohm directly to PICAXE output pin
2) a tvr010 remote
3) a IR LED with 33 ohm via FET to PICAXE output pin

In the first two cases all of the test boards stopped working at exactly the same distance. The warehouse wasn't long enough for 3) to stop working! The spec (and cost) of the receiver made no noticeable difference at all in this (admittedly not very scientific) test.

So yes, we totally agree with you that if you just read the datasheet it should be important to match the carrier frequency and of course there is no harm in doing this if you want to.

However in reality in our practical tests the carrier frequency spec. of the receiver actually made no difference at all, whilst the resistor on the IR LED made a huge difference (many people do not realise the very large currents IR LEDs can take (when compared to normal LEDs), which is why we always highlight this point).

Did you test any Vishay Receivers?

What make & model did you settle on?
What center frequency?
Does it have AGC ?
Is there a real datasheet available?

If the receiver drops to 50% sensitivity at +2KHz, what was mine doing when it was receiving at 70KHz ?.

As I said before, IR remote control is very tolerant of misconfiguration and misuse, otherwise we'd all be still getting off the sofa to change channels.

I quite understand the desire to match components for optimum performance, but sometimes le mieux est l'ennemi du bien.

I do not believe that "responsivity" and "sensitivity" are the same thing. Two different words with two different meanings.

I stand corrected.

http://www.its.bldrdoc.gov/fs-1037/dir-031/_4587.htm ( But they are close ! )

Edit : Full frame version http://www.its.bldrdoc.gov/fs-1037/fs-1037c.htm

We took a TVR010 apart and hooked up a LOGIC and found in Sony it runs at about 40.8 kHz as shown by the attached screenshot.

The LED020 datasheet is here www.picaxe.com/docs/led020.pdf. Although the LED020 is indeed centred around 38MHz it will give at least 90% distance efficiency to anything from approx 36 to 41.5kHz and at least 80% from 33 to 44kHz, so has a very large useable range around the centre frequency. It also has quite a large voltage range (2.7V to 5.5V) compared to some other parts.

So in summary a 38kHz receiver like LED020 is perfectly fine for use with a PICAXE chip 'irout' and for a TVR010, but if you really want to be more 'centred' then use a 40kHz receiver for both. In reality it will probably make very little difference to your project.

I've never bothered reading the spec of the IR receiver before, but looking at it now it's obvious that this device is designed to have a wide tolerance on just about all of it's parameters.

This means it's probably not the 'best' choice for maximum range work, but it does mean it will still do about 11 metres with a transmitter which is off frequency by 5KHz. This is still plenty enough range for most of it's intended applications.

Technical said:

We took a TVR010 apart and hooked up a LOGIC and found in Sony it runs at about 40.8 kHz as shown by the attached screenshot.

The LED020 datasheet is here www.picaxe.com/docs/led020.pdf. Although the LED020 is indeed centred around 38MHz it will give at least 90% distance efficiency to anything from approx 36 to 41.5kHz and at least 80% from 33 to 44kHz, so has a very large useable range around the centre frequency. It also has quite a large voltage range (2.7V to 5.5V) compared to some other parts.

So in summary a 38kHz receiver like LED020 is perfectly fine for use with a PICAXE chip 'irout' and for a TVR010, but if you really want to be more 'centred' then use a 40kHz receiver for both. In reality it will probably make very little difference to your project.

Click to expand...

Technical, thank you for taking the time to do this. It certainly clears the air authoritatively in a way that earlier discussions on the subject failed to do. I will proceed with my project and not worry about carrier frequencies.

I've got my circuit working, the IR controller works well from a long way away from the receiver, at least with no housing around the receiver.

Eventually, I want to locate the receiver behind the front panel of an amplifier, so must find a way to expose it to the IR beam that is directed at the amplifier, and that means drilling a hole of some kind in the panel and putting the IR receiver behind the hole. I got to wondering if a light pipe would work well in this situation. There are lots of them out there, but intended to direct the light from inside a box to be visible on the outside. Will it work as well to direct IR from outside the box into the box to the IR receiver?

Something like this is what I had in mind, does anyone have experience with doing it this way?

http://www.digikey.com/product-search/en/optoelectronics/optics-leds-light-pipes/524541?k=light%20pipe

http://www.ebay.co.uk/itm/5pcs-5mm-Bezel-LED-Clip-Holder-Mount-Chrome-Metal-Rubber-Base-Solid-Construction-/281398118157?pt=UK_BOI_Electrical_Components_Supplies_ET&hash=item4184a2870d Why not mount the IR LED on panel

I don't think i've seen a light pipe used in any gadgets that I've taken apart. Usually they have the sensor behind a tinted 'window'. My gut reaction tells me a light pipe would knock the sensitivity way down. Could you drill a hole and then use window film to cover the hole from the inside for a clean appearance on the outside?

JimPerry said:

http://www.ebay.co.uk/itm/5pcs-5mm-Bezel-LED-Clip-Holder-Mount-Chrome-Metal-Rubber-Base-Solid-Construction-/281398118157?pt=UK_BOI_Electrical_Components_Supplies_ET&hash=item4184a2870d Why not mount the IR LED on panel

Click to expand...

He's wanting to mount the receiver, not the LED.

Most products such as TVs, would simply place the sensor behind a red plastic panel.

You can buy the correct shade plastic as add-ons for small hand held enclosures e.g.
http://uk.farnell.com/hammond/001150/infrared-front-panel/dp/1244229

I think I found my solution this morning.

My goal is to install a small window for the receiver in the aluminum front panel of my stereo preamp. I need something about 1/2" in diameter that I can cut to be perfectly flush with the front of the panel. I think I've found my solution at Amazon. I will ask a friend who is a machinist to cut a hat-shaped piece from this with a 1/2" diameter front with height that exactly matches the panel, and leave a raised edge at the back that I can glue to the back of the panel. The front should then be quite elegant-looking.