just wondering if anyone has figured out what the zenith ir code format is? i'm still trying to "crack" the codes for my old zenith tv. i'd love it if i could figure out the code format so i could add the appropriate lines to the .ini file for prontoutil

if anybody's bored enough, here are a few examples:

zenith tv power toggle:
0000 0066 0000 0011 0014 0013 0014 00a7 0014 00cf 0014 0013 0014 00a7 0014 0013 0014 00a7 0014 00cf 0014 0013 0014 00a7 0014 00cf 0014 0013 0014 00a7 0014 00cf 0014 00cf 0014 0013 0014 1421

zenith tv channel up:
0000 0066 0000 0011 0014 0013 0014 00a7 0014 00cf 0014 0013 0014 00a7 0014 0013 0014 00a7 0014 00cf 0014 00cf 0014 0013 0014 00a7 0014 00cf 0014 0013 0014 00a7 0014 0013 0014 00a7 0014 1449

zenith tv channel down:
0000 0066 0000 0011 0014 0013 0014 00a7 0014 00cf 0014 0013 0014 00a7 0014 00cf 0014 0013 0014 00a7 0014 00cf 0014 0013 0014 00a7 0014 00cf 0014 0013 0014 00a7 0014 0013 0014 00a7 0014 144a

any help appreciated! :)

i've been fiddling with this again. my attempt at deciphering the code doesn't go very far, but here's what I've got:

the 0066 is the carrier frequency.
the 0011 (dec 17) is the number of burst pairs (including the leadout)
*apparently* 0014 0013 is the burst pair representing a "0", while 0014 00a7 represents a "1". but if this is true, then what the heck is that 0014 00cf that's all over the place? is there such a thing as a "trinary" code? it doesn't appear to be any kind of lead-in or -out since it appears all over the place (particular in the codes for the number keys, not posted here).

has *anyone* messed with this? I'd love to hear some suggestions.

thanks again for any help

I'll take a crack at it, but it would be helpful if you would post the number keys (at least 0,1,2) also.

Nevermind on those codes, I got them from the files section. If you or anyone else could post the same codes from another Zenith product (like a VCR?) it would be helpful.

Here are a few of the number codes:

number 0:
0000 0067 0000 0011 0014 0013 0014 00a5 0014 0013 0014 00a5 0014 00cd 0014 00cd 0014 0013 0014 00a5 0014 00cd 0014 0013 0014 00a5 0014 00cd 0014 0013 0014 00a5 0014 00cd 0014 0013 0014 13ef

number 1:
0000 0066 0000 0011 0014 0013 0014 00a7 0014 0013 0014 00a7 0014 00cf 0014 00cf 0014 0013 0014 00a7 0014 00cf 0014 0013 0014 00a7 0014 00cf 0014 0013 0014 00a7 0014 0013 0014 00a7 0014 1449

number 2:
0000 0066 0011 0000 0014 0014 0014 00a7 0014 0014 0014 00a7 0014 00cf 0014 00cf 0014 0014 0014 00a7 0014 00cf 0014 0014 0014 00a7 0014 0014 0014 00a7 0014 00cf 0014 00cf 0014 0014 0014 1d5d

thanks so much for the help!!!
jason

just wondering if anybody has had any luck cracking these.

I cracked these codes long ago... the 00a7 and the 00cf are *the same* code... I don't know why it is, but I ran out a listing of the Zenith codes using all 00a7 (and no 00cf), and the buttons worked as if it was the original remote... Just ignore such small differences in codes...

Although, calculating it out, 00a7 is 20% less than 00cf, which is twice as much difference as the 10% difference listed in the IR code deciphering guide, so there *could* be something to it...

Actually, that reminds me!!! I just remembered that I had figured this out before... Notice, if you will, that the 00a7 ALWAYS comes at the first of a string of repeat codes (namely, there is a 0014 0014 before 0014 00a7) and that there is *ALWAYS* a 0014 00a7 or 0014 00cf before a 0014 00cf. This is probably an error check built into the code... if the bit sent is a 0014 00cf, the unit knows that the previous one was the same state bit.

So, I guess my codes I ran were based on that criteria instead. Try it, it works! :)

My remote worked my 20 y.o. zenith console tv right out of the box using the philips default code set. I don't know the history there, but wondered whether philips owned Zenith at the time, or whether Magnavox may have manufactured the set for Zentith or what!

I hadn't looked at the Zenith code format before, but this thread piqued my interest. The good news is that I have figured it out, the bad news is that I doubt any code generators will be able to format it without having specific code in place for it as it's not like any other IR format that I've seen.

It's true that the "0014 00CE" and "0014 00A6" strings both translate to ONEs, and "0014 0014" translates to ZERO. There doesn't appear to be a device code as such in the code, so I'm guessing that the slight differences in the three formats shown (for cable, TV and VCR) tell the units which one the signal is suppoed to be for.

Anyway, the code contains a 5 digit function code and it's compliment in MSB (most significant bit) first format. Now here's where it gets messy. The code is broken up into fragments. In the following syntax, F1 means the first digit of the function code, F2 the second, etc. C1 is the first digit of the compliment, etc, and ONE means ONE. Here's the syntax for the function code portion of the signal:

ONE C1 F1 ONE C2 F2 ONE C3 F3 ONE C4 F4 ONE C5 F5

When a ONE is needed for the compliment, it is always in the "0014 00CE" format. When a ONE is needed for the function code, it is always in the "0014 00A6" format. The fixed ONE that follows the code will be in the "0014 00A6" format if the preceeding code used "0014 00CE", and vice versa. I'm guessing this is to bring the timings back into line.

Here is the binary for the numeric buttons of a Zenith cable, TV and VCR. I am representing "0014 0014" by "O", "0014 00A6" by "i" and "0014 00CE" by "I". Then following the code I have shown the real binary for the compliment followed by the function code:

Zen Cable:
0: I Oi I IO i IO i IO i IO = 01111 = 10000
1: I Oi I IO i IO i IO i Oi = 01110 = 10001
2: I Oi I IO i IO i Oi I IO = 01101 = 10010
3: I Oi I IO i IO i Oi I Oi = 01100 = 10011
4: I Oi I IO i Oi I IO i IO = 01011 = 10100
5: I Oi I IO i Oi I IO i Oi = 01010 = 10101
6: I Oi I IO i Oi I Oi I IO = 01001 = 10110
7: I Oi I IO i Oi I Oi I Oi = 01000 = 10111
8: I Oi I Oi I IO i IO i IO = 00111 = 11000
9: I Oi I Oi I IO i IO i Oi = 00110 = 11001

Zen TV:
0: O i Oi I IO i IO i IO i IO = 01111 = 10000
1: O i Oi I IO i IO i IO i Oi = 01110 = 10001
2: O i Oi I IO i IO i Oi I IO = 01101 = 10010
3: O i Oi I IO i IO i Oi I Oi = 01100 = 10011
4: O i Oi I IO i Oi I IO i IO = 01011 = 10100
5: O i Oi I IO i Oi I IO i Oi = 01010 = 10101
6: O i Oi I IO i Oi I Oi I IO = 01001 = 10110
7: O i Oi I IO i Oi I Oi I Oi = 01000 = 10111
8: O i Oi I Oi I IO i IO i IO = 00111 = 11000
9: O i Oi I Oi I IO i IO i Oi = 00110 = 11001

Zen VCR: 
0: I Oi I IO i IO i IO i IO i IO i IO  = 01111 = 10000
1: I Oi I IO i IO i IO i Oi I IO i IO  = 01110 = 10001
2: I Oi I IO i IO i Oi I IO i IO i IO  = 01101 = 10010
3: I Oi I IO i IO i Oi I Oi I IO i IO  = 01100 = 10011
4: I Oi I IO i Oi I IO i IO i IO i IO  = 01011 = 10100
5: I Oi I IO i Oi I IO i Oi I IO i IO  = 01010 = 10101
6: I Oi I IO i Oi I Oi I IO i IO i IO  = 01001 = 10110
7: I Oi I IO i Oi I Oi I Oi I IO i IO  = 01000 = 10111
8: I Oi I Oi I IO i IO i IO i IO i IO  = 00111 = 11000
9: I Oi I Oi I IO i IO i Oi I IO i IO  = 00110 = 11001

As you can see, the code preceeding and following the function code is slightly different in the three examples, I'm guessing that this is what the devices use to differentiate between the three formats, otherwise someone with a Zenith TV, cable box and VCR would have all three units working off of all three remotes, and that just wouldn't do! :)

I do wonder, based on the strange format of the signal, if occassionally, the code accidentally falls into another devices format, meaning when you press "Menu" on your Zenith VCR's remote, you stereo changes inputs, or something.

Later,
Rob.
http://www.hifi-remote.com/

Whew ... it's been so long ago that i even messed with figuring out IR codes manually that it would take me awhile to a) get back into the swing of things, and b) figure out how to use the info posted above with any of the prontoutil or code generation utilities ... if i remember, i think my original goal was to generate a bunch of possible IR codes, then try them all out on the tv to see if there was a discrete on/off.

if anybody has any ideas how to proceed, particularly considering the odd nature of the code, please reply!

thanx,
jj

I found the discrete on and off codes for my Zenith VCR, so I could probably figure out the codes for your TV if you are interested.

Regarding the various code generators out there, I doubt that they would be able to create this code without having special code added. Most of them simply create a device code of so many characters in this or that format, followed by a command code, etc. They wouldn't be able to put all those ONEs in the middle of the code like Zenith does without having Zenith specific code added.

Having said that, i brought this thread to the attention of Barry Gordon who wrote some of those generators and he is looking into adding code to handle it.

Rob.

I haven't had time yet to look at the Zenith files I downloaded when I first saw this thread. Going on just my understanding of what you have written here, it ought to be easy to write an IRP file that would let either my MakeHex program or Barry's similar program generate this protocol. I don't think any programming change should be required.

Maybe you don't know that the defined sequence for ONE or ZERO in an IRP file can be longer than a single pair.

If I understand correctly, you have identified pairs that you call "0", "I" and "i". You have determined that when a ONE is meant it sends 0iI and when a ZERO is meant it sends I0i. I think you could just define each of ZERO and ONE to be three pairs.

You also seem to indicate that Cable, TV, and VCR each have different prefix and suffix structures. An IRP file can represent any of the prefix or suffix structures, but it can't represent more than one at once. You probably would need to write three different IRP files to cover that.

I hope I can find the time to investigate those CCF files and form my own opinion about the protocol structure, and modify my CCF2EFC program to handle decoding this protocol. I don't see (yet) why any programming change should be required for encoding it.

If anyone ever comes up with a discrete power on/off for Zenith TV's, I would be forever in your debt. Discrete codes for switching from TV to video 1, video 2 and back to TV would earn super bonus points as well.

Rob, you rock! I'm sitting here, pulling what's left of my hair out of my head trying to crack the silly code, & you did it! Wow.

Anyway, the good news is, I have made an Excel spreadsheet that correctly decodes AND encodes Zenith VCR hex IR codes, using the rules & formats established by Rob. I haven't tested them yet, but am trying to incorporate a way to make them into a ccf. I'm working on it, but my wife's in the hospital, and the neighbors let a tree fall on my power line. (bad week!). I'll let you know when I finish, though.

1 Question for Rob: I found that you are correct, there is a 5 digit function code, but that only yeilds 31 (hex=00001 to 11111) possible codes. Make it 6 digits, and you get like 64 (I think: 011111-111111). How do we get to 255 codes?

-Viscouse

Viscouse,
Basically you don't get to 256 codes, the reason that you may see the 256 number thrown around is because most IR signals use an 8 bit command code which translates to 256 possible commands, if there are less bits in the command code, you will get less possible commands. With only 5 bits you will only be able to generate 32 functions (from 00000 to 11111), which is probably enough for older TV's and cable boxes. Further research has shown me that the VCR actually uses a 7 bit code, the last two bits (which are constant in the examples shown above) actually change for some of the other functions.

John's idea about combining the three codes to form the ONEs and ZEROs is excellent, but you would need to overwrite the final burst with a longer off time.

Here's the codes for the ONEs and ZEROs:

1=0014 00CE 0014 0014 0014 00A6
0=0014 0014 0014 00A6 0014 00CE

The cable and VCR formats use the following lead-in:
0014 00CE

The TV format uses the following lead-in:
0014 0014 0014 00A6

It looks like a final burst string of something like 1700 would be more than enough.



Now, it looks like the 5 most significant bits of the 7 bit VCR command code match up with the 5 MSBs of the cable and TV formats, with the final 2 bits of the VCR code being set to ONEs, for simlar functions. Look at the numeric examples shown above.

I was hoping to use this principal to generate the discrete power codes for the TV and VCR formats, based on the known discrete codes for the VCR format, but unfortunately, the discrete on and off codes are ones where the final two bits of the VCR code is something other than ONE ONE. Which leads me to believe that Zenith didn't think of using discrete on off codes until they came up with the VCR format, but the only way to be sure would be to generate the TV hex and try it out.

Rob.