Jump to content
Due to a large amount of spamers, accounts will now have to be approved by the Admins so please be patient. ×
  • 0
IGNORED

CRT TV EEPROM & Micom hacking for RGB (+tube swap for 48cm to 51cm TV)


buttersoft

Question

This thread started out being about swapping a 48cm tube onto a chassis that originally had a 51cm tube, but has turned into a discussion of EEPROM and microcontroller hacking to enable RGB on old Australian (possibly NZ) SD CRT TV's on which it appears to have been disabled by the microcontroller (Micom). Discussion of the latter begins a few pages in, and it's all very much a work in progress.

 

Note that not all old sets need the above, in progress hack in order for you to RGB mod them. As a very rough guide, if the neckboard has colour pots for RGB bias/gain, you probably don't need to worry, injecting signals into the jungle IC should just work. For newer sets without neckboard pots, try inserting RGB, and see if it works.

 

Remember that the shumps thread is a better place for information on the subject of RGB hacking!

 

I've been hunting for more RGB-moddable chassis to drive 20"/48cm TV tubes - the standard 2ohm/14.8ohm ones in old cabs. They're getting harder to find though.

 

After @namastepat's just-make-it-fit suggestion I was wondering about swapping out the chassis from a 51cm TV and using it on a standard 48cm tube. Do the 51cm tubes typically have the same yoke resistances and pinout? Is the required HV going to be much different? I know I've seen chassis that will suit multiple tubes.

 

At the moment I'm looking at this job here - https://my.mixtape.moe/rnatkf.pdf.

 

That service manual says the chassis can handle 14", 20" and 21" tubes. (In this particular case I suspect the 20" and 21" tubes might be the same thing but US vs Aus names? i.e both the same 51cm tube, not a 48cm one?)

Edited by buttersoft
Link to comment
Share on other sites

  • Answers 299
  • Created
  • Last Reply

Top Posters For This Question

Top Posters For This Question

Posted Images

Recommended Posts

  • 0
My eyes are killing me from measuring and experimenting with more resistors than I can possibly count, but I am getting close to the magical sync that makes the LG happy (for RGB to component that is)...

 

I prefer to use pots at about 1K, that way you can cover a bigger range easily. Then just meter the pots and replace with resistors.

Link to comment
Share on other sites

  • 0
I prefer to use pots at about 1K, that way you can cover a bigger range easily. Then just meter the pots and replace with resistors.

 

I started off that way (500 rather than 1k), but turned out it was wanting a hell of a lot more resistance than I expected... narrowing down was difficult though as too low it went whacko, too high went a different whacko, so took much jumping around to eventually find the sweet spot is somewhere around the 5.8k-ish mark... Not what I expected at all. Now that I'm in the right territory I'll go back to pots to tune it from there. Might be a few nights before I can get back to it.

Link to comment
Share on other sites

  • 0
I started off that way (500 rather than 1k), but turned out it was wanting a hell of a lot more resistance than I expected... narrowing down was difficult though as too low it went whacko, too high went a different whacko, so took much jumping around to eventually find the sweet spot is somewhere around the 5.8k-ish mark... Not what I expected at all. Now that I'm in the right territory I'll go back to pots to tune it from there. Might be a few nights before I can get back to it.

 

Sounds like hard work. Glad you're making progress.

 

 

Sent from my iPhone using Aussie Arcade

Link to comment
Share on other sites

  • 0

Looks like I won't get the chance to play with it until I return from holidays but psyched to see if I can do this. I'd consider it a pretty good achievement to communicate with the Jungle I/C and tell it to do something the factory didn't want it to do.

 

No certainty yet though.

 

 

Sent from my iPhone using Aussie Arcade

Link to comment
Share on other sites

  • 0

CRT TV EPPROM & Micom hacking for RGB (+tube swap for 48cm to 51cm TV)

 

Tag me when you start the process up again ;)

 

I'll be sure to. Likely three weeks from now.

 

The plan is to try and hook up my PC to the TV's I2C lines via the existing serial port, try and read the registers from the TDA8842 Jungle I/C. If I can get that to work then I'll try and send the command to change the Source Select as a test case, then move on to RBL and IE1 flags (depending on what they read as).

 

On a side note I think they call the "Enable Fast Blanking RGB-1" flag "IE1" because it stands for "Interrupt Enable 1".

 

If I can get my PC to change the Jungle I/C registers and that successfully enables Scart RGB to work I will be looking to program a micro controller that can be embedded in the TV and will set the Jungle chip registers a short time after TV start up. I'd like to do this in the cheapest possible way. Programming an Arduino is really quite straightforward, there is a "Wire" library designed to work with I2C.

 

Who knows what road blocks I'll encounter on the way but have to have a plan.

Edited by MarkOZLAD
Link to comment
Share on other sites

  • 0
I was looking up USB-to-I2C, trying to find something that can act as master or slave, or even better just sit in between reporting traffic. While looking at the datasheets for the USB-I2c dongle you linked, and for the IC it uses, I'm reading that the device can only operate as a slave? Does that mean it can't send instructions to another slave like the jungle IC, or am I reading something wrong? So you'd read what the EEPROM is sending using the dongle, and then use an Arduino to send the info you want to the jungle chip, is that the plan? I presume you can program a particular slave address into the dongle?
Link to comment
Share on other sites

  • 0

Damn, Slave Only? I've stuffed up there then.

 

I need something that is the master.

 

[EDIT]

 

Yes, you're correct as usual. I need a USB breakout board with the FT232R chip instead.

 

Getting very confusing to find the right breakout board to do I2C. I'll get it though.

 

If I can't find a cheaper local alternative I'll just get this:

 

http://www.robot-electronics.co.uk/products/communication-adapters/interface-modules/usb-to-i2c-interface-module.html

Edited by MarkOZLAD
Link to comment
Share on other sites

  • 0

If you google the datasheet of the IC listed in the product datasheet, the FT201XQ, it says slave only :(

 

Cheapest master I can find is here - or anyway it says it's a master, and win7/8 x32/x64 compatible. But it's aliexpress, so it'll take a month at least to get to you.

 

I'd agree that trying to read the registers of the jungle IC would be a shitload easier than interpreting traffic between it and the Micom. That way you'd have a map of the set as it's functioning normally, to feed back in. How were you planning to do this, by wiring the extra master unit into the existing i2c bus as is and simply addressing the jungle IC as slave? You won't need to disconnect the micom lines first, from what I can see?

 

And then what? You just put the read request to the jungle IC and see what it spits back? And that's where we'd expect the data to match the jungle IC datasheet map 1:1, yes?

 

Sorry for the interrogation, but if I'm starting to understand the process I'd really like to give it a go myself :)

Link to comment
Share on other sites

  • 0
If you google the datasheet of the IC listed in the product datasheet, the FT201XQ, it says slave only :(

 

Yep, definitely a slave device. I hadn't done my due diligence, thanks for finding that out.

 

Cheapest master I can find is here - or anyway it says it's a master, and win7/8 x32/x64 compatible. But it's aliexpress, so it'll take a month at least to get to you.

 

I'll check it out. Will be a while before I get home....

 

I'd agree that trying to read the registers of the jungle IC would be a shitload easier than interpreting traffic between it and the Micom. That way you'd have a map of the set as it's functioning normally, to feed back in.

 

Exactly what I was thinking, read the registers and see if I can make sense, see if it matches the data sheet...

 

How were you planning to do this, by wiring the extra master unit into the existing i2c bus as is and simply addressing the jungle IC as slave? You won't need to disconnect the micom lines first, from what I can see?

 

The LG has a service port with easy access to SDA1/2 and SCL 1/2 and ground. I was going to connect wires there.

 

The set has an I2C serial bus so all devices can be connected and only the device that you address in your message will respond. Should be no need to disconnect anything. The service port suggests to me it is designed to allow another master device to be attached to the board to do exactly this.

 

 

And then what? You just put the read request to the jungle IC and see what it spits back? And that's where we'd expect the data to match the jungle IC datasheet map 1:1, yes?

 

Precisely. Send read commands to get all of the data from the Jungle I/C, work out what's what in the Jungle I/C registers (hopefully it matches datasheet), then send write commands to the Jungle I/C to get it to change it's registers and hopefully do our bidding.

 

Sorry for the interrogation, but if I'm starting to understand the process I'd really like to give it a go myself :)

 

No problem at all, I would much prefer to work as a team on this. I can't do anything for a while, if you can progress this while I'm away I'd be stoked.

Link to comment
Share on other sites

  • 0
Hope it works! Total generic brand, no codes or datasheets or anything. Still, if they claim it works on Windows they must have software for it.

 

I'll give it a crack. Added a message to vendor to please provide links to software. $11 gamble.

 

I'm headed to Vegas in two weeks, thought I'd start early.

 

 

Sent from my iPhone using Aussie Arcade

Link to comment
Share on other sites

  • 0

CRT TV EPPROM & Micom hacking for RGB (+tube swap for 48cm to 51cm TV)

 

Just checking the schematics for the Teac CTM5122H. Haven't found a serial port yet but should be able to join the I2C bus from the empty teletext chip port or the Z86129 port.

 

 

Sent from my iPhone using Aussie Arcade

Edited by MarkOZLAD
Link to comment
Share on other sites

  • 0

I'm now getting confused by the addressing required. From the Jungle IC datasheet, what is the actual address of the top left bit of the register map, the INA bit? 0x07? 00x7? Or am I writing/reading the whole byte, so 0x00 then the byte values.

 

And once that's sorted, how are the input bits addressed differently from the output bits? 0x00 vs 1x00 for the starting address?

Link to comment
Share on other sites

  • 0

I believe you have to read or write a whole byte at a time.

 

I'll look into it to confirm.

 

As for the output bytes, that is a very good question indeed. I have noted on other datasheets I've read that there are usually two addresses, one after the other. I would hazard a guess that for the TDA8842 where the slave address is 8A, the output registers are available at 8B.

 

I have read about a technique where you try every possible I2C address looking for responses so that you can identify devices in the bus. I think that's called bit-banging. May have to give that a shot.

 

 

Sent from my iPhone using Aussie Arcade

Link to comment
Share on other sites

  • 0

I've been thinking about multi master. It may well be an issue. For the LG I think this might be solved by using the FS (Enable) - Factory Service line on the serial port.

LG Facilitate Service.png

 

I think that if 5V is applied to this line the on-board micro controller will give up it's master status and the external master we attach can take over.

 

This is just a theory of course.

Edited by MarkOZLAD
Link to comment
Share on other sites

  • 0

I2C documentation states that the bus is definitely multi-master, and that slave devices will treat requests from different masters serially and won't get confused? I presume this means you would need to actually take the Micom out of service, and feed the jungle IC your preferred values as the set powers up. Otherwise the Micom would keep overwriting your RGB-enabling data. But that may not be the case. Or you could just have your master write it back every second, or something. I'm going to go through the schematics I have and see if I can see any obvious with different methods re: AKB or other feedback/control loops required for safety reasons.

 

Did you actually try getting RGB up on the CT-M5122H? Were you going to do that first, or just dive in with the I2C? This might be after you've played around with the LG, of course.

 

In other news I finally figured out that the 0x I keep seeing is prefix indicating hex numbering. That makes things easier to understand.

Link to comment
Share on other sites

  • 0
I2C documentation states that the bus is definitely multi-master, and that slave devices will treat requests from different masters serially and won't get confused? I presume this means you would need to actually take the Micom out of service, and feed the jungle IC your preferred values as the set powers up.

 

I'm a little confused. Did you mean to say it's NOT multi master?

 

I propose to take the MICOM out of the loop by sending 5V down the Factory Service line. Not sure if that's right but worth a shot.

 

Otherwise the Micom would keep overwriting your RGB-enabling data. But that may not be the case. Or you could just have your master write it back every second, or something.

 

I'm presuming the MICOM only writes data to the Jungle on startup and then as needed. I wouldn't think it would be constantly writing the same information to the Jungle I/C. At startup the Micom reads it's EEPROM for the stored settings and then pumps that to the Jungle I/C. After that the MICOM only sends changes to the Jungle if the user changes something via the controls, otherwise it just loops doing nothing.

 

Again, just my theory.

 

Did you actually try getting RGB up on the CT-M5122H? Were you going to do that first, or just dive in with the I2C? This might be after you've played around with the LG, of course.

 

For the CT-M5122H all I have done is study schematics and opened it up to photograph the chassis. I have too many unfinished projects ongoing already! It's on the list...

 

In other news I finally figured out that the 0x I keep seeing is prefix indicating hex numbering. That makes things easier to understand.

 

Yes in programming code if i say

 

int i = 10;

 

the compiler will interpret this as a decimal value 10

 

If I do

 

int i = 0x10;

 

the compiler will interpret this as Hexadecimal 10, which is decimal 16.

 

- - - Updated - - -

 

Here's an example of what I was saying about the slave addresses being sequential for input and output. Not sure if it is for the TDA8842 but I have seen it for numerous others.

Slave address for I2C TB1238N.jpg

 

TDA8842 has very similar pattern

 

Slave address for I2C TDA8842.jpg

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    • No registered users viewing this page.

×
×
  • Create New...