|
It is currently 19 Apr 2024, 14:06
|
View unsolved topics | View unanswered posts
| Author |
Message |
Jump to forum Jump to topic |
| vt4000 |
|
Posted: 20 Mar 2016, 06:35
|
|
|
So, for anyone that's interested how this turned out, I was able to dig up the service manual for this player, and located the analog video signal that happycube was describing (pin 22 of IC500 digital video processor, helpfully labeled "V2" in a box on the board).
I decided to go for broke and soldered a wire (30awg insulated) to the pad for this pin, and connected it to a newly added panel mount RCA jack on the back of the unit.
And what do you know, IT WORKED! Comb filtering artifacts are greatly reduced if not totally eliminated when connecting directly from this new jack to the TV. All features of the player seem to be retained (digital memory, etc) except for the "picture enhancement" which was completely useless anyway.
I hope this ends up being useful to someone else with a CLD-D606 or CLD-D604 (I presume it would work on this player also, as it's functionally the same as the 606 except with no karaoke).
http://i.imgur.com/gSSqw0Z.jpg http://i.imgur.com/ZJUVQ9w.jpg |
|
 |
Jump to forum Jump to topic |
| singmai |
|
Posted: 26 Jan 2021, 09:38
|
|
|
Hello,
Just to clarify about the output from the SM03. The output is SDI (serial digital interface) which is commonly used in broadcast environments. It is a component output, uncompressed, but consumer equipment does not accept it. However, it is possible to convert SDI to HDMI using a low cost box from companies such as Blackmagic. (We do not provide HDMI output because the companies that 'control' the HDMI standard require an annual payment from us so we can purchase the HDMI interface ICs).
The output is 525i (NTSC or 480i active picture lines) or 625i (PAL or 576i active picture lines). We do not provide deinterlacing because modern day displays now have quite sophisticated deinterlacers (as opposed to older monitors that just did line doubling or bob and weave for example) and some have motion compensated deinterlacers. What the SM03 does do is try to produce as artifact free image as possible which makes the job of the deinterlacer simpler (and therefore produces better image quality).
What I am unsure about is how the deinterlacers handle cadence and it might be possible for us to 'help' in that respect.
The SM03 was designed to work with VHS tape players. The method to record the video (colour under) means that a comb filter cannot be used to recover better video. Or at least a conventional comb filter cannot - it is possible to design one that will recover better video. I am less familiar with Laserdisc but we could look to do something similar if the recording format does something similar. For the VHS project we are also replacing all the electronics and are processing the head signal directly to produce better image quality. Again we could consider that in the future. One of the incentives to doing this is to produce faster than real time archiving of the tapes. At present a 3 hour tape takes 3 hours to archive - faster playback corrupts the image. We are hoping to enable faster than real time playback without corruption - essential when you have tens of thousands of tapes to archive.
So I have to read up about laserdiscs which I will do over the coming weeks, and take as much feedback as you wish to give me on what your requirements are for a laserdisc 'processor' to produce the best image quality at an affordable price. |
|
|
Jump to forum Jump to topic |
| whatchitfoool |
|
Posted: 04 Mar 2021, 03:00
|
|
|
| No love during the pandemic? |
|
|
Jump to forum Jump to topic |
| pbiancardi |
|
Posted: 26 Apr 2021, 00:23
|
|
|
I finally made something I find useable and presentable. I gutted an old Faroudja processor and put the eval board, Raspberry Pi (to auto run script), HDCP stripper (my HTPC has an issue with it), and power supply all inside.
Everything works with a front mounted switch and I also wired in the front "on" LED light and replaced the stock (loud) fan with a silent Noctua.
This is all in one solution, front switch powers on eval and Raspberry Pi which runs the eval board script with no user interaction required.
Eval 480i -> Crystalio II 480p -> HTPC / MadVR 2160p -> Projector
https://i.postimg.cc/TPhcKxnw/0425211502.jpg
https://i.postimg.cc/Z558Qm7c/0424210637.jpg
https://i.postimg.cc/wj4X8rnd/0424211355a.jpg
https://i.postimg.cc/NMmmXjFg/0424211355b.jpg
https://i.postimg.cc/Jh9jtk16/0424211355f.jpg
https://i.postimg.cc/HLJQp2C8/0424211358.jpg
https://i.postimg.cc/VvPT9vF4/0424211408.jpg |
|
|
Jump to forum Jump to topic |
| chrisf16 |
|
Posted: 18 Sep 2021, 16:45
|
|
|
EDITED: I had left out the all important step #5!
I have successfully pulled this AC-3 daughter board and fired it up as a standalone RF Demodulator!
Let me preface this with the fact that my old VSX-07TX was kind of toast and was putting out a loud 60hz hum from the amp and pre-outs. I'm not saying the audio had a 60hz hum on it which is a common problem. I mean it had a LOUD 60hz hum INSTEAD of the program audio. On all of the pre-outs. If I banged the unit on the counter a few times while turning it off and on I could sometimes get a clean output and it made a fine AC-3 decoder but it was in its death rattle. The LEDs always faithfully showed it was demodulating and decoding AC-3 RF into 5.1 so I was confident the daughter board was ok.
I considered just adding a Toslink AC-3 output to the unit and let it dangle out the bottom. I was able to find a service manual for the VSX-07TX (and other models) which had full schematics of the AC-3 board but no diagram of the physical layout so I wasn't going to be able to identify any of the soldering points I needed by just looking at the belly of the board. This meant I would have to pull the board out to figure out how to connect everything. If you think you'd like to preserve your receiver and take this approach, I will propose a simple way to do that at the end. Maybe somebody can try what I suggest and see if they are successful at just adding a demod out jack without removing the board. I believe it would be easy to do.
I fully sacrificed my receiver to harvest the board. Here is a description of what I did.
1) First let me say getting TO the board is easy. It has its own cover just under the belly of the unit. Getting the board OUT is another story. This is a ONE WAY TRIP. You will most assuredly render the receiver unusable doing this. There are 3 ribbon connectors from the main unit to the board and they do not have plugs on them. The smaller one at the back end of the board has bare wire ends that are pushed into locking sockets which you have to just rip out. The other two along the side have clip-down "backstop" caps just holding the bare wire ends into the socket. You will want to preserve this backstop on CN9801, the one along the side, closest to the back of the board (the middle of the 3 ribbon sockets if you want to think of it like that). You will have to release the tips of the plastic mounting posts that hold the board to the unit, then unscrew the screws on the back plate of the unit that secure the coax RF and digital input jacks. You need to unscrew the next nearest small backplate support screws like those nearby so you can bend the back plate away far enough to clear the RCA jacks so that you can finally separate the board from the receiver. Now off to the electronics dump with the main unit unless you want to harvest any of its other organs...
2) Now you need to build an S/PDIF output. I chose Toslink because 1) it is a super simple circuit with very few parts needed and 2) I am making NO electrical connection to my modern receiver from my "Frankenstein" board. Don't trust myself to not send some major zapping death pulse to the new receiver... If your decoding receiver only has coax S/PDIF input, then you will need to build the 75 ohm coax type output. There are many examples on the web. I did a small PCB with the Toslink transmitter, one .1uF capacitor bridging the +5 and GND wire like so:
https://i.imgur.com/InSCFpr.jpg
https://i.imgur.com/nLoCcyV.jpg
Black = GND; White = +5Vcc; Grey = Vin (S/PDIF signal) Ignore the double wiring, I changed that later. Just +5 and GND DIRECTLY from power supply.
3) Next you must solder the Vin signal wire of the Toslink out to the board. The point is on the path right before pin 9 into the CD0004AF chip. This chip takes the 4 inputs and does the switching (using receiver's "signal select" button; the "auto" mode is smart switching: AC-3 RF if present, if not pass the other coax or optical) into the DSP for decoding. As a future project, I may try to figure out how to control that chip and pickup my signal on the output side (emulating the Pioneer RFD-1). But for now, I am just picking up the demodulated AC-3 RF and going straight out to the transmitter. !!!!UPDATE!!!! My first attempt at soldering the Vin for the Toslink out was downstream near the input switcher chip, but it gave me trouble so I looked farther "upstream" and found a surface solder point labeled "DEM" (you can see this label under my wire end) right next to the output pin 70 of the PM4007A Demodulator chip. This makes a much more solid solder point and gives me a steady signal, but again, this is only accessible if you have fully removed the daughterboard. Here is a closeup:
https://i.imgur.com/21rbYMz.jpg
This backside solder spot would be the choice for adding demod out without removing the board from the unit (see proposed add-on mod below). This is a very tiny through hole and it will be difficult to solder to solidly.
https://i.imgur.com/kQlbCAh.jpg
If soldering to the "bottom side", you may have a better chance soldering to my original spot:
https://i.imgur.com/opMtLKn.jpg
4) Now to provide power to the orphaned AC-3 board. This circuit only needs +/-5V DC, +15V DC, -15V DC and their common to function. All the other voltages are for the DSP and A/D converters that we are bypassing now. All of these power connections are on CN9801 (that "middle" ribbon connector socket that you hopefully salvaged the backstop crimp holder for). Bring these 5 wires DIRECTLY from the power supply. I will use the pin numbering from the schematic in the service manual. They are numbered in the picture below from right to left. The GND for the 5V goes on pin 9, the +5V supply goes on pin 8, the -15V goes on pin 5, the +15V goes on pin 4 and the common for the 15s goes on pin 3. That backstop cap can now be jammed back down to hold your wires in place.
https://i.imgur.com/WiUv1QD.jpg
5) The demodulator chip PM4007A needs to be told to turn on. This last step was figured out by comparing the schematic of the VSX receiver to the schematic of the Pioneer RFD-1. The circuits are nearly identical up through the demodulator chip. The RFD-1 just keeps it turned on by a steady +5V on pins 3 & 4, but apparently pins 3 & 4 are signaled by the DSP decoder chip in the VSX receivers. I guess if the AC-3 input isn't selected, it turns off the demodulator. Luckily pin 2 right next to them has the Vcc +5V so I just soldered 2, 3 & 4 together.
https://i.imgur.com/i4N4Oqb.jpg
Now the board should operate as a dedicated AC-3 RF demodulator to Toslink out. Make sure you split the +5V and GND(5v) for the board and the Toslink transmitter AT the power supply. As you see in my first picture I was splitting at the Toslink and had ground path problems. Once I split at the power supply it consistently fired up and worked every time. Here is my (interim) completed project:
https://i.imgur.com/M1qmzPP.jpg
I have ordered a 3 voltage AC/DC adapter power supply that I will wire into it and an enclosure to house it all.
PROPOSED DEMODULATOR OUTPUT MOD TO OPERATING RECEIVER:
In other words, if you don't want to destroy the host receiver, I believe you could just add the demodulated S/PDIF out.
DISCLAIMER/WARNING!!
!!!! I haven't tried this. I do not guarantee that this will work as proposed and you may still end up deceasing your receiver !!!!
1) Of course, unplug everything and remove the AC-3 daughterboard cover from the receiver
2) Build the Toslink transmitter module as in step 2 above
3) Solder the Vin wire from the Toslink out module onto the pickup point described in step 3 above. Here is a broader picture to find the location. Refer to closeups in 3) above for more detail.
https://i.imgur.com/z6FF7TQ.jpg
The spot on the left is directly opposite the surface solder spot shown above in step 3. The spot on the right is where I tried the first time (note my first sloppy solder attempt) and is right before the switcher chip input. There is more blank board surface at this secondary spot so less chance of solder spilling onto a another trace and causing a short so you may prefer this location.
4) Solder the Toslink +5V (Vcc) onto the back side of socket CN9801 pin 4 and the GND wire on the backside of pin 3.
https://i.imgur.com/e15pC0k.jpg
Don't be confused by the plastic mounting pegs in my pictures. You will see them coming into the board from the other side while yours is still mounted to the receiver. I reclaimed them from the receiver and used them on this side of my board as "feet" to hold it up in whatever project box I end up enclosing it with.
5) Place the cover back over the board while routing the 3 wires of the Toslink transmitter out to wherever you can find to affix it.
I suspect this solution will require the VSX receiver to be 1) powered on, 2) LD/SAT input selected and the signal select forced to "AC-3 RF" (not auto). Just guessing.
Advanced research opportunity for the real electronics experts in this community: Find out how to control the source selector chip CD0004AF to do the smart pass through function and then pickup Vin on the DATA pin (24) of the switcher chip. My guess is some combination of low or high voltage on pins 4 & 5 which seem to be "selector" pins. There are 4 inputs so 2 binary value pins would seem to cover it.
Good luck and please let me know if anybody else makes this work or finds the advanced solution. |
|
|
|
Page 1 of 1 [ Search found 5 matches ]
|
Thanks list