|
|
Post by robin36mac on Feb 1, 2019 3:21:36 GMT -5
Hello All ! We do have a project, with a friend, that may see the day or not... That will gave a second life to my 808. This time, it won't be HT use but simulation use. But before doing so, and as it will the day later in the year, I still have time to try upgrading them, as it will be profitable for their future use. I do have a BG808, 22K hours chassis, almost nothing from the original serial number... I also have too a BD808S/E, which is in mint condition, 2600hours with P16LNQ. I make that post as I have some questions, I surely will have others, and also to share that process. So I go for the BG808, as it has a higher H frequency, and let's obtain the best from it. I first planned doing 3D, so I put 120MHz video amplifier : (it's in progress, there, sure the three are done)  Then I've been able to get a fast phosphor green tube, and that's the result, you can see before (standart tube) and after (fast green) how it compares with the red...   (to be continued in next post, to add more attachment as I prefer doing pictured posts) |
|
|
|
Post by robin36mac on Feb 1, 2019 3:49:08 GMT -5
As it was a Marquee tube, I had to mod and link pin 2 & 6 together for G2, as they have different pinout. Then it is compatible with both types.  Ok that works well there  Then having the left side correction would be very nice to have (and the why of my post on the EPROM thread). I have finally found a R762518 that has all the chips implemented : (and I'll get Seagull too)  and by the way, I also got the convergence on green card !  Attachments:
|
|
|
|
Post by robin36mac on Feb 1, 2019 4:13:47 GMT -5
Ok, then it's time to upgrade the firmware to get the settings, and you shared the last 808S V7.61, but difficult to find binary of the 6.30 or so for the stock 808. Then... why not going to upgrade as a 808S ? That mean I'll have to swap the input board, and deflection board are not the very same. I have the R762720, and then I figured out that I have some R7627195, as they are in common with the 909 (and enough of them), which is supposed compatible according the "5-7-9-Hulio-rule" Looking at schematics, difference between the 719 and 7195, is they went for using higher bandwith Op Amps in the video matrix switcher instead discrete transistor, and minor change in menu highlighting stuff. Then I swap the input board, keeping the original defl board as I don't have different ones anyway. I don't have the corresponding quad decoder, but don't need it anyway also. It will be fine without it. Just for curiosity, first with the 6.10 Firmware. The set start-up, picture is there, but no brightness/contrast command. Not very surprising, that's different. Then I put 7.61 firmware. And that works very well  I do well have the left side correction, Seagull, convergence on green... and working. I also have an IRIS2/3 kit, so now it is a 7.61, it will work with it, unlike the with 6.10 (which needs IRIS 800). The other BD808S/E have one mounted, so I did the same holes in the convergence tray  And "voilà ".  And it works |
|
|
|
Post by robin36mac on Feb 1, 2019 4:51:20 GMT -5
And then... we got a different idea and changed our mind. Instead of doing 3D, why not use both 808 and blend them ? That will give great and better immersion. And let's come to the harder part. I got a SEMUII kit too, coming from some BG1209S (previous pieces comes from them also). Nicely, the vendor put also BCL and Contrast modulation options in the package, I'll back to them later. (Unlike the first SEMU, designed for stock 808, the SEMUII connect the the input board R762720 and don't need modulation contrast ; first SEMU needed modulation contrast which came in port 3 slot initially) Good also, he put the input board R762720 coming with them. So it's easy to swap as there are swap depending if SEMU board is there or not. Not all 720 board have both Soft Edge connectors.  It's easy to find information on the net about removing them from a PJ, but installing them, really not much ! Anyway, let's go. In an original 808S I would not have any problem (SEMUII being design for them), but as it "was" a plain 808, things are different. There is a J20 connector which is implemented or not on the different controllers. And that's a needed connector for the SEMU II. Last revision I found show it as this in schema : Two logic gates, and then it goes to the ASIC.  All are R7621115, but different index (label is not showing it ?).  - Upper right is the BD808S controller, both logic and J20 are implemented - Upper left is a BG1209S controller, J20 and one logic gate and the second one on a sub-PCB (seems it didn't worked at first time !) - Both lower are controller from the BG808 and another one, there is no J20 |
|
|
|
Post by robin36mac on Feb 1, 2019 5:20:42 GMT -5
So I then tried to put the BG1209S controller board instead, using the 808(S) EPROMs on it. It starts ! It works ! Almost in fact.  SEMU is well there, and the test pattern is "soft-edged" here :  Then I try higher resolution/frequencies to see if everything follow well (we plan to use the 808 not far of its max), and then issue starts. Trying 15/32/47KHz patterns, everything is fine. But when I go up in 60KHz (or more), the menu in not synced anymore (being a test pattern or the menu itself is the same).  The input video signal is well synced however. I removed the options, and redid some swap, and come to the point that is the 1209 controller board that is the issue. With the stock controller board, the menu and paterns works well even near maximum, or at 1600x1200@85Hz input signal. Putting back 1209 controller then issue is back. 1209 deflection being a little different with two retrace time, I thought it could be linked the the HFM line, which is floated in the 808, and connected to ground, or +9V, that's not changing anything. With the controller from the other 808S, the set just don't even boot and scan fail (still with same EPROM, sure). As the controller have very little difference, only thing different I con think of is the EEPROM IC6, I would think it only store counters and block setting, maybe there is some other info ? I'm not getting what is the difference, if you have any experience to share ? I'm also planning doing a sub-PCB to add the J20 on the stock controler, let's see how it goes. So here I am. I'll tweak the other 808 after. |
|
|
|
Post by robin36mac on Feb 3, 2019 11:38:34 GMT -5
So, the last findings :
Somethings goes wrong near the PLL MC14046, IC15.
For a internal pattern at 61KHz :
- Pin 14 : 61KHz signal (which is signal input)
- Pin 3 : 58KHz signal, which explain the menu is not synced...
- Pin 13 : No pulses ! Vcc-like signal
- Pin 2 (unconnected) : Messed pulses which is normal behavior considering the sync issue
Next step, changing that IC...
If there anything to be checked, let me know !
Robin
|
|
|
|
Post by km987654 on Feb 4, 2019 3:33:34 GMT -5
The SEMU board you have is very noisy. The SEMU board from the 909 works well with no noise in the picture. You will also need the contrast modulation board to make this work and the one that is designed for the 808s also creates much picture noise but again the contrast modulation from the 909 works and doesn't introduce noise. The contrast Modulation board will cut image contrast by a good margin making the image dull.
If you intend to blend the best solution may well be an external blend box like a Diventix by Analog Way.
|
|
|
|
Post by robin36mac on Feb 4, 2019 12:31:01 GMT -5
My SEMU boards are R762948, and the SM of the BR909 talks about the R7629482 for SEMU II, it's this one you talks about ? and R7629483 being the SEMU III (only for 909)
About the modulation contrast I have one R7624845 and one R763406, but this last one I don't think will run on the 808S ; I just hope that one modulation contrast only will be enough just to make them match enough.
Sure there are better way to do this, but I'll do also with what I have, or cheap enough.
There is also some curiosity to see how SEMU behave ;-)
I don't expect perfect HT-picture for this project, juste something enjoyable enough and fun.
But I'll try to check the effects on the video path, if it is that bad or not, anyway I'll have plenty of time before the Barco will be monted on the spot.
Robin
|
|
|
|
Post by km987654 on Feb 5, 2019 3:01:04 GMT -5
Attached are the two sets of boards. The one with the fan attached is from the 808. These boards create a lot of picture noise. The other image is of the 909 boards. Much better to use.
909 board numbers (clean)
SEMU R7629482
Contrast Modulation R763406-04 808 board numbers (noisy) SEMU R762948 Contrast Modulation R7624845 Attachments:
|
|
|
|
Post by robin36mac on Feb 11, 2019 14:37:20 GMT -5
Hello, Okay, so they are indeed different board and different version of the SEMU II. That's clear so. Have you tried to get the 909 version on a 808S ? I've tried to add the J20 with 74AC32 and 74AC74, with the stock controller, but no luck so far. The PJ start up, show the hourglass (as when SEMU is loading or redoing the edge calculation), then black picture. Whatever the remote button I push on, It shows the hourglass for some seconds and then black. So no menu, no way to switch to an input. (Except standby button which do its job.) That's a fail !  Using RWI v5.09 , that's not better. To be continued... Robin |
|
|
|
Post by robin36mac on Feb 14, 2019 14:14:58 GMT -5
Going back to the controller coming from 1209S, changing PLL 4046 IC15, still same issue. All fine at low resolutions, then internal/genlocked controller-generated menu or test patterns are messy.  Robin |
|
|
|
Post by km987654 on Feb 15, 2019 2:47:21 GMT -5
Hello, Okay, so they are indeed different board and different version of the SEMU II. That's clear so. Have you tried to get the 909 version on a 808S ? I've tried to add the J20 with 74AC32 and 74AC74, with the stock controller, but no luck so far. The PJ start up, show the hourglass (as when SEMU is loading or redoing the edge calculation), then black picture. Whatever the remote button I push on, It shows the hourglass for some seconds and then black. So no menu, no way to switch to an input. (Except standby button which do its job.) That's a fail ! Using RWI v5.09 , that's not better. To be continued... Robin Yes I have added both the 909 SEMU and CM boards to a 808s Controller and they work just fine. You need firmware v 7.61 and RWI 5.09 and a controller that already has J20. |
|
|
|
Post by km987654 on Feb 21, 2019 2:59:44 GMT -5
The SEMU board you have is very noisy. The SEMU board from the 909 works well with no noise in the picture. You will also need the contrast modulation board to make this work and the one that is designed for the 808s also creates much picture noise but again the contrast modulation from the 909 works and doesn't introduce noise. The contrast Modulation board will cut image contrast by a good margin making the image dull. If you intend to blend the best solution may well be an external blend box like a Diventix by Analog Way. Just wanted to add some information to my above comment about Contrast Modulation Board and how it reduces contrast by a good margin. Its been a while since I have used the CM board and after making the above comment I decided to take another look at it and its impact on the image. While there is a change in the contrast level I would say its not big as I first mentioned and I think this is because when I first installed the CM board the other boards contained 20 year old capacitors, I have over time replaced a significant number of capacitors right across the projector and this makes the projector different to when I originally tested the CM board. |
|
|
|
Post by robin36mac on Feb 27, 2019 3:03:02 GMT -5
Hello All,
Then if someone has some of these SEMU-II 909-style, I'm open to offers... But I guess they will be quite difficult to find.
I hadn't much time yet, troubleshooting of the controller is still ongoing :
Comparing this messy controller to the working one on the BD808S, VCO output is 35MHz on the messy controller and 36MHz on the fine one, with a 44,2KHz pattern.
The VCO_0/1/2 signals from the logic part are well the same (which select the VCO range)
Output from low-pass filter LF353 is around -8V (so should be at its max frequency ?)
To be continued...
Robin
|
|
|
|
Post by hulio on Feb 27, 2019 7:01:20 GMT -5
Hello Robin,
Nice project. I still have a spare SEMU II- Advanced (R7629482) for you.
Send me a e-mail with 909 stuff you can miss and think is worth trading. If i'm not wrong, you also need the CM board in order for SEMU to work (i have it as spare too). Do you already have the special connector in the convergence tray for the SEMU ?
|
|
