|
As this forum is rarely used anymore, we've locked it. Feel free to browse and read. Questions? Please reach out to us directly. Cheers! |
|
 |
|
|
| Author |
Message |
jask
Joined: 17 Mar 2006
Posts: 10187
Location: kamloops BC
|
| Posted: Thu Jan 31, 2013 5:19 am Post subject: |
|
|
| gjaky wrote: | Ok, the work has been done. The board and the concept works in general, I made to fit the CXA1779 in place of the CXA1209, without the need of recalibrating the whole machine. However, I'd lie if would say the new board looks better, but it's not worse either (at least), they do look exactly the same to me.
Also as usually an unforseen side effect occurred in the new board: somehow it doesn't like if any portion of the horizontal back porch is blanked out, otherwise it will streak badly and shifts the color balance, which I may try to eliminate later or may not.
Of course these are only first impressions, I'll throw the board in normal use, and further testing will come. |
Can you describe the streaking more specifically? or post images?
|
|
| Back to top |
|
 |
gjaky
Joined: 05 Jun 2010
Posts: 2802
Location: Budapest, Hungary
|
| Posted: Thu Jan 31, 2013 7:17 am Post subject: |
|
|
| Quote: | | What happens if you use position shift? My 9PG doesnt like too much of that being used. |
I've already tried that, and has no effect on streaking, the problem is clearly connected somehow to the extra blanking circuit that I've added for adjusting the appropriate blanking level during horizontal retrace. But frankly I don't feel this would have much importance at the moment, because there isn't any performance gain with the new board. However I don't give up my original goal, so instead of dicking with that board I look for other opportunities to find the bottleneck of the signal chain.
| Quote: | | Can you describe the streaking more specifically? or post images? |
Well that's tricky, a video would be the best to explain, it is very interesting, I'll try to catch it.
My pj is table top mounted, and the streaking comes up when light objects are near to the right edge of the raster. Also adjusting the blanking causes to light up the raster (as it should) but gives also a bad red tint to everything (the red raster becomes visible)
_________________
projectors in the past : NEC 6-9PG xtra, Electrohome Marquee 6-7500, NEC XG 1351 LC ( with super modified Electrohome VNB neckboard !!!)
current: VDC Marquee 9500LC
The MOD: VNB-DB, VIM-DB
|
|
| Back to top |
|
|
gjaky
Joined: 05 Jun 2010
Posts: 2802
Location: Budapest, Hungary
|
|
| Back to top |
|
|
CasetheCorvetteman
Joined: 09 Nov 2008
Posts: 6326
Location: Australia
|
| Posted: Wed Feb 20, 2013 4:01 am Post subject: |
|
|
Sounds good  Although i have no XTRA or XG 0
|
|
| Back to top |
|
|
gjaky
Joined: 05 Jun 2010
Posts: 2802
Location: Budapest, Hungary
|
| Posted: Wed Apr 10, 2013 12:37 pm Post subject: |
|
|
Now let's see a bit of the theoretical side. For this the wikipedia page about Rise time come in hand.
The two most important formulas:
For calculating rise time:
And for calculating the resultant rise time of serial blocks with each a given rise time:
So given the signal path in the PG xtra consists of three main stages, the IC on the GAIN CTL has a 100MHz advertised bandwidth, but in its actual application it should be good for up to 250MHz so its rise time is: 1.4ns
VIDEO OUT PWB has about 200MHz BW -> 1.75ns
Neck board 130MHz -> 2.7ns
The resultant rise time is 3.5ns so the theoretical bandwidth is 99.7MHz, which is pretty close to the factory specs.
With the CXA1779 (or CXA1299) on the GAIN CTL board, its theoretical bandwidth was increased to 380MHz -> 0.92ns
Recalculating the overall rise time it gives 3.35ns which means 105MHz BW for the machine, this explains why I wasn't able see any improvement with this mod.
The neckboard mod that I am working on:
The PG xtra (and XG750, 1100, 1350) neck boards are based on the so called push-pull configuration, which is consist of the VPA13 ic, and a video amplifier made from discrete parts. The first thought was to replace the transistors with faster ones, but this didn't made much difference alone (I haven't played with peaking so far) but there is one other (and last) possibility, to replace the VPA13 with the VPA15H ic, which can be found on Sony 1292 neckboards for example, in theory the VPA15H has 150MHz BW. If all of that bandwidth could be utilized and with the modded GAIN CTL board the overall -theoretical- bandwidth of the machine would be 114MHz, the problem is the 1292 boards are not cheap to get, and not sure if that extra bandwidth worth the effort/money...
_________________
projectors in the past : NEC 6-9PG xtra, Electrohome Marquee 6-7500, NEC XG 1351 LC ( with super modified Electrohome VNB neckboard !!!)
current: VDC Marquee 9500LC
The MOD: VNB-DB, VIM-DB
|
|
| Back to top |
|
|
mp20748
Joined: 12 Sep 2006
Posts: 5689
Location: Maryland
TV/Projector: 9500LC Ultra / Super 02 and 03 VIM
|
| Posted: Wed Apr 10, 2013 1:00 pm Post subject: |
|
|
|
Can you post a diagram of that board, without your mod involved (stock). I may be able to help out here.
|
|
| Back to top |
|
|
gjaky
Joined: 05 Jun 2010
Posts: 2802
Location: Budapest, Hungary
|
| Posted: Wed Apr 10, 2013 1:26 pm Post subject: |
|
|
Sure. I had the idea to change the Q3805 and Q3804 with BFQ232 and BFQ252, Scott suggested to increase the (dis)charging current through Q3802, and lowering R3809, but did not give much more performance, but generated much more heat instead.
_________________
projectors in the past : NEC 6-9PG xtra, Electrohome Marquee 6-7500, NEC XG 1351 LC ( with super modified Electrohome VNB neckboard !!!)
current: VDC Marquee 9500LC
The MOD: VNB-DB, VIM-DB
|
|
| Back to top |
|
|
mp20748
Joined: 12 Sep 2006
Posts: 5689
Location: Maryland
TV/Projector: 9500LC Ultra / Super 02 and 03 VIM
|
| Posted: Wed Apr 10, 2013 1:47 pm Post subject: |
|
|
No, I meant the first stage board. the one that has your board attached to it.
The neck boards are fine. you only need to replace C3810 and C3812 with 22uf electrolytics. Nothing fancy, but they would need to be 105 temp.
|
|
| Back to top |
|
|
flipdog
Joined: 22 Dec 2011
Posts: 156
|
| Posted: Wed Apr 10, 2013 2:02 pm Post subject: |
|
|
|
On those neckboards with a VCC of 110v your actual bandwidth is probably around 110 to 125mhz but your video signal is around 55Vp-p. If you decrease your VCC to 90v your bandwidth goes up to around 130 to 135mhz but your video signal goes down to about 45Vp-p. So it would then depend on how it would respond to the lesser signal and if enough adjustment compensation is available. Or increase the V-in signal slightly.
|
|
| Back to top |
|
|
gjaky
Joined: 05 Jun 2010
Posts: 2802
Location: Budapest, Hungary
|
| Posted: Wed Apr 10, 2013 4:56 pm Post subject: |
|
|
| flipdog wrote: | | On those neckboards with a VCC of 110v your actual bandwidth is probably around 110 to 125mhz but your video signal is around 55Vp-p. If you decrease your VCC to 90v your bandwidth goes up to around 130 to 135mhz but your video signal goes down to about 45Vp-p. So it would then depend on how it would respond to the lesser signal and if enough adjustment compensation is available. Or increase the V-in signal slightly. |
I think it's not that simple, at least to say if you increase the signal at preamp stage you loose bandwidth there...
Mike, here is the input board:
_________________
projectors in the past : NEC 6-9PG xtra, Electrohome Marquee 6-7500, NEC XG 1351 LC ( with super modified Electrohome VNB neckboard !!!)
current: VDC Marquee 9500LC
The MOD: VNB-DB, VIM-DB
|
|
| Back to top |
|
|
mp20748
Joined: 12 Sep 2006
Posts: 5689
Location: Maryland
TV/Projector: 9500LC Ultra / Super 02 and 03 VIM
|
| Posted: Wed Apr 10, 2013 7:51 pm Post subject: |
|
|
|
Let me know when we can shoot back and forth with each other, because i have a few questions, and then I'll give a few recommendations that will make a difference.
|
|
| Back to top |
|
|
kabuby77
Joined: 28 Mar 2011
Posts: 147
Location: Italy
|
| Posted: Thu Apr 11, 2013 12:33 pm Post subject: |
|
|
| gjaky wrote: | .....The neckboard mod that I am working on:
The PG xtra (and XG750, 1100, 1350) neck boards are based on the so called push-pull configuration, which is consist of the VPA13 ic, and a video amplifier made from discrete parts. The first thought was to replace the transistors with faster ones, but this didn't made much difference alone (I haven't played with peaking so far) but there is one other (and last) possibility, to replace the VPA13 with the VPA15H ic, which can be found on Sony 1292 neckboards for example, in theory the VPA15H has 150MHz BW. If all of that bandwidth could be utilized and with the modded GAIN CTL board the overall -theoretical- bandwidth of the machine would be 114MHz, the problem is the 1292 boards are not cheap to get, and not sure if that extra bandwidth worth the effort/money... |
Oh, someone speaking of 1292  .
If you see the formulas, you understand that the slower device limits the maximum speed.
So you have 3 choices:
1 - Change the neck board amp chip (the slower device)
2 - Change the input stage.
3 - Change both.
You've already done the calculation for the first case. I'll note that despite the use of a 150MHz chip, you get a band of 114MHz , that is very close to 120MHZ of VPA13. So it comes the second choice: if you are able to build an input stage with GHz performance ,when you do the calculatuion with the sum under the square root the dominant term is that of vpa13.
You get the same result as the first point. I suggest you look at the input stage of 909 with eisemann mods. I think he changes the original through hole mux chip (@250MHz) with the pin equivalent (@500MHz). If I remember correctly there is a SMD version with 2 GHz performance avaible on the market 
You can find some VPA18 180MHz @90V supply on ebay.
|
|
| Back to top |
|
|
gjaky
Joined: 05 Jun 2010
Posts: 2802
Location: Budapest, Hungary
|
| Posted: Thu Apr 11, 2013 1:29 pm Post subject: |
|
|
| kabuby77 wrote: |
You've already done the calculation for the first case. I'll note that despite the use of a 150MHz chip, you get a band of 114MHz , that is very close to 120MHZ of VPA13. |
But the point is, with the VPA13 the sum bandwidth is around "only" 100MHz, and if you put a very high frequency input stage you only converging to 130MHz absolute maximum bandwith (the slowest) while using a GHz capable input stage may introduce high frequency oscillations, radio interferences, noise, which are very hard to get rid of, not to mention the PCB (traces) wasn't designed to cope with that high frequencies. NEC used highly application specific ics which are hard to replace with a better one, this would practically mean that you'd have to redesign the whole input stage, if you want a similar configuration like the Barco 909, that is a thing in what I don't want to get involved. My idea is to gain a little more performance in every (or most) stage, that don't need a massive redesigning, yet causing an overall improvement.
The problem with VPA18 is it only can stand 90Vcc, while the VPA13 gets +110V, so you'd have to modify the power supply lines in order to get the required driving voltage for the tubes, this is also a more complicated thing to jump into it. The VPA15H is some special chip that Sony used, I haven't found any datasheet from that (the VPA15 -hithout H at the end, can only stand 90Vcc), but Sony fed the VPA15Hs from +100V in the 1292, and from +105V in the G70, so it seems they should be more or less plug-and-play replacement to VPA13s.
_________________
projectors in the past : NEC 6-9PG xtra, Electrohome Marquee 6-7500, NEC XG 1351 LC ( with super modified Electrohome VNB neckboard !!!)
current: VDC Marquee 9500LC
The MOD: VNB-DB, VIM-DB
|
|
| Back to top |
|
|
kabuby77
Joined: 28 Mar 2011
Posts: 147
Location: Italy
|
| Posted: Mon Apr 15, 2013 9:01 am Post subject: |
|
|
| gjaky wrote: | But the point is, with the VPA13 the sum bandwidth is around "only" 100MHz, and if you put a very high frequency input stage you only converging to 130MHz absolute maximum bandwith (the slowest) while using a GHz capable input stage may introduce high frequency oscillations, radio interferences, noise, which are very hard to get rid of, not to mention the PCB (traces) wasn't designed to cope with that high frequencies.....
|
I've got both moome internal ad external. The internal unit replaces the input stage of the projector, and it is connected directly to the neck board, so you can delete a summand in the formula. It uses a 1,65 GHz buffer amp, it seems logical that the next stages should maintain the same performance. I am agree that not easy to handle circuits with these frequencies.
| gjaky wrote: |
The problem with VPA18 is it only can stand 90Vcc, while the VPA13 gets +110V, so you'd have to modify the power supply lines in order to get the required driving voltage for the tubes, this is also a more complicated thing to jump into it. The VPA15H is some special chip that Sony used, I haven't found any datasheet from that .... |
I've got somewhere, the H chips works at 120V with same performance of 100V.
Modification of power supply is easier than you think, I' ve bought the TL783 from TI, It's like LM317 adjustable voltage regulator. Since no one uses the tubes at 100% of brightness, you can supply the VPA also less than 100V.
|
|
| Back to top |
|
|
gjaky
Joined: 05 Jun 2010
Posts: 2802
Location: Budapest, Hungary
|
| Posted: Mon Apr 15, 2013 9:14 am Post subject: |
|
|
| kabuby77 wrote: | I've got both moome internal ad external. The internal unit replaces the input stage of the projector, and it is connected directly to the neck board, so you can delete a summand in the formula. It uses a 1,65 GHz buffer amp, it seems logical that the next stages should maintain the same performance. I am agree that not easy to handle circuits with these frequencies.
|
In a sony maybe, but in a NEC PG xtra there is no room for optional input board, it has only one RGBHV input, you can't bypass the first stage as it plays important role in contrast adjusting, beam limiting and test pattern switching, just to mention the most.
_________________
projectors in the past : NEC 6-9PG xtra, Electrohome Marquee 6-7500, NEC XG 1351 LC ( with super modified Electrohome VNB neckboard !!!)
current: VDC Marquee 9500LC
The MOD: VNB-DB, VIM-DB
|
|
| Back to top |
|
|
digitalayon
Joined: 02 Mar 2009
Posts: 921
|
| Posted: Mon Apr 15, 2013 3:52 pm Post subject: |
|
|
|
By the time this project is finished, we'll all be selling used digital 4K projectors for 50 bucks!!
|
|
| Back to top |
|
|
AnalogRocks
Forum Moderator
Joined: 08 Mar 2006
Posts: 26706
Location: Toronto, Ontario, Canada
TV/Projector: Sony 1252Q, AMPRO 4000G
|
| Posted: Mon Apr 15, 2013 3:58 pm Post subject: |
|
|
| digitalayon wrote: | | By the time this project is finished, we'll all be selling used digital 4K projectors for 50 bucks!! |
Meh, I might even buy a digital if it's under $100 bucks
_________________
Tech support for nothing
CRT.
HD done right!
|
|
| Back to top |
|
|
gjaky
Joined: 05 Jun 2010
Posts: 2802
Location: Budapest, Hungary
|
| Posted: Mon Apr 15, 2013 8:19 pm Post subject: |
|
|
| digitalayon wrote: | | By the time this project is finished, we'll all be selling used digital 4K projectors for 50 bucks!! |
Hah, Mike has some parts to change, but he have to look up what were those in the XG (it surely helps PG xtra too) then there are the VPA15H ics, after that I'll stop pursuing this... Not that I want to retire my NEC anytime soon 
_________________
projectors in the past : NEC 6-9PG xtra, Electrohome Marquee 6-7500, NEC XG 1351 LC ( with super modified Electrohome VNB neckboard !!!)
current: VDC Marquee 9500LC
The MOD: VNB-DB, VIM-DB
|
|
| Back to top |
|
|
CasetheCorvetteman
Joined: 09 Nov 2008
Posts: 6326
Location: Australia
|
| Posted: Tue Apr 16, 2013 3:48 am Post subject: |
|
|
Ive already got 2 digitals and one was free!! Lucky that, cause that's about all it's f***en worth too 
|
|
| Back to top |
|
|
|
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum
|
Forum powered by phpBB © phpBB Group
|
|
