| Author |
Message |
the dude
Joined: 11 Sep 2013
Posts: 179
|
| Posted: Sat Feb 13, 2016 8:20 pm Post subject: |
|
|
|
Yes. You can see them, but very faint. The vertical lines are dark. I have a hd fury2. I testet it with a computer monitor and it performs very well. I was sure with new neckboards this would be better.
|
|
| Back to top |
|
 |
the dude
Joined: 11 Sep 2013
Posts: 179
|
| Posted: Sat Feb 13, 2016 9:44 pm Post subject: |
|
|
Checked with 300 Mhz scope...
At switcher output the signal has almost the sample amplitude as a 2:2 pattern. The ringing is from the proble ground cable i guess.
Next thing will be checking the driver. Its a little bit complicated because the machine is hanging on the ceiling...
| Description: |
|
| Filesize: |
152.32 KB |
| Viewed: |
5316 Time(s) |
|
|
|
| Back to top |
|
|
redfox001
Joined: 16 Mar 2009
Posts: 2257
Location: The Netherlands
|
| Posted: Sat Feb 13, 2016 10:17 pm Post subject: |
|
|
| the dude wrote: | | Yes. You can see them, but very faint. The vertical lines are dark. I have a hd fury2. I testet it with a computer monitor and it performs very well. I was sure with new neckboards this would be better. |
The biggest improvement in bandwidth came from the barco full hd internal hdmi. I have seen and tested the hdfury3 and it is MUCH worse really. There are more threads on that even a scope measurement from the hdfury somewhere.
_________________
701s->runco933->8500ultra->cinemax->9500mp->919 splitpack + cinemax
|
|
| Back to top |
|
|
the dude
Joined: 11 Sep 2013
Posts: 179
|
| Posted: Sat Feb 13, 2016 10:33 pm Post subject: |
|
|
Hmmm, on my Iiyama LCD monitor the pattern from the fury looks perfekt.
Just checked the driver, its fine. So i think the neckboard is the problem.
But with a white pattern the vertical lines are brighter and well visible. Maybe green only fools the eye?
But they much darker than the horizontal ones. Even for a 2:2 pattern you can see a they are a little bit darker.
I think the neckboard is not really 180 Mhz. They didn't told us how they measured that... -20 db maybe? And the hybrid is for sure the weakest part.
| Description: |
|
| Filesize: |
185.4 KB |
| Viewed: |
5302 Time(s) |
|
| Description: |
|
| Filesize: |
176.78 KB |
| Viewed: |
5302 Time(s) |
|
|
|
| Back to top |
|
|
redfox001
Joined: 16 Mar 2009
Posts: 2257
Location: The Netherlands
|
| Posted: Sat Feb 13, 2016 10:45 pm Post subject: |
|
|
We mostly use this pattern (see one of my earlier posts) and we only check the top squarre for even lightness. You should get something like this or better with your setup. I do not have the hdfury pics but the left vertical squarre is much darker. Use only green so convergende does not disturb the pic.
_________________
701s->runco933->8500ultra->cinemax->9500mp->919 splitpack + cinemax
|
|
| Back to top |
|
|
redfox001
Joined: 16 Mar 2009
Posts: 2257
Location: The Netherlands
|
| Posted: Sat Feb 13, 2016 10:52 pm Post subject: |
|
|
He this calculator shows how all parts work together. There is not one bottleneck.
| Description: |
|
Download |
| Filename: |
BW_calculator.xlsx |
| Filesize: |
9.22 KB |
| Downloaded: |
259 Time(s) |
_________________
701s->runco933->8500ultra->cinemax->9500mp->919 splitpack + cinemax
|
|
| Back to top |
|
|
redfox001
Joined: 16 Mar 2009
Posts: 2257
Location: The Netherlands
|
|
| Back to top |
|
|
the dude
Joined: 11 Sep 2013
Posts: 179
|
| Posted: Sat Feb 13, 2016 11:24 pm Post subject: |
|
|
Thanks for the infos!
Here comes the test pattern
| Description: |
|
| Filesize: |
160.6 KB |
| Viewed: |
5268 Time(s) |
|
|
|
| Back to top |
|
|
gjaky
Joined: 05 Jun 2010
Posts: 2802
Location: Budapest, Hungary
|
| Posted: Sun Feb 14, 2016 8:06 am Post subject: |
|
|
| the dude wrote: | Checked with 300 Mhz scope...
At switcher output the signal has almost the sample amplitude as a 2:2 pattern. The ringing is from the proble ground cable i guess.
Next thing will be checking the driver. Its a little bit complicated because the machine is hanging on the ceiling... |
What probe do you have there? I found out that passive probes are utterly useless for video signal probing. I have an old HP1720A oscilloscope (275MHz), and have also Tek 6106 probe (250MHz), according to Tektronix this probe should deliver the claimed bandwidth combined with the oscilloscope, but my experience is the opposite.
Every part you put in your measuring chain will lower the bandwidth -including the probe.
Then comes the gounding lip and the tip capacitance of the passive heads, the ground clip adds considerable ammount of inductance to the ground path, you should avoid them. Also a typical 12pF tip capacitance means a considerable impedance at high frequency even in 50Ohm terminated applications.
In these applications you are better with directly probing to the signals with coaxial cable terminated at 75/50 Ohm.
I took the more expensive way as I wanted to play with neckboard outputs too, so invested in an Active FET probe (in my case a HP 1120A 500MHz probe with dividers -which perfectly matches to my scope), this probe has less than 1pF tip capacitance when a divider is used (without divider the maximal input voltage is only +/-0.5V), still the grounding wire have to be kept minimal.
_________________
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 |
|
|
redfox001
Joined: 16 Mar 2009
Posts: 2257
Location: The Netherlands
|
| Posted: Sun Feb 14, 2016 9:35 am Post subject: |
|
|
| the dude wrote: | Thanks for the infos!
Here comes the test pattern |
Yes I recognise this and got the same with the hdfury.
_________________
701s->runco933->8500ultra->cinemax->9500mp->919 splitpack + cinemax
|
|
| Back to top |
|
|
redfox001
Joined: 16 Mar 2009
Posts: 2257
Location: The Netherlands
|
| Posted: Sun Feb 14, 2016 9:40 am Post subject: |
|
|
| the dude wrote: | Checked with 300 Mhz scope...
At switcher output the signal has almost the sample amplitude as a 2:2 pattern. The ringing is from the proble ground cable i guess.
Next thing will be checking the driver. Its a little bit complicated because the machine is hanging on the ceiling... |
Great idea this test. But I think it is in the almost. A number of these almost together explain the end. Is it possible to inject the video in the neckboard and than measure this?
_________________
701s->runco933->8500ultra->cinemax->9500mp->919 splitpack + cinemax
|
|
| Back to top |
|
|
the dude
Joined: 11 Sep 2013
Posts: 179
|
| Posted: Sun Feb 14, 2016 11:35 am Post subject: |
|
|
| Quote: | | What probe do you have there? I found out that passive probes are utterly useless for video signal probing |
Jep, its passive only. Its a Tek P6133 with 10.8 pf capatitance. I just wanted to see the amplitude of the vertical lines, i know the waveform has nothing to do with the real signal.
| Quote: | | Is it possible to inject the video in the neckboard and than measure this? |
No, at least not at the ouput. A typical CRT cathode and socket circuit may have 12 pf or something like this, so even a 1pf active probe will degenerate performance a little. You need a active FET probe, this probes are quite expensive and can also be damaged by overvoltage, so measuring a CRT cathode is a quite demanding task.
I think what we really need is a push pull driver like other projectors use. The late barco design with single ended cathode drive is not good. The 1209s ist the worst of all, because they put in a 85 Mhz chip (VPA08) and pushed it with exessive peaking. Look at the peaking network! Someone must spend many days days on it working that out  . The 909 boards are much better because of a newer high performace hybrid driver wich may be the most powerfull cathode driver ever, but the design still sucks. The marquee uses discrete transistors and push pull, the older 1209 uses also push pull (but i never have seen a circuit diagramm), sony G70 uses push pull and i think the G90 also (not sure). There is a good reason for it.
Someone with the nessserary skills should add a true fast high frequency low voltage driver to the control grid. I have no experience with such drivers, but i guess someday i will try. The worst thing that can happen is frustration and magic smoke.
|
|
| Back to top |
|
|
gjaky
Joined: 05 Jun 2010
Posts: 2802
Location: Budapest, Hungary
|
| Posted: Sun Feb 14, 2016 12:20 pm Post subject: |
|
|
indeed all but Barco used push-pull output, but the Electrohome solution os the best.
I myself trying to mate the electrohome VNB with my XG135LC, and the project is almost done, only a few things have to be done, once it is finished it would mean it is possible to mix boards between different brands, so theoretically it even would be possible to mate the VNB with the barco...
https://www.curtpalme.com/forum_archived/viewtopic.php@t=35694.html
_________________
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 |
|
|
the dude
Joined: 11 Sep 2013
Posts: 179
|
| Posted: Sun Feb 14, 2016 12:38 pm Post subject: |
|
|
Respect! Nice work 
|
|
| Back to top |
|
|
redfox001
Joined: 16 Mar 2009
Posts: 2257
Location: The Netherlands
|
| Posted: Sun Feb 14, 2016 12:41 pm Post subject: |
|
|
Need some dummy help with scoping. I put up a two pixel and one pixel vertical pattern all over the screen and than scoped the point on the RGB driver where the video comes in on green. Now I get this.
The dc in between must be the blanking so the ripple must be the one or two pixel pattern. Seems the two pixel amplitude is bigger. Anything stupid done here? Thanks!
_________________
701s->runco933->8500ultra->cinemax->9500mp->919 splitpack + cinemax
|
|
| Back to top |
|
|
redfox001
Joined: 16 Mar 2009
Posts: 2257
Location: The Netherlands
|
| Posted: Sun Feb 14, 2016 12:47 pm Post subject: |
|
|
Closer look at the ripple with 3 2 1 pixelpattern.
_________________
701s->runco933->8500ultra->cinemax->9500mp->919 splitpack + cinemax
|
|
| Back to top |
|
|
redfox001
Joined: 16 Mar 2009
Posts: 2257
Location: The Netherlands
|
| Posted: Sun Feb 14, 2016 12:50 pm Post subject: |
|
|
Now my guess the difference in amplitude is caused by the probe that is on 1x and has some paracitic capacity?
_________________
701s->runco933->8500ultra->cinemax->9500mp->919 splitpack + cinemax
|
|
| Back to top |
|
|
the dude
Joined: 11 Sep 2013
Posts: 179
|
| Posted: Sun Feb 14, 2016 12:56 pm Post subject: |
|
|
|
A little decrease in amplitude can be caused by your scope, since it is 150 Mhz.
|
|
| Back to top |
|
|
redfox001
Joined: 16 Mar 2009
Posts: 2257
Location: The Netherlands
|
| Posted: Sun Feb 14, 2016 1:02 pm Post subject: |
|
|
| the dude wrote: | | A little decrease in amplitude can be caused by your scope, since it is 150 Mhz. |
Yes it is at the edge. Fun to do this. I still do not know for sure what I am looking at with these pictures below. Was it a 3 2 1 testpattern or something? The last line is shorter.
| Nashou66 wrote: | Scope pics of HD Fury III and Moome respectively.
EDIT: These were scoped by Craig Rounds of CIR Engineering. The best Calibration expert in the world!!  Sorry for not giving you credit buddy.
Fury at 1080p@60
Moome at 1080p@72 !!!
Athanasios |
_________________
701s->runco933->8500ultra->cinemax->9500mp->919 splitpack + cinemax
|
|
| Back to top |
|
|
the dude
Joined: 11 Sep 2013
Posts: 179
|
| Posted: Sun Feb 14, 2016 1:06 pm Post subject: |
|
|
|
I think so. Looks like a series of patterns, so you can see degeneration in performance at one glance.
|
|
| Back to top |
|
|
|
|
