DRags you need to make yourself a couple extender boards buddy!!!
Athanasios
the Mobo has +24VDC plainly marked next to pin 17 so I think that's pretty clear. Biggest issue now is cleanly soldering the wire to the pin on P12 so it looks good. I was having trouble last night with the 16 gauge wire which matches the yoke lead but is a little large and rigid which makes it difficult to line up and solder to the pin _________________ MARQUEE HD MODS: Bring your Marquee CRT projector (or clone) back up to spec and offer even greater performance!
Joined: 08 Mar 2006 Posts: 5086 Location: Colorado
Link Posted: Sun Jan 08, 2012 12:54 am Post subject:
it works first try. Dennis and Scott you guys are awesome.
Machine fired right up and only thing I noticed was Red convergence had drifted a hair on the left side. The one thing that does stick out
Dennis did you see any difference in the noise level on G + B tubes? I know MP had posted the noise on those 2 tubes was coming from the 800V sweep but I swear it's no longer visible, at least not obvious like it used to be, I can't see it.
I'll have to pop a stock board back in to verify but I plainly remember seeing small white horizontal flashes, like sparks from a grinding wheel, all over blue and green. Could this have been witching noise from the old FET's, or something to do with the old FET's and GTO's?
Anyway, after 2 hours the Heat-sink is still room temperature _________________ MARQUEE HD MODS: Bring your Marquee CRT projector (or clone) back up to spec and offer even greater performance!
Joined: 08 Mar 2006 Posts: 5086 Location: Colorado
Link Posted: Sun Jan 08, 2012 1:37 am Post subject:
Nashou66 wrote:
Nice work Drags!!!
Athanasios
thanks,Almost makes me want to go back to college for an EE degree,.,,,,,,,,, almost. For now it will just have to be Monkey-see, monkey-do but at least I'm able to do
Theoretically on any marquee you should see it since the noise is coming from the sweep. I doubt it's something they were able to filter out on the late model boards but who knows?
Other possibility is is starts showing up only on high hour boards where the big power FET's are aging. Most of these things have 20 to 50,000 hours on them by now.
I dunno, still have to run 1080P as the machine was set for 720 at this time but this one looks like a winner. thumbsup: _________________ MARQUEE HD MODS: Bring your Marquee CRT projector (or clone) back up to spec and offer even greater performance!
Joined: 08 Mar 2006 Posts: 5086 Location: Colorado
Link Posted: Mon Jan 09, 2012 5:25 pm Post subject:
tse wrote:
The capacitors that parallel the mosfets set the retrace time. Because C4 and C5 are being removed from the circuit the total capacitance is increased which will increase the retrace time from about 1.6us to about 1.84us (in short retrace mode). Not a big change but it is a change.
If the caps have the values shown in this schematic (some HDM versions are different) then the capacitance of C6 and C7 will be different than C10 and C11. The two mosfets will have different flyback voltage across them. Ideally these capacitances would be the same.
If C6 was changed to a 2.2nF cap like C10 the values across each switch would be the same and the retrace time would be about 1.72us short retrace) which is pretty close to original.
To calculate the retrace time determine the resonant frequency of the deflection yoke and the parallel/series value of the flyback caps. The time of this is one complete cycle. The flyback pulse is one half the time of one cycle.
L = yoke inductance in henries or 67 x 10^-6
C = parallel / series value of the flyback caps (don't forget about C8 and C13 which are switched in during long retrace) in farads.
So:
1 / (2 x pi x square root of L x C) = resonant frequency
resonant frequency / 2 = retrace time (add about 0.3us to this to make up for mosfet capacitance, yoke resistance, the inductance of the small balancing coils, and turn off delay time).
Scott
ok, so the second half of all that blew my little brain apart but I think I found a problem yesterday with Short retrace and the current caps on this board, where C6 and C8 = 6.8nf. Initially I read the above as "make sure the caps are the same value" but I see now a higher value on the caps = longer retrace time and by removing C4 and C5 it's even more exaggerated. In checking, all my boards here from 1995/96 have 6.8nf in positions C4, C6, and C10. I'm guessing the schematic Dennis posted is from a newer board?
So on older boards both C10 and C6 might need replacing with 2n2? In Short retrace I'm getting a little "tail" on the test pattern grind at the beginning on the scan sweep (left side / table mount) looks like this
Joined: 08 Mar 2006 Posts: 5086 Location: Colorado
Link Posted: Mon Jan 09, 2012 6:34 pm Post subject:
Nashou66 wrote:
I get that too Drags,
really? I've never seen that before
Nashou66 wrote:
changing timings with a VP or HTPC fixes it easier than replacing caps etc.
But i can see the value in a plug and play 1908p CRT projector by changing caps to adjust the sweep trace times to accommodate 1080p more efficiently. Athanasios
Joined: 12 Jan 2007 Posts: 12985 Location: West Seneca NY
TV/Projector: Marquee 8000, 8500 And a 9500LC RetroIV , 2 Longbow 8500 Ultras(2004!!)Hd145's , Ampro 3600, a G90!!
Link Posted: Mon Jan 09, 2012 8:41 pm Post subject:
Drags in answer to your PM, I didn't find the pic with both side by side but I found the link in my Longbow thread where i posted both HDM's in consecutive pics.
Joined: 08 Mar 2006 Posts: 5086 Location: Colorado
Link Posted: Mon Jan 09, 2012 8:58 pm Post subject:
thanks for looking. I found that one but it doesn't show the difference nearly as well as the profile view. I wish there was a way to search just pictures.
Also, I wish we knew the name or part number of the new heat sink in it's raw form, the part number from the company doing the Extrusion. I would buy a 6 foot piece, cut it to lengths, drill/tap the 2 holes, and add the tape. _________________ MARQUEE HD MODS: Bring your Marquee CRT projector (or clone) back up to spec and offer even greater performance!
Joined: 12 Jan 2007 Posts: 12985 Location: West Seneca NY
TV/Projector: Marquee 8000, 8500 And a 9500LC RetroIV , 2 Longbow 8500 Ultras(2004!!)Hd145's , Ampro 3600, a G90!!
Link Posted: Mon Jan 09, 2012 11:15 pm Post subject:
Man just had some nostalgia here. Went back looking all over for that pic and came across the SAB and CMM issues with my Longbows in another thread. That was where you accidentally bought the SOIC chips instead of DIP chips for the SAB. Did you ever buy those SOIC to DIP adaptors to test those out?
Joined: 08 Mar 2006 Posts: 5086 Location: Colorado
Link Posted: Wed Jan 11, 2012 5:37 pm Post subject:
So dropped in a stock HDM last night and the problem I posted above is not there. With the stock board there is a very small ripple at the beginning of the grid line but nothing like the big tail I posted above which is roughly to scale. I'm going to change the caps at C6 and C10 with 2n2 and see if that fixes it.
Where's Dennis? I would like to hear if he's running short retrace and if he's seeing the above problem? _________________ MARQUEE HD MODS: Bring your Marquee CRT projector (or clone) back up to spec and offer even greater performance!
Joined: 08 Mar 2006 Posts: 5086 Location: Colorado
Link Posted: Wed Jan 11, 2012 7:46 pm Post subject:
Had a little extra energy last night to go out into the shop after work and found another interesting observation, I wonder if this is where the bigger VDC HDM heat-sink came from?
the heat-sink on the left is from the FCM. To get it to fit you need off nip off the 2 existing mounting tabs (not pictured) , drill and tap a pair of 6-32 holes on the center of the main center section, and also mill or hacksaw off
1) the lowest fin on the right side to clear the existing parts on the HDM.
2) Half of the top/left fin to clear the card-cage
All in all I think the amount of heat-dissipating surface area is increased by 300%
Joined: 08 Mar 2006 Posts: 5086 Location: Colorado
Link Posted: Thu Jan 12, 2012 5:39 pm Post subject:
HK-Steve wrote:
Here is a pic,
nearly the same as your proposed idea.
thanks, that's exactly the picture I was looking for. You can see by the rounded edges on the fins that the VDC heat-sink is not simply a trimmed/cut Focus board piece but a specific extrusion. However as you said by the time you finish modifying the FCM piece it looks almost identical. I got it to fit last night after work and it looks good. Last thing left to do is measure some temp's after like 2 or 3 hours run time at 1080P short retrace, will do that this weekend.
Joined: 12 Jan 2007 Posts: 12985 Location: West Seneca NY
TV/Projector: Marquee 8000, 8500 And a 9500LC RetroIV , 2 Longbow 8500 Ultras(2004!!)Hd145's , Ampro 3600, a G90!!
Link Posted: Thu Jan 12, 2012 5:42 pm Post subject:
draganm wrote:
HK-Steve wrote:
Here is a pic,
nearly the same as your proposed idea.
thanks, that's exactly the picture I was looking for. You can see by the rounded edges on the fins that the VDC heat-sink is not simply a trimmed/cut Focus board piece but a specific extrusion. However as you said by the time you finish modifying the FCM piece it looks almost identical. I got it to fit last night after work and it looks good. Last thing left to do is measure some temp's after like 2 or 3 hours run time at 1080P short retrace, will do that this weekend.
1) Temp of stock 07P board
2) Temp. of 07P board with Cree FET modification
3) Temp. of 07P board with Cree FET modification + bigger HS
There should be a temp. drop as I go down the list but I need to be sure and take pictures I'll post here
Drags add a temp of the stock 07P amps with bigger heat sink as well so we can see how much the larger heat sink
dissipates heat.
Nashou _________________ Don't blame your underwear for your crooked ass~ unknown Greek philosopher
" To be truly Free is to be reliant on no one, but the Author of our Destiny"~~~ Elbert Guillory"
One Smart Dog!!! Tom Bauerle to Andre Cuomo" Elliot Spitzer Screwed some Prostitutes, you screwed the Constitution" On his call to Cuomo to Resign!!!
I'm not publishing this in a scientific journal and have no plans to log data over the course of 9 hours . oh yeah i know, it's real simple just set-up a computer in the shop and hook up the Thermocouple device with a USB cable into Labview program and ........... ummm, yeah, not going there
This is going to be very simple before/after temp readings with a typical movie run. I think Gladiator is like 2:30 run time?
honestly I don't believe there's a market for this modification and the parts are fairly pricey so my only goal at this time is to help the 2 or 3 guys who are having problems with oscillating or unstable rasters. My bigger concern is bike racing season which starts march 3rd and I'm hauling around an extra 5 or 6 pounds from the dam Holidays _________________ MARQUEE HD MODS: Bring your Marquee CRT projector (or clone) back up to spec and offer even greater performance!
Joined: 08 Mar 2006 Posts: 5086 Location: Colorado
Link Posted: Fri Jan 20, 2012 6:16 pm Post subject:
so bad news , before testing the temp. on the Cree'd HDM I wanted to lower the value on the caps at C10 and C6 to 2n2 from 6n8. (the schematic posted earlier showed 2n2 and 4n7 in those locations but the 95 through 97 boards have a higher value here). The other 2 caps in parallel at C11 and C7 remain 3n9 on all the boards and the schematic as well.
Well the PJ started up fine but as soon as a source was applied and machine switched to a recall with short retrace the machine shut down with H fail lit. So it looks like this is not a good way to go.
Originally on this board I started with
1)original configuration- three of the big power fets and 32.1nf total capacitance in parallel (six caps)
then,
2) 2 cree fets and four caps (two 3n9 + two 6n8) = 21.4nf , machine ran fine
and finally
3) 2 cree fets and four caps (two 3n9 + two 2n2)= 12.2nf = H-fail
So it seems to me like we have dropped the capacitance of the ciruit down too low. However Scott say's
tse wrote:
The capacitors that parallel the mosfets set the retrace time. Because C4 and C5 are being removed from the circuit the total capacitance is increased which will increase the retrace time from about 1.6us to about 1.84us (in short retrace mode). Not a big change but it is a change.
Joined: 12 Jan 2007 Posts: 12985 Location: West Seneca NY
TV/Projector: Marquee 8000, 8500 And a 9500LC RetroIV , 2 Longbow 8500 Ultras(2004!!)Hd145's , Ampro 3600, a G90!!
Link Posted: Fri Jan 20, 2012 7:05 pm Post subject:
Caps are a funny thing, if you look at the schematic there are caps in Parallel as in C4 and C5 but you also have caps in series and when in series capacitance decreases. So removing those two caps you take two caps out of series which will raise the total capacitance.
From all about circuits:
When capacitors are connected in series, the total capacitance is less than any one of the series capacitors' individual capacitances. If two or more capacitors are connected in series, the overall effect is that of a single (equivalent) capacitor having the sum total of the plate spacings of the individual capacitors. As we've just seen, an increase in plate spacing, with all other factors unchanged, results in decreased capacitance.
Thus, the total capacitance is less than any one of the individual capacitors' capacitances. The formula for calculating the series total capacitance is the same form as for calculating parallel resistances:
When capacitors are connected in parallel, the total capacitance is the sum of the individual capacitors' capacitances. If two or more capacitors are connected in parallel, the overall effect is that of a single equivalent capacitor having the sum total of the plate areas of the individual capacitors. As we've just seen, an increase in plate area, with all other factors unchanged, results in increased capacitance.
Thus, the total capacitance is more than any one of the individual capacitors' capacitances. The formula for calculating the parallel total capacitance is the same form as for calculating series resistances:
As you will no doubt notice, this is exactly opposite of the phenomenon exhibited by resistors. With resistors, series connections result in additive values while parallel connections result in diminished values. With capacitors, its the reverse: parallel connections result in additive values while series connections result in diminished values.
REVIEW:
Capacitances diminish in series.
Capacitances add in parallel.
Nashou _________________ Don't blame your underwear for your crooked ass~ unknown Greek philosopher
" To be truly Free is to be reliant on no one, but the Author of our Destiny"~~~ Elbert Guillory"
One Smart Dog!!! Tom Bauerle to Andre Cuomo" Elliot Spitzer Screwed some Prostitutes, you screwed the Constitution" On his call to Cuomo to Resign!!!
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 FAQ
Search
Memberlist
Photo Albums
Register
Profile
Private Messages
Log in
Joined: 12 Jan 2007
Joined: 15 Jul 2006
Link
Dennis and Scott you guys are awesome.
