SA-9800 Dim Display: Options for Fix

[AlwaysAnalog]

Like many of you I have an SA-9800 with a dim display, which is still somewhat visible and functioning. I have done much reading over the past few months and now understand considerably more than I did but I am a non-electrical type so some of the reading has been confusing.

After many voltage / current checks as suggested by MTF and others, I believe my issue is simply an old display that is drawing more current than it did 40+ years ago. That being said, I plan to implement MTF voltage regulator plan (see Jan 2010 thread below) but wanted to know ...

Question 1: Can use a step down Buck Converter instead of the transistor described by MTF described in the Jan 2010 thread quoted below? For a non-electrical type like me it just seems easier to adjust the output voltage to Pin 10 on the Indicator Assembly board (AWV-002) through the Buck Converter than adding resistors to find the best display results, albeit the converter is a bit more expensive to purchase.

Question 2: I believe the VFD is static driven (i.e. common grid across the display and driven by a single voltage input at pin 10). If that is the case, it seems like I could simply remove the voltage to the grid and allow only the anode (segment) positive voltage to attract the electrons from the cathode (filament) wires. The grid does "defuses" the electrons as they pass through the grid towards the segments so there may be some visual difference without the grid voltage present but maybe not enough to be distracting. This was tried by Backlight in the JAN 2010 thread (see below) but there was no discussion about the issues in doing this. I do realize that the DIM/BRIGHT function will no longer work because that button changes the grid voltage and if the grid voltage is removed than no functionality will exist through the button.

Background information is below for those of you that are new to these threads and have the same issues with their display. Hopefully the below will save you some searching time. ALSO, I think I want to build a drop in LED version of the display. I will start another thread on that subject when I am ready.

https://en.wikipedia.org/wiki/Vacuum_fluorescent_display
What is a Vacuum Fluorescent Display (VFD). Read this to understand how the VFD is made and operates.

JAN 2010: http://www.audiokarma.org/forums/index.php?threads/sa-9800-fluorescent-display-issue.277855/
Definitive thread and worth reading several times. It contains the voltage regulator concept proposed by MTF.

FEB 2013: http://www.audiokarma.org/forums/in...-it-try-to-fix-it-or-flip-it-thoughts.503200/
Contains a reference to clean the "metal contacts" to improve display performance. I could not determine what metal contact being referred to but I did try to clean the contact between the VFD and Indicator Board as well as using Deoxit on the Dim/Bright switch. Unfortunately, no improvement.

MAR 2015: http://www.audiokarma.org/forums/index.php?threads/another-sa-9800-dim-display-issue.650700/
Contains a sentence from MTF that reiterates his belief "as the tube ages, more current is drawn, shifting the voltage points in the original circuit. It's NOT designed for the tube's aging over 30 years!"

JUN 2015: http://www.audiokarma.org/forums/index.php?threads/pioneer-sa-9800-weird-display.664259/
Latest thread I could find on the subject. Interesting read.

 

[leesonic]

Mine has the same problem. I tried grounding pin 10 on the display board, didn't make much different. I tried replacing the 390 ohm resistor on dim/bright switch (more resistance makes the display brighter), but that didn't make any difference either.

Since 9800 parts command a premium, I thought about using the display from an SA-8800 or even 7800.

Lee.

P.S. Is Mark still around? I asked a question on my 9800 restoration thread, and then I sent him a PM, but I didn't hear anything.

 

[AlwaysAnalog]

Mark's father passed away recently. I am sure we will hear from him in the near future but for now I hope he is taking time to care for himself and his family first.

Yes ... there has been threads about replacement displays but NONE are exact matches. I plan to make a very close match with LEDs (as best I can) because I assume my current VFD display will die eventually. I posted on BarterTown for a DEAD SA-9800 that will help me start the mechanical design for a drop in replacement (as much as I can make a drop in replacement). I will start a new thread about the LED project at some point in the future once I get started. First project is to make the voltage regulator suggested by MTF and see the improvement.

 

[petehall347]

unless you can somehow rejuvenate the original display whatever you do will just be a band aid until it finally gives up all hope .

 

[AlwaysAnalog]

Yes .. I know that the voltage regulator is a band-aid fix for a display that is "dying of old age". However, I may get many more years after I install the voltage regulator suggested by MTF. My replacement will be the LED version at some point in the future.

 

[jcnash]

Looking forward to the LED replacement. Both my 9800s and my 8800 and 7800 will ultimately need these...

 

[AlwaysAnalog]

Hi Folks,

EDIT: I made few mistakes in my original post which I corrected after doing the testing again and taking photos.

I went ahead with an experiment to understand the grid voltage affects on the display. I installed the first part of MTF's voltage regulator by placing a 2200 ohms variable resister across pins 9 and 11 on the protection board along with a 47 uf (80v) capacitor across the resister. I chose to use 2200 ohms instead of 1600 ohms since I also needed the voltage drop to be under the input voltage of my Buck Converter. See the Jan 2010 reference in my original post.

I turned up the variable resistance and measured a grid voltage at pin 9 on the protection board (pin 10 on the indicator board) of 35'ish volts. I then turned down the variable resistance to zero ohms and watched the grip voltage drop to nearly zero. Interestingly enough, below approximately 5 volts the display became fuzzy around the segments edges and the VU bars above 0.1W had a definite change in the glow pattern. Above 5 volts the segments became sharper and the glow went back to a nice uniform pattern (see photo attached).

This experiment showed me that a grid voltage above 5 volts definitely helps to "defuse" the electrons coming from the cathode (wire) onto the anode (segments), which is what the grid is suppose to do. Hence, running without a grid voltage does give you a fuzzy, non-uniform display although I am still not sure it harms the VFD.

As far as brightness goes, there was definitely an improvement going from the 60+ vdc grid voltage (with the unit warm) I was measuring without the variable resistor installation to my final grid voltage which I set at 19 vdc, which I selected to match my segment voltages (no good reason, just seemed OK to me). Having said this, the brightness did not change much between 10 vdc and 35 vdc so I could have lived with a grid voltage of, say 30 vdc as easily as 19 vdc. See photo attached.

I do not plan to install the buck converter or transistor (per MTF) until I run the unit for a while in this configuration.

 

[txturbo]

I had been working on a Carver receiver last week that had a dead VFD display, turns out to be a bad cap.

Anyway, in my online studies I came across this on rejuvenation of a FVD...
" I hooked up my DC power supply to the filaments with the 3.0VAC transformer taps disconnected. I did 3x short ramps with higher current till they started to slightly glow, each ramp for a few seconds. Hooked everything back up, and fired it up. The display was much much brighter, 100% better."

I had removed my display from the Carver and did what the article suggested while bench testing the display. No ill effects that I can see at the moment and the display is nice and bright.

http://www.electronicspoint.com/thr...cent-displays-revitalized-observation.227632/


http://www.vintage-radio.net/forum/showthread.php?t=94870


Looks like the trick is to burn the oxidation off of the filament.. Similar to the way a picture tube is revitalized.

 

[AlwaysAnalog]

HOLY CRAP! This sounds great but I do not know enough about VFDs (or electronics for that matter) to comment intelligently. I also do not have the equipment to give this a try but it would be great is someone out there with a non-functioning Pioneer VFD would try to duplicate this. Seems like it should work for any VFD if it worked for one. Anything to get a few more years, eh!

 

[petehall347]

i know nothing about it ... rejuvenation would be good ...does it really work ?
i remember tv tubes could be done some times ...

 

[txturbo]

It's not that hard to do... You just need to gain access to the filament lines to the display. I did it with a variable power supply, I just increased the voltage till you can see the filament glow a bit...

This is a video I shot when I was testing the display for the Carver.. The display is nice and bright now however my display was dead for other reasons.

 

[markthefixer]

Remind me later when I am back in, done with dealing with my dad's estate (and MY personal/financial repercussions AFTER that), to get into this.

An additional point is to look at the sx-3900 and sx-d7000 display circuitry, they run on different voltages, and don't seem to have this problem.

Nice work again, txturbo!!!

 

[1TUFSS]

Bumping this thread in case anyone's given this any thought. I do have an SA-9800 with a working display, but I suspect that one day in the future, I will face this issue too. Let's see what we can do to save these units.

 

[AlwaysAnalog]

It has been 9 months since I changed my grid voltage to improve the brightness. So far, everything is still operating perfectly!! I think I will leave it alone until I start seeing the brightness fade again. At that point, I will try to rejuvenate the display like txturbo suggested above. From my perspective, I think we have several good options from markthefixer and txturbo (discussed in this thread) to keep our displays going for a while yet.

 

[1TUFSS]

Also @ Mark, one major point of difference is that the SX-3900, SX-D7000 (and possibly CT-F1200) use a pulse-width modulated dimmer, whereas the SA-9800 uses constant voltage. Okay, just checked again and that's only 50% correct - they use PWM for the tuner display, the power meters are still static drive, with brightness adjusted by grid voltage. Still, I have never seen a VFD die in any of those units. SA-9800s seem to die fastest... I've heard of a couple of SA-8800s doing it too, but definitely not to the severity or frequency of 9800s.

@AlwaysAnalog: So to sum it up, 2200 ohm trim pot between pins 9 and 11 on the protection board in parallel with a 47μF capacitor? I'd test this myself if I had a dead display to try it on, but for now, let's see whether we can obtain a repeatable result from others and hash out a sure-fire way to bring these displays back to life.

 

[AlwaysAnalog]

For me, the 2200 ohm resister was a better choice than the1600 ohm resister that markthefixer suggested. My reasons are stated above.

By the way, my VFD was not dead ... just very dim. I suspect, from information provided by markthefixer, that there will be no way to extend the life forever ... just a way to improve the brightness. At some point my VFD will die. I believe the SA-9800 suffers more failures because of the DIM switch capability. I think I read something about this in a thread elsewhere.

 

[1TUFSS]

Yes, nothing lasts forever. However, these are good VFDs - they should NOT die this quick. Pioneer VFD clocks, for example, are regularly cited as running constantly for years or even decades without failing.

I don't have a 7800 or 8800 to test, but it seems that the voltages used to drive these displays are the deciding factor. 3000 series receivers don't fail, SA-7800s don't fail, SX-D5000s and 7000s don't fail. It's only the 8800s I've seen starting to go dim, and the 9800s actually doing it.

 

[txturbo]

Has anyone tried to rejuvenate a display?

 

[1TUFSS]

Has anyone tried to rejuvenate a display?

Not so far.

 

[markthefixer]

I'm probably first in line to try, once the bench is cleared for action.

That's still a little ways off....