• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

SA-9800 fluorescent display issue

markthefixer said:
I also have the preliminary information about the addition of the transistor regulator:

On the awm-163 protection assembly, pins 9 to 11, a 1600 ohm 2 watt resistor is added.

The wire from pin 9 on the awm-163 protection assembly going to pin 10 of the awv-002 indicator assembly is interrupted. This wire will now go to the base of an NPN power transistor.
Instead, the emitter of that NPN power transistor will be connected to pin 10 of the awv-002 indicator assembly

Finally a NEW wire is needed from the collector of that NPN power transistor to the +68 volts, either from pin 6 of the awm-163 protection assembly or pin 13 of the awr-194 power supply assembly.

Then all you need to do is to figure out where to mount the transistor, and either put a heatsink on it, OR do an INSULATED mounting to the metal chassis somewhere convenient to carry away the heat.

The transistor will be an NPN power transistor capable of handling 100 volts, over 100 milliamps and dissipating up to 4 watts of power (or more). It can be to-220, which will need insulating - because the metal case is energized, or it can be a TO-126 case, which would NOT need insulating.

We MIGHT want to add a small capacitor 20 to 47 uf (80v) across the 1600 ohm resistor, which will then closely imitate the sx-950's Q5 regulator circuit and preclude any chance of oscillation.
Click to expand...

I have build this regulator with success! The regulator keeps the voltage stable under all circumstances. But the 32VDC was not giving the best results I found out after testing with various settings, including the voltage on pin 7 on the indicator board which is now 19.5 volts with the replaced 18V zener and 330Ohm resistor.

I did some little tweaking and ended up having an additional 330Ohm resistor between the emitter and pin 10 on the indicator board. This keeps the voltage on pin 10 around 28VDC in bright position, which makes the display look better than the 32VDC it was without the additional resistor. And I preferred this solution over replacing the 1.6K resistor with a lower value which would not only cause a lower voltage on the transistor base, but also draw a lot more current away from the 68VDC line...

I also installed a 47uF/80V (Nichicon UPW) cap across the 1.6K resistor.

I positioned the transistor on the side chassis panel below the speaker switch. It can dissipate its heat easily. But is this the best position since it is not shielded and is quite near the tone amp board? Any comments here?

And another thing: does the additional usage of the 68VDC line affect the sound in any way? Is that possible?

Nevertheless, the display, in bright position, looks quite similar to the displays of my SA-7800 and SA-8800 now. It is slightly less bright on the right end of the screen, but acceptable. The display is not functioning according specs anymore, that is sure. But I hope it will as it is...which is acceptable.

Any comments or questions on this modification?

And if there still is somebody out there with voltage readings of a properly functioning SA-9800 display, please reveal yourself!
 
Last edited:
Backlight, glad to hear it is up and running. The fact you have a left to right luminous gradient suggests the cathode (heater) may have some DC on it (unlikely in that circuit), or the grid isn't at the same potential from left to right. The display is only grid connected at one point on the pcb- pioneer only soldered one grid connection on the left hand end of the tube pins- the others go straight through the pcb to no lands or are cut off- I'd hook up the right hand end grid pin too and see what happens. I am almost tempted to design a adjustable duty cycle grid/anode brightness pcb for this amp.
 
Hi restorer john,

Thanks very much for this new insight. If I am correct you mean the 3.6VAC directly from the small transformer? How can I detect if there's any DC on it?

And the "grid", what voltage is going to the grid? Is this pin 10 on the board?

Do I understand the functioning of the display correctly:

- The heater is supposed to heat the entire display (3.6VAC in bright position)
- The grid is fed through pin 10 and receives in my tweaked circuit 28VDC in bright position. I don't know what this value should be since I never had the chance to measure the voltage with a correct working display. What is the grid doing?
- The individual segments are fed by 19.5 VDC, which make them light up.

You suggest to connect the right hand (front faced) grid connection to the 28VDC as well and see if the display becomes evenly bright from left to right. Am I correct?
 
From my understanding and reading, it seems that display aberrrations can be caused by several factors, and running a display at the lower end of its operating voltages can also cause non-uniformity like what you are experiencing. DC can be measured by putting your meter on DC and checking the filament terminals. I figured that if the small filament transformer had poor symmetry on its windings or some were shorted, you could conceivable get DC when the zener/resistor/brightness sw was switched in. The Pioneer circuit works, but all the major VFD manufacturers specifically warn against using varying grid voltages to change brightness- and that's what ultimately their circuit does. Varying duty cycle is better, but perhaps might not work so well in a peak reading display like the power meter due to the fast fluctuations of segments. The centre tap is to 'float' the filament potential above the peak to peak swing and therefore ensure uniform brightness.
I cannot test the one here as the display is dead, although I am interested to know that yours is fine at 28volts.
Look, the grid connection should be fine- if you look into the display, there is about 5 grid connections, all to the single grid. I don't know why they didn't use the middle one for symmetry, but it shouldn't matter. Exactly what the grid does apart from accelerate the electrons on their way to the anodes, I don't know. In fact, I'm confused as to why you need a grid at all if the electrons are emitted by the filament and absorbed by the doped anode. My triode theory is pretty ordinary.
 
Thanks.

So what I can try is to connect the other grid connections to the 28VDC directly too?

For your information, If I disconnect the 28VDC for the board (resulting in 0V to the grid), the display is bright, but the segments are blurry at the edges. The switch has only a very little effect in this situation.

I don't know how much the 28VDC is below spec. But the display looks decent with it.

Another small observation worth mentioning is that in current situation, not only the display is slightly brighter on the right side, also the wattage number segments become slightly brighter when a neighbor power indicator segments lights up. Might this be caused by the grid voltage or irregularity?
 
I would bet on it being a function of the relationship of the segment voltages and the grid voltage.

In other words that 19v that drives each segment.

Might even have something to do with the heater voltage on dim, the resistor to ground and the zener diode across it.
 
So, after feeding the grid voltage to all connections, what is a possible next experiment? Lowering the 19V to the segments and increasing the voltage to the grid?

It would really be helpful to have reading of a correct working display!

The influence of neighbor segments also happens in bright position. So I think the resistor and zener don't play a role.
 
restorer-john said:
From my understanding and reading, it seems that display aberrrations can be caused by several factors, and running a display at the lower end of its operating voltages can also cause non-uniformity like what you are experiencing. DC can be measured by putting your meter on DC and checking the filament terminals. I figured that if the small filament transformer had poor symmetry on its windings or some were shorted, you could conceivable get DC when the zener/resistor/brightness sw was switched in. The Pioneer circuit works, but all the major VFD manufacturers specifically warn against using varying grid voltages to change brightness- and that's what ultimately their circuit does. Varying duty cycle is better, but perhaps might not work so well in a peak reading display like the power meter due to the fast fluctuations of segments. The centre tap is to 'float' the filament potential above the peak to peak swing and therefore ensure uniform brightness.
I cannot test the one here as the display is dead, although I am interested to know that yours is fine at 28volts.
Look, the grid connection should be fine- if you look into the display, there is about 5 grid connections, all to the single grid. I don't know why they didn't use the middle one for symmetry, but it shouldn't matter. Exactly what the grid does apart from accelerate the electrons on their way to the anodes, I don't know. In fact, I'm confused as to why you need a grid at all if the electrons are emitted by the filament and absorbed by the doped anode. My triode theory is pretty ordinary.
Click to expand...

Restorer-John, are you still around in this thread?

I plan to start some work on the SA-9800's display PCB this weekend and next week. Are all "non-connected" pins to the display going to the grid? If yes, it should be safe to feed them all the 28VDC. I can then change the voltage by replacing a resistor in the regulator circuit I have built according to mtf's specs.

Once I did this I will report back here the results. But please confirm if it is safe to connect 28VDC to which non-connected pin(s). If you need pictures let me know.

Thanks!
 
Hi Backlight, I've just completed the SA-9800, I ended up tracking down a replacement from a parted out SA-9800. I got the whole board and VFD only to find one of the VFD driver ICs was dead- didn't matter I used one from the old display pcb and it is 100% perfect.
I have a pdf I made with the voltages measured with a perfectly functioning display. see the attachment.
Hope that helps you, note I previously had +73 volts on my grid pin with essentially no current flow leading me to believe the tube had lost vacuum.
 

Attachments

Thank you very much Restorer-John!

This is very helpful! And congratulations with your good display replacement!

It looks like my voltages are very similar on pins 1 to 5 and 11 to 13, all 19VDC when lit. Pin 6 is 17VDC on my PCB. Pins 14 and 15 have 3.6VAC on my PCB, but I have to check tonight if that is in dim or bright position.

The added regulator feeding pin 10 is providing less voltage. I can increase this to easily to match the 33.7/43.3 VDC you measured.

My display is looking good on the left side, gradually dimming to right. I hope this can be changed by feeding pin 10 (grid voltage) to more than one grid connectors. Following the lanes only pin 13 (to the display itself) in receiving grid voltage. Do you know which of the other pins go to the grid as well?

I will try and connect the grid using all available connectors and hope to create an evenly bright looking display. Perhaps this also solves the grid to have increased resistance while in use. This was the reason to build the regulator, keeping the voltage stable, regardless the resistance of the grid.

Thanks again!
 
If it's gradually dimming left to right, it is more likely there's DC on the filament somehow- check that first.
I have some other SA-9800 hints for you as well- send me a PM if you want them.

John
 
If I measure AC over pin 14 and 15 on the indicator board it reads between 3.6 and 3.7 VAC. I compared it to my SA-8800 (with bright and clean display) which has the same readings.

Reading DC on the same pins is 0 Volts, so it should be ok. Or is there another measurement I need to do to find out if DC is present?

I do notice there's no change in this reading when switching to dim position, while this should affect the voltage since the middle transformer pin is no longer directly connected to ground, but goes through the resistor/zener regulator. Could mean the zener no longer works, but shorts...

Any idea?
 
A little addition: In dim position I measure 6.9VDC over the resistor/zener regulator. I measure the same 6.9 VDC on pin 14 and 15 (indicator PCB) to ground. I now understand the way the switch influences the power to the filament: AC is unaffected, but it gets an additional 6.9 VDC, which helps dim the brightness, together with increasing the DC voltage to the grid. That display is truly a piece of magic... :) Haven't got a clue how this "electronically-physically-chemically" works.
 
Left to right dimming could be caused by the following things:
DC - you've checked that
Displaced grid - unlikely as it's part of the assembly fixed down firmly
Filament coating loss at right ends - possible but again unlikely due several individual heater wires
Read the attached application notes from Futaba (they made the display and still make heaps of VFDs)
Note second part, part 5, although a gradual L-R change doesn't manifest itself as what I think you are describing.
 

Attachments

after reading all of this thread was there a final fix for this display fault?
I have just got my 9800 running again and yes i have the problem, i have close to 60v on pin 10 on the indicator board. can someone come to my rescue, thanks Barrie
 
barrie 2707 said:
after reading all of this thread was there a final fix for this display fault?
I have just got my 9800 running again and yes i have the problem, i have close to 60v on pin 10 on the indicator board. can someone come to my rescue, thanks Barrie
Click to expand...

have you seen the thread that checks the voltages and adds a transistor to regulate a display voltage?
 
thanks for the reply, not being an electronics engineer i am slightly out of my depth,but i did all the checks on the AWR194 board, that was fine . the part of the thread about the transistor i must admit i did cast my eye over but could not be sure if this was the final "fix".
As i found on the 163 the voltage was alarmingly high ,60v the display only dim,and changing between the two positions only changing the brightness slightly.
I am now going back to reread that part of the thread. Cheers! Barrie


I have reread that section


Attached Files sa-9800_indicator_board.pdf (484.8 KB, 48 views)

before i do this mod. I'm going to remove and check/replace the 390ohm and 1.2k resistors on the protection board,
 
Last edited:
I also have the same problem with my VFD on my SA-9800. It's working but is very dim. If I switch the dim button there's a very little difference.

As I'm only an amateur I removed the cable from pin 10 and got a nice bright display but of course the dim switch does not work.

Question: What can happen if I just leave pin 10 open? Will it affect or damage anything?

I also observe when the meters are "running" that they are blurry/diffuse on the right hand side and sharp on the left side and some letters are "bleeding".
 
Last edited:
You would get responses starting a new thread. Resurrecting an old thread only confuses things.
 
You must log in or register to reply here.
Back
Top Bottom