Making an SPI VFD work with PICAXE

Buzby

Senior Member
I would say that this is normal. Having spent many years repairing video recorders which often used this type of display, being able to see a faint orange glow (often only visible in near darkness, as you describe) from the heaters was a good sign! .....


If that does not help, it could be inter-electrode leakage in the actual VFD itself.
Hi mkstevo,

Thanks for the input.

My VFDs are much smaller than nick's , and the heaters don't glow, but I'm happy that you can confirm that the glow is normal for bigger ones.

Nick has already tried replacing the electrolytics.

I don't understand how inter-electrode leakage could occur inside the glass, can you explain ?.

If you look at the pictures in post #1 you can see the graduated ghosting.
My suspicion was either too high HT, thus causing grid punch-through, or too high heater current causing an electron cloud to form.

Another fault might be the zener, mentioned in section 6.2 of this useful webpage : http://www.noritake-elec.com/vfd_operation.html

Whatever it is, I don't think it's going to take much to fix.

Cheers,

Buzby
 

nick12ab

Senior Member
Hi Nick,

If only two effective pins, it's probable that the transformer has a centre tap.

Can you trace the heater pins back to the transformer ?
Then measure resistance again, this time to find 1 or more transformer pins which have low ohms to the known heater pins.
It could be a centre tapped coil, with just 3 pins. Or it could be two series connected coils, with the tap at the joint. In this case using 4 pins.
Resistance between the two heater terminals is 0.6 ohms.

The transformer has 10 pins, like an IC, with pin 1 indicated by a dot. Pin 8 is connected to the right end of the heater and pin 10 to the left end of the heater. The resistance between pins 8 + 9 and 9 + 10 is 0.3 ohms in both cases. No other pins have a low resistance between these pins.

Then, with power on, what is the voltage across the heater ?.
( This will be AC at way higher than 50Hz, so it's not a very accurate measurement. )
A cheap multimeter claims 0.6V on the 200VAC range but then just shows a '1' at the left of the display (overload indicator) when changed to the 20VAC range.
 

Buzby

Senior Member
Hi Nick,

It looks like transformer pin 9 is the centre tap.

From what mkstevo said, I don't think we have an overdriven heater.

It might be the zener mentioned in the webpage I posted in #4.
If this, or the capacitor across it, was bad it might give those background speckles you can see.

Time to trace the centre tap !.

Cheers,

Buzby
 
M

mkstevo

Guest
Inter-electrode leakage could be caused by either physical damage to the electrodes inside the VFD or due to contamination of the electrodes caused by invisible particles of stray materials, attracted by the voltage potentials has been deposited on the electrodes (in a similar way to putting a magnet in your pocket might become covered in metallic particles).

As the phosphors that glow, some of this material can become trapped between the electrodes which in order to work at the relatively low (for a vacuum tube) voltage are very close together. By the same token, if the components that make up the display have not been cleaned completely before being assembled they can give off contaminants which could compromise the integrity of the internal vacuum and cause electrical leakage to occur between the electrodes.

The impedance of the grids, anodes and cathodes is very high in a vacuum tube, in the order of millions of ohms and the slightest resistive path (even a build up of gas or slight loss of 'air' vacuum) between them will cause it to malfunction.

I hope that none of these is the cause of the problem, as this can't be fixed, I just offer it as a possibility.
 

nick12ab

Senior Member
The zener is a small three-pin surface mount device that looks like a transistor. I will investigate it some more tomorrow.
 

nick12ab

Senior Member
The zener (which is marked 682), when in circuit, has a resistance of 4 ohms no matter which way it is measured. The side of the zener diode with only one pin is connected to pin 9 of the transformer and the other side to VFL on the chip.
 

Buzby

Senior Member
This might be it !.

Either the zener, or maybe a capacitor across it, looks to be short circuit.

I've no idea what voltage it should be, but it can't be very high, maybe 6.8v ?


Can you try replacing it ?.
 

Buzby

Senior Member
I'd try the 5.6v first.

It won't blow up.
Too low a zener voltage will give the symptoms you have now.
Too high and it will be a very dim display, but definitely no ghosting !.

When you take the old one off, just check the PCB traces again, make sure there is not still a low resistance where the zener was.
It could still be something in parallel.
 

nick12ab

Senior Member
I'd try the 5.6v first.

It won't blow up.
Too low a zener voltage will give the symptoms you have now.
Too high and it will be a very dim display, but definitely no ghosting !.

When you take the old one off, just check the PCB traces again, make sure there is not still a low resistance where the zener was.
It could still be something in parallel.
The zener was the cause of the low resistance.

Well it turns out I only have 5.1V ones - don't know why I thought I had 5.6 ones. I replaced the zener and now there's only a very slight dim glowing at the edges when nothing is on the display and slightly less ghosting than ther was before when there's stuff on the display.

So can I stick a normal diode with a 0.7V drop in series with the zener (but in reverse bias) to reduce the glow further?
 

nick12ab

Senior Member
So can I stick a normal diode with a 0.7V drop in series with the zener (but in reverse bias) to reduce the glow further?
Yes I can! Now the ghosting's all gone but the display is now quite dim compared to how it wa sbefore. Maybe I need to use a skottky diode instead to get the right brightness?
 

nick12ab

Senior Member
I made it brighter by putting three 5.1V zeners and one 12V zener in series (total 27.3V) and now there's only a very slight ghosting. So the voltage of the original zener would have been something like 30V, not 6.82V.

Thank you Buzby for all your help. Thank you mkstevo for your contributions too. Hippy, thanks for the suggestion of using a gosub to reverse the bit order.
 

Buzby

Senior Member
So is it looking like you are on the right track.

Just need to find the right zener voltage.
 

nick12ab

Senior Member
I've stuck in an extra 5.1V zener bringing the total up to 32.4 and there's now only a tiny bit of ghosting on the rightmost character only and visible only when the display is shielded from external light.

And some improved code:
Code:
#picaxe 28x2
#no_table
symbol cs = C.2
symbol dataoutpin = pinC.5
symbol vfddata = b4

main:
	high cs
	hspisetup spimode00,spimedium
	low cs
	
	vfddata = %00010000 : gosub sendcommand
	for b0 = 0 to 15
		lookup b0,("     PICAXE     "),vfddata	'Initial message goes here
		gosub senddata
	next
		
	vfddata = %01100000 : gosub sendcommand
	vfddata = %01010111 : gosub sendcommand
	vfddata = %00110000 : gosub sendcommand
	vfddata = %00000000
	for b0 = 0 to 15
		gosub senddata
	next
	vfddata = %01110000 : gosub sendcommand
	
	
	'more stuff can go here
	
	
	do : loop
		
sendcommand:
	pulsout cs,1
senddata:
	vfddata = vfddata rev 8
	hspiout (vfddata)
	return
 

Willie...

New Member
This is an interesting discussion! :) I have a couple of Noritake dot matrix VFD's that I have tinkered with, but couldn't get working right. Their website has some very useful info, even referring to the "ghosting" you guys are talking about. I can't imagine that the PICAXE has anything to do with that, tho, unless you're using it to DIRECTLY communicate with the display (via transistors for the high voltage parts) because normally, they have a dedicated controller chip that "talks" to the VFD elements. Read the info on their site, "Inter Digit Blanking" section 4.2 to see if you can do anything with that. https://www.noritake-elec.com/technology/general-technical-information/vfd-operation
 

nick12ab

Senior Member
Top