Searching \ for 'Interfacing to vacuum fluorescent displays' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: techref.massmind.org/techref/index.htm?key=interfacing+vacuum
Search entire site for: 'Interfacing to vacuum fluorescent displays'.

Truncated match.
PICList Thread
'Interfacing to vacuum fluorescent displays'
1997\03\12@015251 by Prashant Bhandary

flavicon
picon face
I've just obtained a few vacuum fluorescent displays(VFDs) from a surplus
dealer and am looking at interfacing them to the PIC. The part no. is
Futaba 4-LT-105Z. It is a timer(88:88) display which seems to have been
made for an oven. What's the best way of driving them from a PIC 16C84?
I haven't used VFDs for a while now and any pointers to resources on VFDs
in general would be great.

Regards

Prashant
--------------------------------+---------------------------------
 Prashant Bhandary             | Tel:  +61-2-9662 5299
 Spatial Information Solutions | Fax:  +61-2-9662 5348
 Roads and Traffic Authority   | Email: spam_OUTprashbTakeThisOuTspamrta.nsw.gov.au
 Rosebery NSW 2018, AUSTRALIA  | "2b|!2b" - William Shakespeare
--------------------------------+---------------------------------

1997\03\12@082223 by Kalle Pihlajasaari

flavicon
face
Hi Prashant,

> I've just obtained a few vacuum fluorescent displays(VFDs) from a surplus
> dealer and am looking at interfacing them to the PIC. The part no. is
> Futaba 4-LT-105Z. It is a timer(88:88) display which seems to have been
> made for an oven. What's the best way of driving them from a PIC 16C84?
> I haven't used VFDs for a while now and any pointers to resources on VFDs
> in general would be great.

You will have a fillament that will draw significant current 40 .. 120 mA
would be a likely figure.

The fillament is at the lowest potential, preferably the midpoint is at
ground and the two ends have an AC drive voltage byt this is not common
these days as it would mean an extra transformer winding.  So instead one
puts one end to ground and the other end to a DC voltage 2 .. 6 V typically
either positive or negative.  The fillaments should heat up enough to just
be visible in total darkness with no other signals present.  They take about
1 .. 3 seconds to heat and about 0.5 seconds to burn out if you over voltage
them too much.

The drive is now referenced to the 'ground' potential and consists of similar
segment and digit drivers.  They are made to work at 0 .. 12V positive these
days while in the past the voltages were higher.

You need to have some form of intensity balancing in long displays if you have
a DC fillament voltage as the drive voltage varies from 12V at one end to
7V at the other end and this can be achieved in two ways at least.  One is to
have the digit grids driven through a resistor of magnitude such that the
digit intensities are all the same as the dimmest digit at the lowest voltage
end of the filament.  Once you drive the display with a PIC you have another
option and that is to duty cycle modulate the digit on times, make the dimmest
digit selected for longer.  The relationship should be close to linear across
the display.

The drive of the segments is set the segments you want lit ti 12V and then
set the digit grid to 12V for the digit in question with the other digit
and segment lines at 0V.

Refresh at faster than the visible flicker rate, the phosphor supplies
extra persistence so you can probably get away with anything over 10 Hz
but 30 Hz will be flicker free.

The drivers on a Futaba type LT display I have are NE594 and have 8 High side
drivers with TTL inputs.  With your 5 digit display you would need 5 + 8
drivers if you want to use the decimal points and the colon, just the
numbers would need 4 + 7 so you will need 2 ICs 8 or 7 drivers per
IC would be enough.  You could suffer the high side drivers with 2 transistors
each but a lot more work.

Beacuse of the persistance you can also go do other things for a short while
in the PIC code without flickering of the display.

Cheers
--
Kalle Pihlajasaari   .....kalleKILLspamspam@spam@ip.co.za   http://www.ip.co.za/ip
Interface Products   P O Box 15775, DOORNFONTEIN, 2028, South Africa
+ 27 (11) 402-7750   Fax: 402-7751    http://www.ip.co.za/people/kalle

DonTronics, Silicon Studio and Wirz Electronics uP Product Dealer

1997\03\13@215623 by Prashant Bhandary

flavicon
picon face
At 03:10 PM 12/03/97 +0200, you wrote:
>The drive is now referenced to the 'ground' potential and consists of similar
>segment and digit drivers.  They are made to work at 0 .. 12V positive these
>days while in the past the voltages were higher.

I hope it is a 12V display. Anything higher would be a pain to use.

>
>You need to have some form of intensity balancing in long displays if you have
>a DC fillament voltage as the drive voltage varies from 12V at one end to

It may be worth driving the filament through some kind of an H bridge.

>The drivers on a Futaba type LT display I have are NE594 and have 8 High side
>drivers with TTL inputs.  With your 5 digit display you would need 5 + 8
>drivers if you want to use the decimal points and the colon, just the
>numbers would need 4 + 7 so you will need 2 ICs 8 or 7 drivers per
>IC would be enough.  You could suffer the high side drivers with 2 transistors
>each but a lot more work.

With 5 + 8 drivers, I would be using up all the PIC 16C84 I/O lines. A
1-of-n decoder for the digit drivers will save 2 lines.

How much current does the grid and anode take? An alternative to high side
drivers could be using open collector TTL gates with pull-up resistors. A
7405, depending on the max voltage it can take, can be used to drive 6 lines.
That would mean 3 of these chips. Other possibilities are an LM339(4 lines,
4 chips) or CMOS gates operating off a 12V supply with a 0-5V signal on
their inputs. The decoder could be an extra chip unless I find an open
collector decoder like the 7441 or a CMOS decoder. I wonder what the input
characteristics of CMOS gates operating on a 12V supply are. If it is
suitable, I could use 2 hex inverters and one decoder(3 chips). I might even
be able to cobble together some kind of an H bridge for the filament with the
spare gates.

Regards

Prashant
--------------------------------+---------------------------------
 Prashant Bhandary             | Tel:  +61-2-9662 5299
 Spatial Information Solutions | Fax:  +61-2-9662 5348
 Roads and Traffic Authority   | Email: prashbspamKILLspamrta.nsw.gov.au
 Rosebery NSW 2018, AUSTRALIA  | "2b|!2b" - William Shakespeare
--------------------------------+---------------------------------

1997\03\14@111133 by Kalle Pihlajasaari

flavicon
face
Hi Prashant,

> >The drive is now referenced to the 'ground' potential and consists of similar
> >segment and digit drivers.  They are made to work at 0 .. 12V positive these
> >days while in the past the voltages were higher.
>
> I hope it is a 12V display. Anything higher would be a pain to use.

The FUTABA displays you have will most likely work fine with 12V.

> >You need to have some form of intensity balancing in long displays if you
have
> >a DC fillament voltage as the drive voltage varies from 12V at one end to
>
> It may be worth driving the filament through some kind of an H bridge.

H-bridge won't be much help except to swap the polarity and not much need
of that on a small display.  What would work nicely is to have a regulator
that supplies 5V across the fillament and then has a floating ground
that you can adjust to a value from 0V to 5V, you could then choose
the fillament reference at 4.5V and make sure the fillament is at
that voltage infront of the digit you are displaying by adjusting the
'ground' reference.  Digits at the higher voltage end of the filament
will be close to 4.5 V if low end is near ground and digits at the
low voltage end will be near 4.5V if you raise the ground to 4V and the
other end will be 9V.  This would give you an even digit brightness
over the whole display with equal multiplex time.  You must not go below
0V or else the Anode/Grid voltahe will be more positive and turn segments
on weakly.

The multiplex duration intensity control is something I thought of, I have
not seen it used and the characteristics of the Phosphor may preclude
the use of it as brightness may be a function of voltage only and
not time, I may try it out one day.  The above sliding ground
method should work but means that you need a analog signal for the
fillament supply regulator ground reference.

> >The drivers on a Futaba type LT display I have are NE594 and have 8 High side
> >drivers with TTL inputs.  With your 5 digit display you would need 5 + 8
> >drivers if you want to use the decimal points and the colon, just the
> >numbers would need 4 + 7 so you will need 2 ICs 8 or 7 drivers per
> >IC would be enough.  You could suffer the high side drivers with 2
transistors
> >each but a lot more work.
>
> With 5 + 8 drivers, I would be using up all the PIC 16C84 I/O lines. A
> 1-of-n decoder for the digit drivers will save 2 lines.

If you leave the colon permanently on or off you save one line.  If you
decide to use a BCD-7seg decoder you save 3 more lines, if you ignore
the decimal points you save one more so you can get away with 2 + 4
on the PIC, you still need 4 + 7 for the display.

> How much current does the grid and anode take? An alternative to high side

Not sure, at a guess Grid 0.1 mA and all the Anodes together say 1.0 mA.

{Quote hidden}

You cannot use 5V levels to switch CMOS inputs running at 12V, the threshold
is near 6V so you would need to use a level shifter.

SPRAGUE and other no doubt have decoder driver IC's part number
UCN4805A might work, it sinks current so I'm not sure if the polarity is
correct but it has a latching 4-7 line decoder and an extra driver for
the decimal point and can work upto 60V.  Polarity is the only concern
and you will probably need pullup resistors.  A UCN4807A is similar
but has a 3-8 line decoder instead.  With these two ICs you could
probably interface in minimum mode with 4 + 2 lines or in maximum
mode with 5 + 3 lines.  I don't have datasheets so cannot verify that
they will work...... Having a better read of the section, they probably
won't work as VF displays are not mentioned but LED & filaments are.
The polarity of all the segments would be wrong :-(

Sprague do also have plain drivers, decoders required extra, with
6 and 8 gates per IC with numbers in the range UDN6116 UDN6118 UDN6138
The UDN6118 is probably the same function as the NE594.

Cheers
--
Kalle Pihlajasaari   .....kalleKILLspamspam.....ip.co.za   http://www.ip.co.za/ip
Interface Products   P O Box 15775, DOORNFONTEIN, 2028, South Africa
+ 27 (11) 402-7750   Fax: 402-7751    http://www.ip.co.za/people/kalle

DonTronics, Silicon Studio and Wirz Electronics uP Product Dealer

More... (looser matching)
- Last day of these posts
- In 1997 , 1998 only
- Today
- New search...