Breaking News
Blog

GPS-Driven, FPGA-Decoded Nixie Tube Speedometer, Part 2

NO RATINGS
Page 1 / 3 Next >
View Comments: Newest First | Oldest First | Threaded View
<<   <   Page 2 / 4   >   >>
Stargzer
User Rank
Author
Re: VHDL and Lookup Table
Stargzer   1/23/2014 2:24:50 PM
NO RATINGS
@bobdvb:  "...but a lookup table for the speed? Was that really the most efficient mechanism?"

"Somewhere in the basement" I have an old book on "Assembly Level Programming for Small Computers" (back when a "small" computer would have been a TTL mini with 4K or so of memory).  It has routines for using lookup tables to do multiplication or trig functions for systems that didn't have multiply or divide instructions or were just plain too slow.  Of course, the 50MHz clock on that FPGA should be fast enough for math, but sometimes a lookup takes a lot less cycles than carrying out the calculation.  The blinding speed and surfeit of memory available these days has for the most part done away with the need for really tight code, but if you're on a budget ... . 

Max The Magnificent
User Rank
Author
Re: Call me...
Max The Magnificent   1/23/2014 2:18:55 PM
NO RATINGS
@David: Like you I find the old displays very pleasing - much nicer than the angular 7-segment LEDs that are everywhere these days.

I think everyone finds them pleasing -- if you have something like that Nixie-Tube clock we were talking about last year, for example, engineers just stand and drool stare :-)

Max The Magnificent
User Rank
Author
Re: Call me...
Max The Magnificent   1/23/2014 2:16:06 PM
NO RATINGS
@zewde: It's worth holding on to old defunct parts. You never know what you'll be able to go back and integrate with...

I could kick myself for all the stuff I've let slip through my fingers.

Stargzer
User Rank
Author
Nixies
Stargzer   1/23/2014 1:48:08 PM
NO RATINGS
Back in Ancient Times I remember a piece of telecomm troubleshooting gear that used Nixies instead of 7-segment displays.  They advertised that when there were quick changes in the low-order digit you could see the change on the Nixie as two or more digits were displayed in the same tube, but on a 7-segment you might only see an "8".  I think it was call the Range Rider, doing Bit Error Rate tests and so on.

Some surplus place used to sell a tube with multiple segments and they included a circuit to display segments at random to create random patterns, like a random blinking light circuit.

 

zewde yeraswork
User Rank
Author
Re: Call me...
zewde yeraswork   12/12/2013 9:08:18 AM
NO RATINGS
It's worth holding on to old defunct parts. You never know what you'll be able to go back and integrate with even, in this case, five decades later attaching a speedometer to an old vehicle.

David Ashton
User Rank
Author
Re: Call me...
David Ashton   12/6/2013 6:21:33 PM
NO RATINGS
@antedeluvian...They had some funny old displays in those days.  My dad used to work for Burroughs and I got some out of some defunct desk calculators.  They were made by Itron.  Alas, I foolishly got rid of them when I left Zimbabwe :-(

I remember reading about integrated injection logic a long time ago....like ECL it could not compete with the behemoth of TTL.   Another case of good technology getting sidelined by a more popular but inferior technology.

antedeluvian
User Rank
Author
Re: Call me...
antedeluvian   12/6/2013 9:53:38 AM
NO RATINGS
David

they had 10 segments and you could make most letters with them as well as numbers.  

I remember doing some work with those "star burst" VF displays. The ones I used had 18 segments including the decimal point and the second point to make up a colon. I was trying to make a cash register for the dry-cleaning industry. I think I still have a display sitting in a drawer. To drive it (at least to decode from binary the the display format) I used a brand new chip from TI- the AC5947N (I probably have a couple of those in my drawer as well) made in a brand new technology (I2L, that is I squared L for Integrated Injection Logic). Neither the chip nor the technology lasted in the market and I had to look for alternatives. But I'm not bitter.

David Ashton
User Rank
Author
Re: Call me...
David Ashton   12/6/2013 4:19:20 AM
NO RATINGS
Thanks Victor.  I used to have a bunch of those old green VFD displays - they had 10 segments and you could make most letters with them as well as numbers.  You find them a lot in old video recordersas well - as you say they are multi-digit in one glass envelope - but they are usually special purpose ones and difficult to use.

Like you I find the old displays very pleasing - much nicer than the angular 7-segment LEDs that are everywhere these days.

Victor Lorenzo
User Rank
Author
Re: Call me...
Victor Lorenzo   12/6/2013 2:46:57 AM
NO RATINGS
Hi @David, "but I'm being a bit pedantic"

In no way you're being pedantic. Thanks for making me note that.

David Ashton
User Rank
Author
Re: Call me...
David Ashton   12/5/2013 6:48:00 PM
NO RATINGS
@victor... "I specially like those green nixies from the old casio calculators"

They're actually vacuum fluorescent displays (VFDs), not Nixies as such, but I'm being a bit pedantic.  They are very nice.  I was lucky enogh to get a few huge (like 9 by 5 inch) 16 char x 4 line dot matrix VFDs not long ago and am working on driving one with a PICAXE soon.

I also have some Sperry 7-segment displays which work like Nixies (neon gas discharge) but with segments instead of the complete  numbers that Nixies have.

There are certainly some tasty displays out there.....

<<   <   Page 2 / 4   >   >>
Most Recent Comments
michigan0
 
SteveHarris0
 
realjjj
 
SteveHarris0
 
SteveHarris0
 
VicVat
 
Les_Slater
 
SSDWEM
 
witeken
Most Recent Messages
9/25/2016
4:48:30 PM
michigan0 Sang Kim First, 28nm bulk is in volume manufacturing for several years by the major semiconductor companies but not 28nm FDSOI today yet. Why not? Simply because unlike 28nm bulk the LDD(Lightly Doped Drain) to minimize hot carrier generation can't be implemented in 28nm FDSOI. Furthermore, hot carrier reliability becomes worse with scaling, That is the major reason why 28nm FDSOI is not manufacturable today and will not be. Second, how can you suppress the leakage currents from such ultra short 7nm due to the short channel effects? How thin SOI thickness is required to prevent punch-through of un-dopped 7nm FDSOI? Possibly less than 4nm. Depositing such an ultra thin film less then 4nm filum uniformly and reliably over 12" wafers at the manufacturing line is extremely difficult or not even manufacturable. If not manufacturable, the 7nm FDSOI debate is over!Third, what happens when hot carriers are generated near the drain at normal operation of 7nm FDSOI? Electrons go to the positively biased drain with no harm but where the holes to go? The holes can't go to the substrate because of the thin BOX layer. Some holes may become trapped at the BOX layer causing Vt shift. However, the vast majority of holes drift through the the un-dopped SOI channel toward the N+Source,...

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)
Like Us on Facebook
EE Times on Twitter
EE Times Twitter Feed