Burned K155ID1 chips?

[Imbanon]

Hey all

Got a bit of a problem over here.
Since I've got my atmegas, I've started repairing my breadboarded
circuit. Protected my fresh chip, checked for any possible shorts, and
fired the damned thing. The results were odd. Some of the nixies lit
up showing number 7, instead of the real time. I suspected something's
wrong with the K155ID1, so I took it out and breadboarded it
separately to test it out. Not to mention the enigma of all the nixies
lit up, when only one was supposed to..
The results weren't as expected. Giving it the certain inputs, outputs
were not right. I didn't test any further.
I figured I should try 5 other chips, of which in the end, only two
were working properly.
Suddenly, one stopped working right. Stupidly, I tried the other
working one, which then worked for some time, but after a few
switchings on and off, it was the last one to act faulty.

Then it hit me. I noticed I forgot the limiting resistor on the Vcc!
Then I tested again with the limiting resistor of 300 ohm on 5.15V,
giving proper 17ma.
These were the results; input 0001 gives 5 instead of 1, and 0011
output 7 instead of 3, etc.

So what can I do now? Is it possible to fry a demux in such way, or is
there still hope for these babies?
I'll never forgive myself for doing such a dumb thing..

Thanks for your help
Imbanon

[John Rehwinkel]

> Then it hit me. I noticed I forgot the limiting resistor on the Vcc!
> Then I tested again with the limiting resistor of 300 ohm on 5.15V,
> giving proper 17ma.

TTL chips like this really want to run from a regulated power supply, instead of adjusting a resistor to produce about the right voltage drop. TTL can draw widely varying amounts of current as it operates, so even if you dialed in the resistor to produce the correct voltage at one time, it could still be wrong other times.

> So what can I do now? Is it possible to fry a demux in such way, or is
> there still hope for these babies?

It is quite possible to cook these chips, but it's also a random thing. I've massively abused some chips (melted one into the shape of a banana) and had them continue to work properly. Others I've abused would stop working, then work again a few minutes or days later. And some, with just mild abuse, stopped working and stayed that way.

These K155ID1 chips are fairly robust, but either their quality control was iffy, or there are a lot of damaged chips on the market, as I've had several that weren't right when I got them, and I've had a few more fail with just a little use. One of these apparently repaired itself after about a week, and has been working fine for years now.

> I'll never forgive myself for doing such a dumb thing..

It's worth picking up an ordinary 7805 style integrated voltage regulator and a few filter capacitors to build a nice stiffly regulated power supply with overcurrent and thermal protection. I generally use a large (2200-4700µf) filter capacitor, another 1µf capacitor right at the input terminal, a second 1µf unit right at the output terminal and a 0µ1f capacitor at the power supply pin on each chip I'm powering. It's kind of a shotgun approach, but it's not difficult or expensive, and my chips seem happy (from that standpoint at least - I always seem to find some other way to make their lives difficult).

Alternatively, if you can find a 5 volt regulated power supply (like a phone charger or USB hub supply or similar), those are dandy for powering logic. I'd still suggest a small capacitor for each chip - especially necessary since the regulator is now at the end of a long cord.

- John

[Mich...@aol.com]

Does 0101 give a 5 as well?

 

Was wondering if 0100 is stuck.

 

Does it count......

4 5 6 7 4 5 6 7 8 9 4

instead of

0 1 2 3 4 5 6 7 8 9 0

 

Michail

[dr pepper]

74 ttl as mentioned really needs a 5v regulated supply, the 7805 is
important.
Also a ceramic decoupling capacitor accross each chip supply pins is a
good idea, I've had similar issues with 74 devices, when the
transistors change state within the device there is a 'rush through',
sort of a temporary short across the 5v, this can deck the supply at
the chip if there is resistance or incductance in the supply to it,
which there allways is and cause a reset or screwup.

It does sound like the chips are fried though.

On 10 Feb, 04:56, Micha...@aol.com wrote:
> Does 0101 give a 5 as well?
>
> Was wondering if 0100 is stuck.
>
> Does it count......
> 4 5 6 7 4 5 6 7 8 9 4
> instead of
> 0 1 2 3 4 5 6 7 8 9 0
>
> Michail
>
> In a message dated 2/9/2012 7:42:23 P.M. Pacific Standard Time,
>

[Tidak Ada]

If you use a 7805 as voltage stabilizer, consider these IC's need an
overvoltage of at least 3V at the input. The show an voltage drop of that
3V. Usually a transformer with an output of 8 to 9 volts AC is advised (This
results in a raw DS of 1.4 × 8 = 11.2 V. So you are safe for line and load
dips. Adequate cooling of the 7800 is required !

eric

--
You received this message because you are subscribed to the Google Groups
"neonixie-l" group.
To post to this group, send an email to neoni...@googlegroups.com.
To unsubscribe from this group, send email to
neonixie-l+...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/neonixie-l?hl=en-GB.

[Cobra007]

I would recommend to use a switch mode 5V regulator, no need to worry
about dissipated heat.

For low current power supplies I still use zener diodes, works very
well if certain applications and no need to worry about voltage drops.

Judging from your data (17mA & 300ohm), your power supply is then
about 10V. A series resistor to go from 10V down to 5V TTL really is
an unusual solution. You can never guarantee that the current will be
17mA in all situations, so the voltage may vary between 3 and 8 for
example, frying your chips in the mean time.

Michel

[Imbanon]

My supply was an old computer supply, giving me 3.3 5.15 and 12 volts,
also all these only in negative "outputs".
I did what you said and set up a 7805. It gave me a steady 5.04 volts.

I did full testings of the chip. I found out that the first two bits
are stuck on 01. So instead of 0000 I get 0100.
This means I only got outputs 4, 5, 6, 7, and so on repeating itself
(instead of 0 1 2 3 4 5 6 7 8 9 its 4, 5, 6, 7, 4, 5, 6, 7, 4, 5,...
even for the last 6 outputs that were supposed to all give 0)

Any way to unstuck these, or are they done?

Thanks people

[Mich...@aol.com]

So, it sounds exactly like what I said earlier, although you have 2 stuck bits.

 

I ran into this once, but it was my bad solder job.   I would double check that first, and if it doesn't fix then simply replace the chip.  The cost of the chip is cheap compare to the time you have spent trying to repair.

 

I don't believe the chip can be repaired anyway.  :(

 

Michail

[dr pepper]

If your using an old pc supply, that in itself could be the issue.
Pc supplied are switch mode and require a minimum load, if you dont
pull enough current on the 5v line then there will eb issues, drop
out, switching noise etc.
You need a 'swamp' resistor on the 5v, to dump about an amp, I've used
an old car headlight bulb on the odd ocaision just as a temporary
measure.

On 10 Feb, 16:32, Micha...@aol.com wrote:
> So, it sounds exactly like what I said earlier, although you have 2 stuck
> bits.
>
> I ran into this once, but it was my bad solder job.   I would  double check
> that first, and if it doesn't fix then simply replace the  chip.  The cost
> of the chip is cheap compare to the time you have spent  trying to repair.
>
> I don't believe the chip can be repaired anyway.  :(
>
> Michail
>
> In a message dated 2/10/2012 8:20:30 A.M. Pacific Standard Time,
>

[Imbanon]

The good news is that one of the chips isn't fried, and is working
well on the current setup.
And I love the car headlight idea! It would be a pretty crazy sight on
a work table :)

[Adam Jacobs]

I've definitely done that, the volkswagon headlights work best, the 6v
kind. Or sometimes you can find a really big rheostat. These kinds of
things are great for when you want to drain a battery or put a lot of
load on a circuit.

-Adam

[Lucky]

I mod quite a few of the supplies and always use a 10 ohm 10W Wire-
wound resistor across the 5v line such as these, ebay listing
400275651219. Find the supplies 12v tend to run rock solid loaded or
unloaded. I normally cable tie one to the metal work inside (through
the vents) to aid cooling the resistor. Worth trying, not as
impressive as the headlight but a lot smaller ;-)

Dave

[Imbanon]

Sorry for bumping an old thread, just want to post a little epilogue.

The chips turned out to be just fine. Do NOT resistor limit their
input lines according to datasheets. Same goes for the supply.
Mine were already "burned", so I figured I don't have anything to lose
if I connect them without resistors. And they now work like a charm.
The first one that I used when something smoked on the breadboard
still doesn't seem to work, so I guess the mystery is solved.
Again, thank you all for help!

[Mich...@aol.com]

It sucks when you let the smoke out of anything.

 

When I started playing around with ICs, I had let the smoke out of several.  :(

 

Michail

[Instrument Resources of America]

And let me tell you!!!!!!   It's a real bitch trying to catch all of those teeny tiny particles of smoke and put them back where they came from.!!!  I've never been able to do it, and I've tried.   Ira.

--