Forums :: Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives
Support This Site: Contributors :: Advertise


It is currently Mar Sun 29, 2020 3:50 pm


All times are UTC [ DST ]





Post New Topic Post Reply  [ 15 posts ] 
Author Message
 Post subject: SN74LS90N IC Decade Counter - How Can I Make It Work?
PostPosted: Sep Sat 07, 2019 4:45 am 
Member

Joined: Feb Sat 16, 2013 5:08 pm
Posts: 957
Location: Ft. Collins, CO.
I just bought 10 SN74LS90N decade counters for a clock and counter project using 7 segment display tubes. I'm using CD4511BE chips to decode BCD into the 7 segments, and drive the tube. The CD4511BE chips work great, and output 0-9 correctly for the BCD inputs.

The SN74LS90N on the other hand, almost doesn't work at all. The data sheet is here: http://www.ti.com/lit/ds/symlink/sn74ls90.pdf

The chip is supposed to be wired as follows:

CKa > Clock input (Push button to + 5 V, with a 1 k ohm resistor to ground)
CKb > QA
RO(1), RO(2), R9(1), R9(2), GND > Ground (-)
Vcc > + 5 V
QA. QB. QC. QD > Outputs to decoder (High = 1 Low = 0)

This diagram uses the same counter (74LS90) but a different decoder. I don't think that should make a difference, as the inputs are the same 4 BCD.

Image


So, what is happening, is that the outputs have more or less random high pins when first applying voltage. With 5 V on Vcc, I can get 3 volts max out of any Q pin. Connecting and disconnecting CKa (clock input) to Vcc should iterate the counter, but it does nothing. The outputs keep their values.

The ~3 v out is not enough to trigger my decoder. I bumped up the voltage of Vcc to the max allowable, 7 V, before I could get the 4.5 volts out required to operate the decoder. I then discovered that connecting QA to Vcc will iterate the counter a seemingly random number of times, and grounding the input CKa directly (not through a resistor) will also iterate the counter a random number of times.

So, what am I missing? How is this chip supposed to work correctly? Why does grounding the input, and applying Vcc voltage to QA iterate the counter?

I'm hoping I'm just missing something obvious.

_________________
-Eli
http://www.FortCollinsB17.org


Top
 Profile  
 
 Post subject: Re: SN74LS90N IC Decade Counter - How Can I Make It Work?
PostPosted: Sep Sat 07, 2019 5:36 am 
Member
User avatar

Joined: Oct Sun 15, 2006 12:57 pm
Posts: 7065
Location: Liberty, Missouri
1. Mixing logic families is generally not a good idea. A lot of parameters to consider to ensure compatibility. Output levels of 3 to 3-1/2 volts is normal for TTL families, and is somewhat dependent on the load. Find some TTL 74LS47 decoder chips. "LS" drive/input parameters are quite different than standard TTL, and as you have found out C-MOS is far different than any TTL. Then there is MECL which is another entirely different breed.

2. you need a power on reset signal. A cap to VCC in series with a resistor to ground. and tie R1, R2 to the junction of the two. then they will come up all Zeros. I'd think they would show a power on reset circuit in the data sheet, but....

You should look for an old National/TI/Motorola TTL data book they usually had applications notes in the back which were quite instructive.

3. They won't live a long time with 7 volts, Too much dissipation, stick with 5V.

4. In what manner are you generating clock. just putting a wire to VCC? A switch, a push button ??? To work properly the clock has to be clean, with a single positive going leading edge. If not it will quite possibly increment several counts with each "clock".

Logic families of old had very specific parameters that needed to be met if you were going to have good results.

You could use a square wave generator or pulse generator for a clock source, but be sure the signal you apply to the ICs compatible with their requirements.

_________________
Pics, Bold & underlined text, are usually links.

https://mikeharrison.smugmug.com/

Mike


Top
 Profile  
 
 Post subject: Re: SN74LS90N IC Decade Counter - How Can I Make It Work?
PostPosted: Sep Sat 07, 2019 8:10 am 
Member
User avatar

Joined: Sep Mon 16, 2013 2:42 am
Posts: 3763
Location: Tucson, Arizona U.S.A.
74LS logic ICs are not compatible with the 4000 series unless you use a level translator between them. Raising the power supply voltage to 7 volts to get more out of the LS chip also raises the input voltage requirement of the 4000 chip so nothing is gained.

74 TTL series parts, including LS, prefer to be driven with a switch to ground, rather than to supply. You can make it work with a supply side switch if the resistor to ground is low enough resistance. Consult the data sheet to find out the input current at the recommended low input and use Ohm's law.

As mentioned, the clock signal has to be clean. The IC is more than fast enough to count the bounces that the switch contacts make when they close and open. Some switches with mercury wetted contacts are clean enough if you can get one and it is suitable in other respects, like size and mounting position. If you want to use an ordinary switch, a debouncer is required between it and the counter.

Your circuit doesn't show any power supply bypass capacitors. Omitting them can cause all sorts of strange behavior.

_________________
Jim Mueller


Top
 Profile  
 
 Post subject: Re: SN74LS90N IC Decade Counter - How Can I Make It Work?
PostPosted: Sep Sat 07, 2019 1:03 pm 
Member
User avatar

Joined: May Wed 18, 2011 2:40 am
Posts: 4434
Location: Littleton, MA
RCAProduction wrote:
The ~3 v out is not enough to trigger my decoder.

Add 1K Ω pull-up resistors to the LSTTL outputs.

_________________
Steve Byan https://www.byan-roper.org/steve/steve-at-play/


Top
 Profile  
 
 Post subject: Re: SN74LS90N IC Decade Counter - How Can I Make It Work?
PostPosted: Sep Sat 07, 2019 1:57 pm 
Member
User avatar

Joined: Jul Sun 17, 2011 1:11 am
Posts: 4460
Location: Tehachapi, CA - Bottom end of the Sierras
If this is your 1st usage of logic chips, get a copy of Don Lancaster's "TTL Cookbook". Its a great book for hobbyists:

https://www.amazon.com/gp/offer-listing ... ition=used

A couple of things missing. These chips are very sensitive to "glitching".

1) keep traces short. Use heavy wire (18awg+) for +5V & Gnd !

2) Put small 0.1uf ceramic capacitors, across +5V to GND, at EACH IC ! Leads as short as possible.

3) TTL inherently goes hi (logic-1), but very softly, in the noise zone (<2V, >0.8V), and it needs to solidly be pulled either high (>2V) or low (<0.8V). That 1K pull-down resistor ain't gonna do it ! Needs to be dropped to 220 ohms. Better still don't use pull downs on TTL. Only pull ups.

A pull-DOWN (resistor) has one leg tied to GND. A pull-UP (resistor) has one leg tied to +5V.

4) inputs respond to very quick signals. That mechanical switch will be noisy, and generate a bunch of pulses for each depression.

LOOK UP DEBOUNCE CIRCUIT

https://www.google.com/search?source=hp ... CAc&uact=5

In the meantime, hook up a squarewave generator, as a temporary substitute.

5) You CAN mix TTL & CMOS (4000 series), if you put pull-ups (1K) from TTL outputs to CMOS inputs. Not needed to other way around, but watch out for "fan-out" limits, from CMOS outputs to TTL inputs.

I use both TTL & CMOS in this following (AND PROVEN) circuit:

https://threeneurons.files.wordpress.co ... 6c_sch.gif

I also use to sell it as a kit, and still sell the bare boards. Kit has been superceded by a new version.

Also note the debounce circuitry following the time set switches.

FINALLY: CMOS is very sensitive to static. Static, even at levels that you can't feel, can KILL IT ! handle it in a very conductive area. Bare concrete floor, that's damp. Do it in your bare feet. Use an anti-static wrist band, that's tied to ground. Basically all the rules opposite to handling high voltage.

_________________
Life is like a roll of toilet paper - The closer you get the end, the faster it goes - A.E. Neuman


Top
 Profile  
 
 Post subject: Re: SN74LS90N IC Decade Counter - How Can I Make It Work?
PostPosted: Sep Sat 07, 2019 8:18 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 5160
Location: Perrysburg, OH, 43551 U.S.A.
The minimum required high input level foe most "TTL" parts is 2.0 volts. As long as the output connected to that TTL input exceeds that 2 volt level, the input is happy.

stevebyan's suggestion is the one most often given in the application notes for CMOS logic. The downside of this is that more current is needed form the power supply when the TTL output is low.

Instead, why not use the MC14518B? It's a dual BCD up counter with just the basics - each counter has Clock, Enable and Reset inputs and the four BCD outputs. Data sheet here: https://www.onsemi.com/pub/Collateral/MC14518B-D.PDF

John

_________________
“Never attribute to malice that which can be adequately explained by stupidity.”
― R. A. Heinlein


Top
 Profile  
 
 Post subject: Re: SN74LS90N IC Decade Counter - How Can I Make It Work?
PostPosted: Sep Sun 08, 2019 12:53 am 
Member
User avatar

Joined: Sep Mon 16, 2013 2:42 am
Posts: 3763
Location: Tucson, Arizona U.S.A.
A pull--up resistor on each TTL output is a simple level translator and it works but it does cause a higher power supply drain.

In the circuit linked above, pay attention that the debounce circuits use Schmitt trigger gates. Ordinary gates may have a tendency to oscillate in that circuit.

Back to the beginning of this thread, the 7 volt rating of 74 series chips is a survival rating. It only guarantees that the device won't burn out. It does NOT guarantee that it will operate properly at that voltage. The guaranteed operating voltage range of 74 series TTL chips is 4.75 to 5.25 volts. Outside of that range all bets are off; it may or may not work.

Besides a ceramic capacitor at each chip, a 10uF tantalum capacitor where the power enters the board is a good idea.

_________________
Jim Mueller


Top
 Profile  
 
 Post subject: Re: SN74LS90N IC Decade Counter - How Can I Make It Work?
PostPosted: Sep Sun 08, 2019 3:45 am 
Member

Joined: Aug Thu 06, 2015 2:20 am
Posts: 217
I just wanted to comment that TTL uses current sinking to determine a zero input and a voltage input somewhat in excess of 2V to indicate a one state (no current sinking). The suggested 1K pullups will provide the necessary pullup if the decoder is CMOS.

CMOS on the other hand uses voltage levels, not current sinking. Voltages somewhat in excess of 1/2 the supply voltage are interpreted as ones, voltages somewhat less than half the supply voltage are zeros.


Top
 Profile  
 
 Post subject: Re: SN74LS90N IC Decade Counter - How Can I Make It Work?
PostPosted: Sep Sun 08, 2019 5:33 am 
Member

Joined: Feb Sat 16, 2013 5:08 pm
Posts: 957
Location: Ft. Collins, CO.
Update: Pull up resistors worked great. I can now power the counter and decoder off of 5 V. I also do believe that the counter is operating correctly, but each time the push button switch is hit, it is so sensitive it reads multiple counts.

In hindsight, picking more compatible chips would be better, as many have suggested. I'll keep that in mind for future projects, but for this, I'm going to try to use what I have already.

I'd like to trigger the counter with a 555 chip if possible. At first I thought this would be fairly simple, but I can't figure out how to wire the 555 to produce any outputs. I've followed various articles online, all of which leave a constant high or low output (no square waves, or switching). The chip is new, and I haven't overvolted it. I'm working on this on a separate board from the rest of the project, so all of the counter and decoder circuitry is not involved yet.

This seems to be the most consistent circuit described:

Image

I also have some ATTiny chips which I could program with an Arduino to have a 1 Hz output, but I'm trying my luck with a more analog approach. I've always wanted to use a 555 chip, but I've never been able to make them work.

_________________
-Eli
http://www.FortCollinsB17.org


Top
 Profile  
 
 Post subject: Re: SN74LS90N IC Decade Counter - How Can I Make It Work?
PostPosted: Sep Sun 08, 2019 4:37 pm 
Member

Joined: Feb Sat 12, 2011 2:29 pm
Posts: 595
Location: Smithfield, Pa. 15478
Your multiple hits from the switch is called "Bounce". It comes from just that, the contact generates noise the instant the contact closes before solid connection is made. This is resolved by a "debounce" circuit. The easiest way to do that is simply an RS flip-flop made with a pair of nand gates. The switch feeds into those gates. When the first pulse, including the bounce hits the input the flip-flop changes state, sending a clean single pulse high to the counter. When the switch changes state the reverse occurs; the pulse drops cleanly. The counter sees a single high followed by a low. One single pulse for each activation, and a single count. You can sample either output from the flip-flop, in that manner you may obtain either a positive or negative going pulse depending on which you wish to use.


Top
 Profile  
 
 Post subject: Re: SN74LS90N IC Decade Counter - How Can I Make It Work?
PostPosted: Sep Sun 08, 2019 6:53 pm 
Member
User avatar

Joined: May Sat 17, 2014 6:55 am
Posts: 220
Location: Eatonville, WA
For simple debouncing with TTL, I like to use a 7414 Schmitt trigger inverter whose input is held high by an electrolytic capacitor charging through a resistor from +5V. A NO button is connected to the + end of the capacitor and the other end is grounded. Pushing the button discharges the cap and the 7414 detects a level change on the input side and provides a clean TTL output. Releasing the button allows the cap to charge again and the output resets.

_________________
The smoke means it's working.


Top
 Profile  
 
 Post subject: Re: SN74LS90N IC Decade Counter - How Can I Make It Work?
PostPosted: Sep Sun 08, 2019 7:38 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 5160
Location: Perrysburg, OH, 43551 U.S.A.
ForkInSocket wrote:
For simple debouncing with TTL, I like to use a 7414 Schmitt trigger inverter whose input is held high by an electrolytic capacitor charging through a resistor from +5V. A NO button is connected to the + end of the capacitor and the other end is grounded. Pushing the button discharges the cap and the 7414 detects a level change on the input side and provides a clean TTL output. Releasing the button allows the cap to charge again and the output resets.

I've used the CMOS equivalents to do the same thing. The nice thing about those chips is you get six in one package to use as de-bouncers for other switches, simple oscillators where accuracy isn't too important, and even (of all things) and simple inverter!
A couple application notes about Schmitt triggers:
CMOS Schmitt Trigger—A Uniquely Versatile Design Component https://www.onsemi.com/pub/Collateral/AN-140.pdf.pdf (no, the ".pdf.pdf" extension isn't an error).
Understanding Schmitt Triggers http://www.ti.com/lit/an/scea046/scea046.pdf (Look at page 3 for the switch debouncer)
John

_________________
“Never attribute to malice that which can be adequately explained by stupidity.”
― R. A. Heinlein


Top
 Profile  
 
 Post subject: Re: SN74LS90N IC Decade Counter - How Can I Make It Work?
PostPosted: Sep Sun 08, 2019 9:59 pm 
Member

Joined: Jan Tue 16, 2007 7:02 am
Posts: 3454
Location: Lexington, KY USA
A 555 timer makes an excellent switch debouncer. A 556 will make two, or one debouncer and one oscillator.

A microcontroller can be programmed to debounce several switches, provide the timing signal, and the counting, display decoding and drive functions as well. As the hardware guy said, it's almost free; just takes a little software to do everything...

If you do try the microcontroller route sometime, you will find that there are very handy serial-in / parallel-out shift registers that have constant current LED driver outputs to eliminate the series resistors for the LED segments.

Ted


Top
 Profile  
 
 Post subject: Re: SN74LS90N IC Decade Counter - How Can I Make It Work?
PostPosted: Sep Mon 09, 2019 3:18 am 
Member
User avatar

Joined: Sep Mon 16, 2013 2:42 am
Posts: 3763
Location: Tucson, Arizona U.S.A.
Here is a data sheet for the 555: http://www.ti.com/lit/ds/symlink/lm555.pdf. It shows the circuit you are using along with the equations for calculating it's properties. I have used that circuit numerous times and it works. If it doesn't work for you, you are doing something wrong.

Do you have power supply bypass capacitors? 0.1 uF ceramic and 10 uF tantalum are appropriate.

How are you checking that it is working? If you are looking at the LED, perhaps it is on for so short a time that you can't see it. Go through the math on the data sheet to make sure that it is on for at least 0.1 seconds.

The circuit won't work if your pot is set to 0. There has to be significant resistance at that point.

Did you count the pins on the IC correctly? IC data sheets usually show a top view as opposed to tube data sheets that are usually bottom views. If you counted the pins backwards, you may have destroyed the chip by reversing the power supply polarity.

These are what comes to mind immediately; there could be other issues.

If you intend to use this to trigger your counter, be sure to operate it from the same 5 volt supply as the rest of they circuit. If you run it from 9 volts, you could damage the counter IC by overvoltaging the input. You should probably also remove the LED and it's resistor to avoid loading the output.

_________________
Jim Mueller


Top
 Profile  
 
 Post subject: Re: SN74LS90N IC Decade Counter - How Can I Make It Work?
PostPosted: Sep Mon 09, 2019 7:58 am 
Member

Joined: Jan Tue 16, 2007 7:02 am
Posts: 3454
Location: Lexington, KY USA
It might be better to move the LED and resistor to connect to Vcc and light the LED when the output is low. The 555 output's pull down is much stronger than the pull up. It's pretty much a slow but beefy TTL output stage.

Image

Ted


Top
 Profile  
 
Post New Topic Post Reply  [ 15 posts ] 

All times are UTC [ DST ]


Who is online

Users browsing this forum: No registered users and 7 guests



Search for:
Jump to:  
































-->


Privacy Policy :: Powered by phpBB