Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Problem with 74LS90.. some help please??

Status
Not open for further replies.

patroclus

New Member
Hello to everyone. I need some help and I don't find it anywhere.

I've using some 7490 IC for counting purposes, with a simple 555 circuit for clock pulses. Everything was fine.
Yesterday I had to replace one of the 7490 counter, so I used a 74LS90 instead, because I had no other. Surprise!! the counter didn't work!! and I changed no other thing.. I'm really confused..

Where the 7490 worked, this new 74LS90 doesn't... I tried everything in my breadboard but don't have any idea why this happends. The counter does not count correctly, and changes state not only in 1 to 0 clock transition, but also in 0 to 1. Counting seems random..

I tried 3 74LS90, and 3 7490; all 7490 worked, none of the 74LS90 did. I bought each one separately. Maybe the 74LS90's are defective??

I REALLY aprecite some help.
Greetings.
 

ChrisP

Member
Of course, the 74LS90 chip is designed for maximum current levels approximately half of those used with the 7490, but the primary difference -- the one probably causing your troubles -- is the minimum pulse widths on the resets. The 7490 calls for a minimum 15ns pulse width, while the 74LS90 requires 30ns pulses for resetting. Here is a TI datasheet that covers both chips...
 

k7elp60

Active Member
Sounds like possible power supply problems. Do you have a electrolytic bypass capacitor from Vcc and ground on the breadboard? Do you have
a 0.1uf at the same points? If not try a 22 uf electrolytic capacitor and a
0.1 ceramic in parallel from Vcc to ground on the proto board
 

patroclus

New Member
Thanks for answering.

ChrisP, I don't thnik it is a reset problem, as I ground both reset inputs and don't use them now, and the pulses coming from the 555 are about 1 Hz.

k7elp60, I didn't try to use a bypass capacitator.. the power supply is +5V stabilized. I'll try adding the 22uF capacitator between V+ and GND, but.. why use a 0.1 ceramic capacitator in parallel?? Wouldn't the 22uF be enough??
 

Phasor

Member
I'll try adding the 22uF capacitator between V+ and GND, but.. why use a 0.1 ceramic capacitator in parallel?? Wouldn't the 22uF be enough??

Electrolytic capacitors have a higher internal resistance than other types - this can cause 'slow' response when working at high frequencies.

The 0.1u ceramic has a lower internal resistance, and is better for 'catching' the high frequency transients.
 

k7elp60

Active Member
I have built a lot of circuits with the 555 timer with TTL circuits,as often as I forget the electrolytic bypass on the proto board I have intermittent problems. I generally use a 100uf because my power supplies have be adjustable ones and they can stand the 100uf on the output lines. Some
fixed power supply's can have problems with a large value of electrolytic
capacitor across it output lines, thats why I recommended a 22uf. I would put it as close to the 555 pins 1 and 8 as possible.

TTL is not only sensitive to supply and ground-line noise, but it also generates a lot of its own noise when any totem-pled output structure changes state and drqws heavy current spike from the supply lines. These narrow spikes must be kept from going through the supply system.
These despikeing capacitors with the shortest leads are recommended.

Don Lancaster, in his TTL cookbook recommends not only the despiking capacitors, but a 10uf 6V tantalum capacitor across the +5 lines where they leave the board
Ned
 

patroclus

New Member
do I have to use a bypass capacitor before every IC??
Or just at the input, where power supply??

I'm going to use a LM7805, so I though on using a capacitor between its common and input pins, and maybe, another one after it, between Output and common pins... I'm I doing well?? Is still necesary a bypass capacitor before the other IC??

Thanks
 

k7elp60

Active Member
The only time you need a bypass capacitor on the input of the 7805 is if you have more than about 3 inches of wire between the powersupply filter capacitor and the input pin of the 7805. If you are using greater than 22uf in the output circuit of the 7805 the manufactures recommend a diode like a 1N4001 connected cathode to input pin and anode to output pin.
This is to protect the IC against discharge currents of the output capacitor.
Especially if the input happens to get shorted to ground.

As far as bypass capacitors for the TTL on one is needed for several packages. But most protoboards have a lot of extra contact area and
longer than necessary wires, so from habit I usually put one 0.1 for every
TTL on the protoboard. I generally use the multilayer ceramics that have 0.1 or 0.2 inch hole spacing.
 

patroclus

New Member
I don't know if I understud exactly the last thing you said. So should I use a small ceramic capacitor close to each TTL?? Is it that?
What happends if it is not close to them?

I'm gonna make the circuit on two universal board (soldering), so the 7805 is on one of them, along with 5 TTLs, and the other one containing another 8 TTLs. Initially I was going to use only an output capacitor along with the 1n4001 (where the 7805), and planned no other. I though that using a good filtering at the output, it was not necesary to put other capacitors along the rest of the circuit...
 

k7elp60

Active Member
I would put a minimum of 2, 0.1uf ceramic capacitors on the board with
5 TTL circuits and a minumum of 3, 0.1uf ceramic capacitors on the board with 8 TTL circuits. They don't have to be the miniature ones. I spoke of these as I worked at a electronic distributor and had access to these at
a real good price. Regular disk will work, but if you get much over 100 volts ones they will be quite large. You can always solder them on the bottom of the board between the Vcc pin and Ground on the TTL pins.
 

Anthony Tursi

New Member
You probably have fixed your problem by now, but I recently had a similar problem. I had an old 7490 chip connected as a bcd counter using a 555 chip for clock pusles, and it worked fine. I substituted a 74LS90, and it didn't work. I got some bizarre count sequence. I remembered that it was recommended that a capacitor be placed between pin 5 and ground for the 555 which I didn't need with the 7490. I added the capacitor, and the the 74LS90 now works as expected.
 

crutschow

Well-Known Member
Most Helpful Member
Welcome to the ancient art of necromancy which seems to be often practiced by newbies on these sites.
Yes, it is likely, that after 11 years, the problem has been fixed or forgotten. :rolleyes:
 

steven7890

New Member
Decoupling capacitors ......I have made it a habit to put a 0.1uF ceramic cap or multilayer cap or tantalum cap (not E-cap) across the Vcc and Ground for each TTL and CMOS IC, plus a 10uF E-cap between +5V rail and ground plane for a group of 6 to 10 IC . ...For high power IC like microcontrollers or LED / Line drivers, a 0.1uF with a E-cap 4.7uF to 10uF is used. And all these always work for me for years. In fact it also apply to analog devices like op-amp
 

JimB

Super Moderator
Most Helpful Member
I am new to this site. I didn't notice that the posting data was in the upper left hand corner!
Welcome to ETO.
Always be aware of the age of a thread I have been caught out like this a couple of time myself.
Old threads are sometimes revived by spammers and those who post nonsense. (I am NOT including you in this category).

JimB
 

EinarA

Member
Anthony's cure for this problem is an interesting quirk of the 555 which I had not encountered before. It does boil down to bypass capacitors, which the 555 needs lots of since it shorts out the supply every time it switches state, something most users of this part seem blissfully unaware of. When building a circuit on a solder less bread board you should use a cap for every IC, connected as directly between the supply pins as possible.
 

Anthony Tursi

New Member
I didn't have any problems using the 7490. It cropped up when I substituted the 74LS90; it is not as forgiving as the old 7490. Anyway, adding a bypass capacitor did solve the problem. Thanks for the advice.
 
Last edited:

jpanhalt

Well-Known Member
Most Helpful Member
Pin 5 is the control voltage. A capacitor there is optional, except at high frequencies (high for the 555 -- the datasheet gives the frequency at which it is recommended). What frequency were you operating at? I don't believe one could rightly call it a decoupling capacitor.

John
 

EinarA

Member
I don't want to get into a long discussion on 555s but pin 5 is connected to the resistor divider that sets the trigger voltages. Without a cap on it when the IC triggers ,the supply sags and the internal voltage drops below the input, making the latch untrigger. This creates a burst of high frequency oscillation that will toggle a counter multiple times. The effect is worse at low frequencies.
 
Status
Not open for further replies.

Latest threads

Top