NE555N help needed

[tool49]

Here's my problem. I built the circuit on the right on the following page:
http://ourworld.compuserve.com/homep...tm#555mono.gif

Now I bought the NE555N from a local shop. In the lot (6 of them) one is from ST and the 5 others are from Philips. They should all behave exactly the same way right? Well no... The ST chip does exactly what is expected; the output goes high for 1 second on a falling signal on the trigger. Now the Philips chips react differently; the output goes high and stays high indefinitely.

I have absolutely no idea why one brand of chip would work and not the other... Any idea will be appreciated...

Thanks,
Sébastien

[skyraider]

bad chips maybe..
Does this happen with all the phillips chips?

[tool49]

Yep with all 5 Phillips chip. And since I'm swapping the chips in the socket, it can't be because I messed the circuit... Otherwise the ST chip wouldn't work either.

I'm starting to suspect counterfeit chips... although for a ¢20 chip it is not very likely...

Thanks,
Sébastien

[peranders]

Originally posted by tool49
They should all behave exactly the same way right?

Advises from the hard world: You always have to read the datasheet from the manufacturer in mind if you want to be totally sure. They are "almost" the same.

I have a bad experience from the LM324 opamp. We at work have used models with approx. 0.1-0.2 V as "low" voltage out. Some day we used an another brand with 0.6 V out as minimum..... not a success if the load was a bipolar transistor and the design was made for less than 0.5 volts.

[janneman]

The most probable cause is differences in bias current at the timing pins (6 & 7 IIRC). You use a very large elcap (100uF). If that cap has a leakage current on the same order as the chip bias current, the cap will never charge up.
Try a smaller cap, like 25uF and/or a low leakage one. You can then get back to right timing with a larger resistor, although this may bring back the same problem. 100uF is really too large for a bipolar 555. Try a CMOS version, that'll work. Or go to 10 x shorter timing and put a 74xx divider chip at the output. That will also restore a 50% duty cycle, which may or may not be important for your app.

Jan Didden

[tool49]

Thanks Jan and Peranders.

I forgot to say that I modified this circuit slightly to accomodate my timing. (By the way I'm using the AC coupled circuit with originaly 22uF and not the other one with originaly 100uF) The parts I've used are as follow: R is 100K and C is 10uF so as to have a monostable circuit with ~1s output. I went back to the store this morning and bought some more chips. I now have a sample of 5 ST chips, 5 Philips and 5 National. The ST are all working properly. The Philips all latch up. The National all work properly but the time response is a bit off (~1.5s). So I'm still puzzeled. The worst part is that I can't pin point any big difference in the datasheet so as to point me to what might be the discrepancy. In all cases, my trouble is solved, I'll simply use the ST chips for now...

Thanks to all for your input.
Sébastien

[skyraider]

I would just return all the other chips and exchange with ST chips.
Wouldnt be bothered to go into the 'scientific R&D'.
But if curiosity gets too much over you,...