From: Winfield Hill
Subject: Re: How to get CMOS counter to drive a relay?
Date: 23 Oct 2002 05:22:07 -0700
Organization: Rowland Institute
References: <0001HW.B9CEF278023A90B81662EAD0@news.covad.net> <3DA9D6CC.firstname.lastname@example.org> <3DA9D9E1.email@example.com> <3DA9EE2A.B7CFD919@juno.com> <3DAAF96D.firstname.lastname@example.org> <3DAB71EC.DB68183C@juno.com> <3DAC2309.email@example.com> <firstname.lastname@example.org> <3DAFFA70.email@example.com> <firstname.lastname@example.org> <3DB1869F.649A02A4@bellatlantic.net> <email@example.com> <firstname.lastname@example.org> <email@example.com> <firstname.lastname@example.org> <email@example.com> <firstname.lastname@example.org> <email@example.com> <firstname.lastname@example.org> <email@example.com>
X-Newsreader: Direct Read News 2.98
Tony Williams wrote...
> Frank Bemelman wrote:
>> Okay, found another 4060. Relay from output to ground,
>> no diode. Vcc = 18V. Outputfrequency is 25Hz, and the
>> relay is clicking like mad. The undershoot is 1V below
>> GND, or 0.7V with diode. Voltage drop from the output
>> is 5V, so the relay gets 13V. [snip]
>> The 4060 is now running at 28,5V, relay draws 40mA,
>> no diode. Oh, and no decoupling cap, and long wires ;)
> Thanks Frank. Strange..... I'd have laid odds that
> throwing a pin below the substrate would have stopped
> the ic from operating, and blown it earlier.
If the current forced into a CMOS pin, whether Vdd or Vss,
is high enough, the chip will go into SCR latchup. In this
condition it shorts out the supply rails with about a 1.0 to
1.2V drop, and the SCR is able to carry several amps if the
power supply insists on delivering it. Of course, the chip
will get very hot. But if the supply current limits at say
750mA or so, the latched chip won't be too badly damaged and
will work properly again after the supply is cycled off - on.
When I began working with cmos in the late 60s, the spec for
incurring SCR latchup was only 20mA, and latchup was not
uncommon. A simple slip of a probe could trigger it. In
the last 30 years manufacturers have been careful to reduce
the gain of the parasitic transistors responsible for SCR
latchup, so the spec is now closer to 50 or 60mA, and actual
latchup onset may occur at even higher currents, say 100mA.
Frank subjected his poor suffering 4060 chip to only 40mA
flyback current, far below 60 or 100mA, etc. I respectfully
suggest that he try running three or four relays in parallel,
that should do it nicely. If the IR-drop from powering four
relays doesn't get his stubborn 4060, the flyback-triggered
SCR latchup will. :-)