ULN2003 - ramping up the control voltage?

[tempus fugit]

Hi all;

I am using a ULN 2003 (multi darlington array relay driver IC) to drive
several relays for an audio bypass application. I'm getting some popping
when the relays switch, and was wondering if ramping up the control voltage
would help to reduce or eliminate this problem.

To achieve this, could I simply put a cap (say 0.1uF) across the across the
input and output of each Darlington pair? I have a data sheet of the ULN if
you need to see one.

Thanks

 

[Simon Dice]

tempus fugit ha escrito:

Hi all;

I am using a ULN 2003 (multi darlington array relay driver IC) to drive
several relays for an audio bypass application. I'm getting some popping
when the relays switch, and was wondering if ramping up the control voltage
would help to reduce or eliminate this problem.

To achieve this, could I simply put a cap (say 0.1uF) across the across the
input and output of each Darlington pair? I have a data sheet of the ULN if
you need to see one.

Thanks

What did you do with pin 9?
ULN2003 has a free wheeling diode that most be conected to the source
in case of inductive loads, as the relay coil is...

 

[tempus fugit]

I'm sorry Simon, I don't understand. Do you mean that pin 9 needs to be
connected to the voltage source?

Thanks

 

[Chris]

Hi all;

I am using a ULN 2003 (multi darlington array relay driver IC) to drive
several relays for an audio bypass application. I'm getting some popping
when the relays switch, and was wondering if ramping up the control voltage
would help to reduce or eliminate this problem.

To achieve this, could I simply put a cap (say 0.1uF) across the across the
input and output of each Darlington pair? I have a data sheet of the ULN if
you need to see one.

Thanks

Hi, Tempus. The short answer is, "no".

Relays will typically pull in at about 70% of rated voltage, and
release at around 15% of rated voltage. That's because it's a lot
easier to pull in the armature as it gets closer.

All you're going to do is delay the turn on of the relays a bit. As
the armature pulls in, it will make contact, and you'll get the "pop"
anyway. Sorry.

Good luck
Chris

 

[tempus fugit]

Hi Chris;

Thanks for the reply.

I read this at the Geofex web site:

"for non-mechanical switches the switch elements may couple a little of
their control signal into the signal path. This is what happens with JFET,
CMOS and relay bypass switches. If any of these are driven with a
sudden-change control voltage, the unavoidable capacitances inside them can
couple the fast wavefront into the signal lines, and this makes a pop. Some
CMOS switches couple only a nanosecond or two of sound in, but several volts
of it. The amplifier could not normally respond to this, but it overloads
other circuits in the effect, and that overload is then heard as a pop."

It's a little confusing there when he starts talking about 'non-mechanical
switches' and later mentions relay bypass switches (I'm taking that to mean
relays, which are mechanical. Is there some type of silicon relay bypass
switch that isn't JFET or CMOS?). He writes the solution later in the page
as ramping up the control voltage:

"Ramp the control voltage up/down slowly (over a mS or two) to keep it from
being coupled through the small capacitances."

So, (whether it will get rid of the pop problem or not) will connecting a
cap across the input and output pins ramp the voltage this way?

Thanks

 

[ehsjr]

Hi Chris;

Thanks for the reply.

I read this at the Geofex web site:

"for non-mechanical switches the switch elements may couple a little of
their control signal into the signal path. This is what happens with JFET,
CMOS and relay bypass switches. If any of these are driven with a
sudden-change control voltage, the unavoidable capacitances inside them can
couple the fast wavefront into the signal lines, and this makes a pop. Some
CMOS switches couple only a nanosecond or two of sound in, but several volts
of it. The amplifier could not normally respond to this, but it overloads
other circuits in the effect, and that overload is then heard as a pop."

It's a little confusing there when he starts talking about 'non-mechanical
switches' and later mentions relay bypass switches (I'm taking that to mean
relays, which are mechanical. Is there some type of silicon relay bypass
switch that isn't JFET or CMOS?). He writes the solution later in the page
as ramping up the control voltage:

"Ramp the control voltage up/down slowly (over a mS or two) to keep it from
being coupled through the small capacitances."

So, (whether it will get rid of the pop problem or not) will connecting a
cap across the input and output pins ramp the voltage this way?

Thanks

The relay is electro-mechanical. The coil is the "electro"
part - the switching part is mechanical. "Ramping up"
the control voltage will do nothing for you - except maybe
burn out the chip. Drive the relays properly - which includes
connecting pin 9 to +. The Geofex site may have a fix for
the popping noise that switching via relays causes - but
it does not involve increasing the control voltage.

Ed