__interview question

  • Thread starter robin.pain@tesco.net
  • Start date
R

robin.pain@tesco.net

Guest
When you connect the load, it oscillates. What could you try to prevent it?


|\
-----|+\
| >--------------------
/\/\-----|-/ | |
| |/ | load
| | |
| | -----
---\/\/\/---\/\/\/-- ---
-


Robin
 
robin.pain@tesco.net wrote:
When you connect the load, it oscillates. What could you try to
prevent it?


|\
-----|+\
| >--------------------
/\/\-----|-/ | |
| |/ | load
| | |
| | -----
---\/\/\/---\/\/\/-- ---
-


Robin
Click to expand...
Put a cap from the neg op-amp input to the op-amp output. Place a small
resister from the op-amp output to the load. Keep the FB resister
connected to the load.


Kevin Aylward
salesEXTRACT@anasoft.co.uk
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.
 
On 22 Oct 2003 00:37:42 -0700, robin.pain@tesco.net
(robin.pain@tesco.net) wrote:

When you connect the load, it oscillates. What could you try to prevent it?


|\
-----|+\
| >--------------------
/\/\-----|-/ | |
| |/ | load
| | |
| | -----
---\/\/\/---\/\/\/-- ---
-


Robin
Click to expand...

Use one of those cool National C-load opamps. A capacitive load just
slows down the dominant pole.

John
 
On 22 Oct 2003 00:37:42 -0700, robin.pain@tesco.net
(robin.pain@tesco.net) wrote:

When you connect the load, it oscillates. What could you try to prevent it?


|\
-----|+\
| >----+-----r----+------+
/\/\--+--|-/ | | |
| |/ c | load
| | | |
| | | -----
+---\/\/\/-+--\/\/\/--+ ---
-
Click to expand...
John
 
robin.pain@tesco.net (robin.pain@tesco.net) wrote in message news:<bd24a397.0310212337.12e3db12@posting.google.com>...
When you connect the load, it oscillates. What could you try to prevent it?


|\
-----|+\
| >--------------------
/\/\-----|-/ | |
| |/ | load
| | |
| | -----
---\/\/\/---\/\/\/-- ---
-


Robin
Click to expand...
I tried putting a capacitor from ouput to -ve input and it made it
worse! How come? More negative feedback should stop it?

Confused Student
 
Put something on the noninverting input such that it doesn't act
like a broadband receiving antenna, capacitively coupled to the
opamp's own output. Also, bypass (decouple) the opamp's power
supplies.

Cheers!
Rich

robin.pain@tesco.net wrote:

When you connect the load, it oscillates. What could you try to prevent it?


|\
-----|+\
| >--------------------
/\/\-----|-/ | |
| |/ | load
| | |
| | -----
---\/\/\/---\/\/\/-- ---
-


Robin
Click to expand...
 
<robin.pain@tesco.net> wrote in message
news:bd24a397.0310222258.1d9584d7@posting.google.com...
robin.pain@tesco.net (robin.pain@tesco.net) wrote in message
news:<bd24a397.0310212337.12e3db12@posting.google.com>...
When you connect the load, it oscillates. What could you try to prevent
it?


|\
-----|+\
| >--------------------
/\/\-----|-/ | |
| |/ | load
| | |
| | -----
---\/\/\/---\/\/\/-- ---
-


Robin

I tried putting a capacitor from ouput to -ve input and it made it
worse! How come? More negative feedback should stop it?

Confused Student
Click to expand...
Huh? You placed a capacitor from the output of the op-amp to where? What
is a "-ve input"? Since it made the oscillation worse I assume you placed
the capacitor from the output of the op-amp to the negative supply rail
(sometimes labeled something like -Vee, or othertimes just V- in the
datasheet). The negative supply rail isn't normally considered an "input"
pin on the opamp. Rather it is a supply pin while the inverting and
non-inverting input pins are referred to as the "input" pins. This type of
arrangement is precisely the type of load most likely to cause op-amps to
self oscillate for no apparent reason. A capacitor from the output to some
low impedance connection (IE: ground, negative supply rail, positive supply
rail) can cause problems. Normally a few tens of picofarads or less isn't a
problem.

For increased stability place the capacitor from the op-amp output to the
op-amp's inverting input pin (IE: from pin 6 to pin 2 on the 741).
Alternatively to, or in addition to this, place a small resistor (such as 47
ohms) directly to the output pin to the load. Both of these circuit
modifications will change the circuit's performance characterisitics,
sometimes in undesireable ways.
 
On Fri, 24 Oct 2003 12:53:20 -0700, "Fritz Schlunder" <me@privacy.net>
wrote:

robin.pain@tesco.net> wrote in message
news:bd24a397.0310222258.1d9584d7@posting.google.com...
robin.pain@tesco.net (robin.pain@tesco.net) wrote in message
news:<bd24a397.0310212337.12e3db12@posting.google.com>...
When you connect the load, it oscillates. What could you try to prevent
it?


|\
-----|+\
| >--------------------
/\/\-----|-/ | |
| |/ | load
| | |
| | -----
---\/\/\/---\/\/\/-- ---
-


Robin

I tried putting a capacitor from ouput to -ve input and it made it
worse! How come? More negative feedback should stop it?

Confused Student

Huh? You placed a capacitor from the output of the op-amp to where? What
is a "-ve input"?
Click to expand...

In certain benighted parts of the world, "-ve" substitutes for the
word "negative", and "+ve" means "positive." But these people also say
"maths", so are clearly incurable.

John
 
John Larkin <jjlarkin@highSNIPlandTHIStechPLEASEnology.com> wrote in message news:<vi4jpvcuveofm1lh9q1eeilcknu7uaqmup@4ax.com>...
On Fri, 24 Oct 2003 12:53:20 -0700, "Fritz Schlunder" <me@privacy.net
wrote:


robin.pain@tesco.net> wrote in message
news:bd24a397.0310222258.1d9584d7@posting.google.com...
robin.pain@tesco.net (robin.pain@tesco.net) wrote in message
news:<bd24a397.0310212337.12e3db12@posting.google.com>...
When you connect the load, it oscillates. What could you try to prevent
it?


|\
-----|+\
| >--------------------
/\/\-----|-/ | |
| |/ | load
| | |
| | -----
---\/\/\/---\/\/\/-- ---
-


Robin

I tried putting a capacitor from ouput to -ve input and it made it
worse! How come? More negative feedback should stop it?

Confused Student

Huh? You placed a capacitor from the output of the op-amp to where? What
is a "-ve input"?


In certain benighted parts of the world, "-ve" substitutes for the
word "negative", and "+ve" means "positive." But these people also say
"maths", so are clearly incurable.

John
Click to expand...
Of all the Americanisms
Perhaps the worst of any:
Radios have aerials
Insects have antennae

Robin
 
"Fritz Schlunder" <me@privacy.net> wrote in message news:<bnbvn6$vr6gj$1@ID-203926.news.uni-berlin.de>...
robin.pain@tesco.net> wrote in message
news:bd24a397.0310222258.1d9584d7@posting.google.com...
robin.pain@tesco.net (robin.pain@tesco.net) wrote in message
news:<bd24a397.0310212337.12e3db12@posting.google.com>...
When you connect the load, it oscillates. What could you try to prevent
it?


|\
-----|+\
| >--------------------
/\/\-----|-/ | |
| |/ | load
| | |
| | -----
---\/\/\/---\/\/\/-- ---
-


Robin

I tried putting a capacitor from ouput to -ve input and it made it
worse! How come? More negative feedback should stop it?

Confused Student
Click to expand...
<snip>

For increased stability place the capacitor from the op-amp output to the
op-amp's inverting input pin (IE: from pin 6 to pin 2 on the 741).
Alternatively to, or in addition to this, place a small resistor (such as 47
ohms) directly to the output pin to the load. Both of these circuit
modifications will change the circuit's performance characterisitics,
sometimes in undesireable ways.
Click to expand...
Yes, these things work ok. The oscillation stops but as you say the
bandwidth is reduced.

If I do this instead:-

|\
-----|+\
| >--------------------
/\/\-----|-/ | |
| |/ | load
| | |
| | -----
---\/\/\/---\/\/\/-- ---
| -
/
\
/
|
|
-----
-----
|
|
-------
---
-

and fiddle about with capacitor and resistor, I get better bandwidth.
How does this work?


Even more confused student
 
On Sun, 26 Oct 2003 23:57:18 -0800, robin.pain@tesco.net wrote:

Of all the Americanisms
Perhaps the worst of any:
Radios have aerials
Insects have antennae
Click to expand...
AFAIK, the agreed international term for a radiator or collector of
electromagnetic radiation in the radio part of the spectrum is "antenna",
plural "antennas".

Cast in stone, I think, (correct me if I'm wrong), by the ITU many years
ago.

Probably enshrined in some ISO document, too.

In German "antenne"
In French "antenne"

Don't know about Italian or Spanish, but I'm sure someone will enlighten
me.

"Aerial" is peculiarly British, and doesn't appear in any modern or recent
engineering books or articles by British authors, either. It's deprecated
in engineering circles.

--
Then there's duct tape ...
(Garrison Keillor)
nofr@sbhevre.pbzchyvax.pb.hx
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top