LM311 PWM circuit

[Terry Collins]

I have some questions on using LM311 comparator op-amps.

1) We are using a pair as a light dimmer. One is the oscillator and the
other is the PWM driving a jfet (?). The problem is that the circuit
given to us doesn't show what to do with pins 5 (balance) and 6
(balanbce/strobe). Thie chips are DIL 8pin.

We are running it off a 12V supply (0 - 12v). For the oscillator; Pins 1
& 4 are tied to gnd. Pin 2 (+ve) is divided between 12V & gnd by two
resistors, Pin 3 (-ve) is RC circuit . Pin 8( Vcc) is 12v).

Pin 7 (output) is connected by diodes to pins 2&3 (I can not find any
circuit like this (anode end at 7).

So does anyone have any idea on where pins 5 & 6 should be connected?
Gnd or Vcc? For both 311's?


2) How do you fault find this circuit?
We know the oscillator works. We used an Amp to pick up the signal (no
scope). We also used it to slow down the frequency of the original
circuit when we built it on the prototype board..

The osc provides a signal (off pin 3 #1-311) into pin 3 of 2nd 311 and
that is reaching that pin. Pin 2 is a pot between Vcc & Gnd) and its
signal (voltage) is reaching the pin.

We are getting something out the output (pin 7) to the Jfet, but no joy.

The whole circuit was built on a prototype board and worked (after
adjusting Rc). We have now built it on a ferro board and tested all
links/soldering, but no joy. We have even swapped the 311's, but still
no joy.

If I check voltages, the jfet turns on (if the put is up (towards Vcc),
and stays on until we turn the pot down.

3)Can anyone give me any idea of the frequency range that we can feed
into the PWM 311? With the original circuit, I estimate the osc was
around 3Mhz and that that was why the PWM was not working. We have
reduced the Osc to a few Khz and it worked on the protype board, but not
on the ferro.

Really are head scratching atm.

TIA

 

[Franc Zabkar]

On Sat, 22 May 2004 22:41:59 +1000, Terry Collins <terryc@woa.com.au>
put finger to keyboard and composed:

I have some questions on using LM311 comparator op-amps.

1) We are using a pair as a light dimmer.

It's not an answer to your question but ...

Figure 3 on page 290 of the 2002-2003 Dick Smith catalogue shows a
light dimmer circuit consisting of a 555 timer IC, an MTP3055E MOSFET,
a potentiometer, and a few minor support components. This PWM circuit
operates at 300Hz from a 12V supply.


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.

 

[Terry Collins]

Franc Zabkar wrote:

Figure 3 on page 290 of the 2002-2003 Dick Smith catalogue shows a
light dimmer circuit consisting of a 555 timer IC, an MTP3055E MOSFET,
a potentiometer, and a few minor support components. This PWM circuit
operates at 300Hz from a 12V supply.

Thanks, luckily I had that on the shelf. I will probably play with it
for one of mine, but not an option for my partner who is most adament
that he wants to build this particular circuit {:-(.

 

[Franc Zabkar]

On Sat, 22 May 2004 22:41:59 +1000, Terry Collins <terryc@woa.com.au>
put finger to keyboard and composed:

I have some questions on using LM311 comparator op-amps.

I don't see any responses from the analogue experts so let me [not an
expert] have a go ...

1) We are using a pair as a light dimmer. One is the oscillator and the
other is the PWM driving a jfet (?). The problem is that the circuit
given to us doesn't show what to do with pins 5 (balance) and 6
(balanbce/strobe). Thie chips are DIL 8pin.

These are used for trimming the accuracy of the comparator (offset
nulling). In your circuit, accuracy is not an issue.

We are running it off a 12V supply (0 - 12v). For the oscillator; Pins 1
& 4 are tied to gnd. Pin 2 (+ve) is divided between 12V & gnd by two
resistors, Pin 3 (-ve) is RC circuit . Pin 8( Vcc) is 12v).

I presume C connects between pin 3 and ground, while R is between pins
3 and 7.

Pin 7 (output) is connected by diodes to pins 2&3 (I can not find any
circuit like this (anode end at 7).

I don't understand this either. The "standard" multivibrator circuit
in the app notes has no diodes, only a second resistor between pins 7
and 2 (for hysteresis). If I understand you correctly, pins 2 and 3
connect to their respective diode's cathode, while both anodes connect
to pin 7.

So does anyone have any idea on where pins 5 & 6 should be connected?
Gnd or Vcc? For both 311's?

Leave them open. (?)

2) How do you fault find this circuit?

Preferably with an oscilloscope.

We know the oscillator works. We used an Amp to pick up the signal (no
scope). We also used it to slow down the frequency of the original
circuit when we built it on the prototype board..

If the diodes are connected as described above, then I can't see how
the oscillator can oscillate at all. Once pin 7 goes high, then both
inputs will immediately go high (via the diodes), resulting in an
indeterminate state. AFAICS, you would at least need to remove the
diode from pin 2. This would then allow the oscillator to produce a
sawtooth at pin 3. However, even in this case there may be a slight
hiccup at the point where pin 7 goes high. Specifically, pin 7 (and
therefore pin 3) would be trying to switch to the supply rail, but the
comparator will be attempting to change state before this happens, ie
at the point where the voltage at pin 3 equals the voltage at pin 2.
To avoid this discontinuity, I suggest a low value resistor (100
ohms?) in series with the diode at pin 3. Or perhaps the output
impedance of the LM311 is sufficiently high not to warrant this
precaution? In any case, I believe you would require a suitable
resistor between pins 7 and 2, as described above. Also, the range of
the pot in the second LM311 would need to be tailored to suit the
amplitude of the sawtooth. I suggest adding suitably valued resistors
at the top and bottom ends of the PWM pot for this purpose.

The osc provides a signal (off pin 3 #1-311) into pin 3 of 2nd 311 and
that is reaching that pin. Pin 2 is a pot between Vcc & Gnd) and its
signal (voltage) is reaching the pin.

I suspect pin 3 is stuck in the high state, perhaps with a tiny amount
of superimposed AC signal. This would be the reason why the second
comparator will only switch when the PWM pot is cranked up to Vcc.

We are getting something out the output (pin 7) to the Jfet, but no joy.

The whole circuit was built on a prototype board and worked (after
adjusting Rc).

I don't understand this.

We have now built it on a ferro board ...

Do you mean Veroboard?

... and tested all
links/soldering, but no joy. We have even swapped the 311's, but still
no joy.

If I check voltages, the jfet turns on (if the put is up (towards Vcc),
and stays on until we turn the pot down.

3)Can anyone give me any idea of the frequency range that we can feed
into the PWM 311? With the original circuit, I estimate the osc was
around 3Mhz and that that was why the PWM was not working. We have
reduced the Osc to a few Khz and it worked on the protype board, but not
on the ferro.

Really are head scratching atm.

TIA

Maybe it's time to borrow a scope.


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.