I have a requirement to add a bi-level clipping circuit to an existing device that uses a single ended 9VDC power supply. I attempted to implement it using two LM358s similar to Crutshow's circuit shown in the third example at https://www.electro-tech-online.com/articles/analog-max-min-value-or-bi-level-clamp-circuits.808/. It failed to function. Have I done something wrong or is this not possible with a single ended power supply? My internet search did not turn up any clipping circuits using a single ended power supply. I would appreciate any help you can offer.
Thanks.
Sam

 

At first glance, I made the following observations:
There is no DC voltage level specified on the input signal. It should be 1/2 VCC.
The negative inputs to the op-amps are referred to VCC. They should be at 1/2 VCC so add a 10K resistor from the negative inputs to supply COM (-).
Those changes should help to get you closer to what you need.

 

Basically you can't clip anything below ground, so the signal would need to have a DC offset added, as KW noted.

Post your .asc file.

 

@Keith Walker & Crutschow
Thank you gentlemen. I'll give it a whirl and let you know how it turns out. It may have to wait until I get some more ICs.
Sam

 

@Keith Walker & Crutschow
I added the 10 k resistor as suggested and reversed diode D2 as shown in Crutschow's third example at https://www.electro-tech-online.com/articles/analog-max-min-value-or-bi-level-clamp-circuits.808/. The new configuration is shown in the attachment.

The circuit still does not function correctly. The circuit output voltage follows the input regardless of max and min settings. The voltage at the output of the max op amp follows the input input as long as it is beneath the max voltage setting. It stays at that level when the input is above the max setting. The minimum op amp output follows the max voltage output until it reaches about 2 volts (which is where the max output stops changing) and stays there as the input decreases. The min voltage adjustment has no effect. I replaced diode D4 in the input circuit with no effect. The original diode was OK.

The next step will be to reverse diode D2 at the min op amp output. If that does not work I will swap the inputs to the min op amp. This is getting to be about as pretty as making sausage in public. If you have other suggestions, I will be most appreciative.
Sam

 

reversed diode D2

Why?
Randomly changing the circuit is unlikely to produce the result you want.

If you have other suggestions

What, exactly, is the input signal amplitude (plus and minus) and its frequency.

 

Crutschow
Thanks for your interest in this problem. I originally had D2 in the same orientation as the others. I reversed it to conform to your previously referenced example. It was in this position for my last post (anode facing op amp output).

For this test circuit I am using a 13.8 VDC single ended power supply. The input for this test circuit is 0/13.8 v from a pot. In the final circuit the power supply will be 9 vdc, and the circuit input will be a 0/9 VDC synthesized sine wave at 60 hz.

 

circuit input will be a 0/9 VDC synthesized sine wave at 60 hz.

Do you mean 0V-9Vpk sine wave?

 

reversed diode D2
can you try to reverse diode D3 and D4, and let us know the results?

 

Below is the sim of my third circuit operating with a single +9V supply, and with a sinewave input of 0-9Vpk, showing the clip voltages going from 9V and 0V, to both =4.5V .
Note that the maximum output is about 7.5V due to the LM358 not being a rail-rail type op amp.
If you want to go to 9Vpk, you will need to either increase the supply voltage, or go to a rail-rail op amp.

Is this what you want?

 

Crutschow
Thank you for the circuit changes. I will implement them and report the result, probably late tomorrow.
I really appreciate what you have done on my behalf. Hopefully others will also benefit from your work.
Sam

 

Thank you for the circuit changes

The circuit was not changed from my original except to go to a single supply, and keep the clamp and signal levels within the supply range.

 

Crutschow
I replaced the minimum setting pot which was open, and found a cold solder joint. After fixing these and changing the circuit to the circuit you last suggested, I checked it without the IC's installed. After verifying that the diodes, 10K resistor and pots were correctly installed, I applied 13.8 VDC and measured the voltages from the max and min pot wipers to be sure they changed properly.

When the circuit was powered with the ICs installed the output changed with the input in a non linear fashion. With the input between the max and min setpoints, the output followed the input. As the input increased above the max setpoint the output stayed close to to the max setpoint (10 V) but crept up as the input increased. When the input got near 13.8V the output suddenly increased to 13.8v.

When the input decreased toward the minimum setpoint (3V) the output became quite non linear around 5V. Just a nudge on the pot providing the input sends the input voltage to below 1 V and the output is about 1.1V. By being VERRRY careful moving the wiper the input voltage can be decreased below 5 v. Between 5v and 3v the output voltage exceeds the input, then the input suddenly drops to a Few millivolts and the output drops to about 1.1v as the wiper on the input pot heads toward zero.

I tried changing the min setpoint to 2 volts with similar results. I tried several different 358 op amps for the setpoints with similar results.

So far it looks like the max setpoint is working and the min setpoint needs a good kick in the rear. I'm getting out my whip and will flog it into submission. I'm thinking about reinstalling the 10K voltage offset resistors as the first crack of the whip.

Meanwhile, any suggestions will be appreciated.

Sam

 

What type of pots are you using?

You might trying adding some resistors in series with the pot elements, either on the top or bottom, depending on what pot output voltage range you want, to make the voltage change less sensitive to pot movement.

 

The max/min pots are enclosed, 5K, 12 turn, screw adjustment, PCboard mounted types. The signal for test purposes is from a 5K, single turn pot normally used for a speed pot in a drive.. When in service, the input signal will be a 0/9V peaK to peak, 60 Hz sinusoidal varying DC signal.
If the circuit is run without the max/min op amps installed, there is no non-linearity problem.

How about this approach? Your idea of adding resistors may do the trick. Since the max & min voltages will never be changed after initial setup, perhaps the 0/9 V signal can be fed to 3 pots as shown in the attachment, or a pot with appropriate resistors on each end instead of pots. The sine wave peaks will be the max and min.

 

How about this approach?

You can try whatever you want, but the circuit should work as I show.
If not, I expect there is a wiring error or bad contact.

 

"..... I expect there is a wiring error or bad contact."
Based on recent experiences with this supposedly simple circuit I'd say you are probably right. I had a funky 358 too.
Thanks again for your help.
Sam

 

The pot approach I suggested yesterday was unworkable. I had not thought it out properly before posting. This looks promising.
Sam.

 

The attached circuit was tested with input signals with and without zero offset as shown in the spreadsheet. The cell formulas in the section on the right side of the spreadsheet use the same column designation as the section on the left so comparisons are obvious. This circuit has not been tested using positive and negative power supplies and signals.
Sam
EDIT: The comment on line 22 now references the input offset voltage. Circuit values in the section on the right hand side reflect those in the entry value section on the left side except for the input offset voltage.
EDIT: This is not a clipping circuit. It simply re-scales a signal that moves between limits. If the signal goes beyond those limits, the output will also go beyond its designed limits.