Maker Pro
Maker Pro

Voltage limiting without altering waveform

arg733

Dec 14, 2010
89
Joined
Dec 14, 2010
Messages
89
Thank you for that circuit. I can get the negative rail from the sound card or from the 9v power supply.
 

BobK

Jan 5, 2010
7,682
Joined
Jan 5, 2010
Messages
7,682
Kris,

Isn't that going to allow about +-8V to the sound card with 9V power?

Bob
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
It is intended to view 400V AC so should be designed for 1kV input unless a divide by 10 probe is used when 100V input should be adequate. Perhaps a divide by 100 probe would be best.

A diode clipper could come after the op-amp.
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
BobK, yes it will. Do you think that will damage it? I doubt it would.

duke37, are you sure he wants to measure 400VAC? If so, yes an external divide-by-10 or divide-by-100 probe would be a very good idea.
There's no point clipping after the op-amp; the op-amp will clip the signal itself! The diodes are mainly to protect the op-amp input.

I will add a transistor-based voltage detector stage with an LED. This will provide an indication that the signal level is too high for the sound card and that the sound card will be clipping the signal. I'll upload it in a few hours.

To the OP. What is the peak-to-peak signal voltage at the sound card's input that causes the sound card to clip?
 
Last edited:

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
attachment.php


This version adds a bipolar voltage detector and indicator LED.
The NPN transistors are all BC547B or 2N3904. PNPs are all BC557B or 2N3906.
You can adjust the LED series resistor value to give the desired current with the voltage rails you are using.

You apply an adjustable signal to the input, and increase the signal level until the sound card just starts to clip, then adjust VR1 (voltage detector sensitivity) until the LED just illuminates. If you reduce the input level, the LED will go out. As you increase the input level again, make sure the LED lights up slightly BEFORE the sound card starts to clip, to give yourself a small safety margin.
 

Attachments

  • ARG733.002.GIF
    ARG733.002.GIF
    11.8 KB · Views: 121

arg733

Dec 14, 2010
89
Joined
Dec 14, 2010
Messages
89
Wow! all that transistors for a led? Can i replace the transistors (don't have a lot of those) in your circuit with mosfets (i have a lot of those)?


To the OP. What is the peak-to-peak signal voltage at the sound card's input that causes the sound card to clip?

I'm not sure but i think it is 2Vpp.
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Wow! all that transistors for a led? Can i replace the transistors (don't have a lot of those) in your circuit with mosfets (i have a lot of those)?
Q1 and Q2, no. The others, yes, but with circuit changes. I suggest you get some transistors! They can be pretty useful. You may be able to find a bulk pack with a selection of NPN and PNP small-signal transistors.

I've just noticed an error. D1 and D2, the zeners across the power supply rails, should be specified for slightly more than the supply rail voltages that you're using. For example if you're using +/- 12V rails from the sound card, then use 13V zeners, instead of the 10V ones shown on the schematic.
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
I've made a few more improvements. Here's an updated schematic. Please throw out the earlier schematics.

attachment.php


I've changed the op-amp supply rail and protection circuitry to completely prevent any back-feeding of input overload voltage into the sound card's supply rails.

I've changed the overload indicator circuit to use a MOSFET to drive the LED. This will give a cleaner indication of overload.

The input resistor is now specified as fusible.

I hadn't noticed that you intend to measure 400VAC. That's potentially very dangerous and I strongly recommend against it. At the very least you will have to ensure that your input resistors and input selector switch are rated for this voltage. Your best option is to use an external x10 or x100 probe, as suggested earlier by duke37.
 

Attachments

  • ARG733.003.GIF
    ARG733.003.GIF
    13.2 KB · Views: 126

arg733

Dec 14, 2010
89
Joined
Dec 14, 2010
Messages
89
I hadn't noticed that you intend to measure 400VAC. That's potentially very dangerous and I strongly recommend against it. At the very least you will have to ensure that your input resistors and input selector switch are rated for this voltage. Your best option is to use an external x10 or x100 probe, as suggested earlier by duke37.

I have an x100 probe and i will also use pots for the attenuator instead of fixed resistors and a switch.

So the transistors should be logic gate right? and can use small ones like 300mA?

Thank you very much.

Edit: Don't draw a new circuit just for the potentiometers I know how to connect them :)
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
I have an x100 probe and i will also use pots for the attenuator instead of fixed resistors and a switch.
Good. The x100 probe is probably designed to work with a measuring device that has a specific input impedance. Check the manual for the probe.

So the transistors should be logic gate right? and can use small ones like 300mA?
Where did you get that idea? They should be normal bipolar transistors. They only need to be rated for 100 mA and 30V. They should have fairly high gain. I suggested some suitable part numbers for them in an earlier post, and on the diagram.

Thank you very much.
You're welcome :)
 

arg733

Dec 14, 2010
89
Joined
Dec 14, 2010
Messages
89
Won't this give a DC instead of an AC signal to the sound card?
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Well, the output is DC-coupled to the sound card input, if that's what you mean. The input is AC-coupled and the op-amp is acting as a buffer, so there will not be any (significant) DC on its output.
 

arg733

Dec 14, 2010
89
Joined
Dec 14, 2010
Messages
89
Are you sure it is TL072? because it has only 2 inputs It's TL071 That has 4inputs.
What should i use?
Also what if i use 1n4007 (1.1v) instead of the 1n914 (0.6v) ?
Thank you.
 
Last edited:

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
TL071 is a single op-amp. Since the circuit only needs one op-amp, you can use a TL071. The TL072 is a dual op-amp; it consists of two TL071 op-amps without the offset null connections. I specified a TL072 because they're more commonly used in audio circuits. They're the same size as the TL071 and the extra pins on the TL071 aren't normally used.

So you can use a TL071 if you change the pin numbers to match, or half of a TL072. Notice that I labelled the op-amp "U1A". The other half, "U1B", isn't used. If you're not using it elsewhere in the circuit, its inputs should be tied to ground, and its output left unconnected. I should have shown that in the circuit diagram.
 
Last edited:

arg733

Dec 14, 2010
89
Joined
Dec 14, 2010
Messages
89
I mean the "triangle" symbol of an TL072 has only 2 inputs (in+ , in-) and out whereas the TL071 has in+ in- offset N1 , offset N2 and out.
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
I know. Reread my post carefully. If you want to use a TL071, you can. Don't connect anything to the offset null pins.
 

arg733

Dec 14, 2010
89
Joined
Dec 14, 2010
Messages
89
Ok thanks now i get it.
I am now finally well enough (i was ill again) to go and by the parts from the nearest electronics shop which is in fact 30 miles away and i was typing a list of components that i need and i got confused when i was reviewing the TL072's datasheet.
 
Top