Over-current Protection Circuit
Re: Over-current Protection Circuit
I've now ordered the same thing from Mouser. The price was a bit higher (International FedEx Shipping), but they had 30 in stock. I'll report back when I have the kit.
Terry
Re: Over-current Protection Circuit
The INA240EVM Boards arrived from Mouser a few days ago, but I had to get some larger diameter solder pins to fit the rather large holes for the terminations on the board. These arrived this morning and I've just assembled the circuit shown below.
Unfortunately, I am no further forward because the PWM still appears at the OUT terminal.
Penri,
If you have time could you have a look at the datasheet for the INA240 and the User Guide for the INA240EVM (see https://www.ti.com/lit/gpn/ina240 and http://www.ti.com/general/docs/suppprod ... %2Fsbou177) and let me know what you think TI mean when they say 'PWM Rejection' on Page 10 of the datasheet. I assumed that the block labelled 'PWM Rejection' in the Functional Block Diagram was there to remove the pulses, but clearly not.
I can see that the EVM board has provision to add input filtering (see Page 5 of the User Manual), but I was expecting the PWM rejection to work without that.
Any thoughts?
- INA240_Based_Current_Monitor.png (75.94 KiB) Viewed 216 times
Unfortunately, I am no further forward because the PWM still appears at the OUT terminal.
Penri,
If you have time could you have a look at the datasheet for the INA240 and the User Guide for the INA240EVM (see https://www.ti.com/lit/gpn/ina240 and http://www.ti.com/general/docs/suppprod ... %2Fsbou177) and let me know what you think TI mean when they say 'PWM Rejection' on Page 10 of the datasheet. I assumed that the block labelled 'PWM Rejection' in the Functional Block Diagram was there to remove the pulses, but clearly not.
I can see that the EVM board has provision to add input filtering (see Page 5 of the User Manual), but I was expecting the PWM rejection to work without that.
Any thoughts?
Terry
Re: Over-current Protection Circuit
Terry
The INA240 seem idea for our application and I'm at a lost to know why the results you are seeing are not bearings that out.
The way I read the PWM rejection information is that the device can cope with fast changing pulsed signals, in fact the illustration they use in figure 9.3 is even more extreme that our application with current direction being changed as first the upper transistor conducts then the lower one.
Three suggestion:
Firstly my old favourite, make sure there is a good solid common ground, a star configuration would be best.
Secondly relocate the PWM modulator in the test circuit so it sits between the INA240 and the load, this more closely replicates our intended application where we want to measure the current drawn from the PSU. The existing configuration should work perfectly well but as I'm clutching at straws I have to suggest it.
Lastly see what difference tying the Reference signals to the INA240's Vs (+ve supply).
Penri
The INA240 seem idea for our application and I'm at a lost to know why the results you are seeing are not bearings that out.
The way I read the PWM rejection information is that the device can cope with fast changing pulsed signals, in fact the illustration they use in figure 9.3 is even more extreme that our application with current direction being changed as first the upper transistor conducts then the lower one.
Three suggestion:
Firstly my old favourite, make sure there is a good solid common ground, a star configuration would be best.
Secondly relocate the PWM modulator in the test circuit so it sits between the INA240 and the load, this more closely replicates our intended application where we want to measure the current drawn from the PSU. The existing configuration should work perfectly well but as I'm clutching at straws I have to suggest it.
Lastly see what difference tying the Reference signals to the INA240's Vs (+ve supply).
Penri
Re: Over-current Protection Circuit
I've been using the ground plane on the INA240 for this. Are you suggesting that all 0 V lines should go to a single pin?
OK. I'll try that once I've fully understood your other points. I can't quite see what difference it would make though.Penri wrote: ↑06/02/2022, 18:38Secondly relocate the PWM modulator in the test circuit so it sits between the INA240 and the load, this more closely replicates our intended application where we want to measure the current drawn from the PSU. The existing configuration should work perfectly well but as I'm clutching at straws I have to suggest it.
Sorry. I don't understand this. What do you mean by 'eference signals'?
Terry
Re: Over-current Protection Circuit
Hello
Taking each in turn.
1. Ideally I'm suggesting that each 0V be connected to Earth via it's own conductor
2. Moving the PWM modulator should not make a difference to the way the INA240 circuit works but I'm clutching at straws. Just from interest what do you see, voltage wise at the output of the 15V PSU when the PWM is in operation?
3. The INA240 has two reference inputs which can be tied to difference levels the evaluation board allows you to set these to 0V, Vs and 1/2Vs via links on board. My reasoning is if the 0V is noisy then using a different reference value may make a difference (although that could be noisy too), section 8.4 of the data sheet gives more detail.
Penri
Taking each in turn.
1. Ideally I'm suggesting that each 0V be connected to Earth via it's own conductor
2. Moving the PWM modulator should not make a difference to the way the INA240 circuit works but I'm clutching at straws. Just from interest what do you see, voltage wise at the output of the 15V PSU when the PWM is in operation?
3. The INA240 has two reference inputs which can be tied to difference levels the evaluation board allows you to set these to 0V, Vs and 1/2Vs via links on board. My reasoning is if the 0V is noisy then using a different reference value may make a difference (although that could be noisy too), section 8.4 of the data sheet gives more detail.
Penri
Re: Over-current Protection Circuit
Penri,Penri wrote: ↑07/02/2022, 13:56 1. Ideally I'm suggesting that each 0V be connected to Earth via it's own conductor
2. Moving the PWM modulator should not make a difference to the way the INA240 circuit works but I'm clutching at straws. Just from interest what do you see, voltage wise at the output of the 15V PSU when the PWM is in operation?
3. The INA240 has two reference inputs which can be tied to difference levels the evaluation board allows you to set these to 0V, Vs and 1/2Vs via links on board. My reasoning is if the 0V is noisy then using a different reference value may make a difference (although that could be noisy too), section 8.4 of the data sheet gives more detail.
I've been busy with the Lady Hanham stuff over the past few days, and my workbench is therefore cluttered with Gate Valves etc. I will try your suggestions in due course. However, I'm not sure that any of this will help because the output that I get is so solid. If there was lots of noise present I would also assume problems with earthing etc. What I actually get is a DC level at half Vs with PWM pulses swinging up to Vs superimposed on it. I would expect to get the DC level of half Vs because without anything connected to the Ref inputs, that is what happens by default. It's the presence of the PWM that's the issue.
I have discovered that Texas Instruments maintain a Forum and have posted a query (https://e2e.ti.com/support/amplifiers-g ... n-the-spec). I'll wait for their response before doing anything else.
Terry
Re: Over-current Protection Circuit
Penri,TerryJC wrote: ↑09/02/2022, 12:35I have discovered that Texas Instruments maintain a Forum and have posted a query (https://e2e.ti.com/support/amplifiers-g ... n-the-spec). I'll wait for their response before doing anything else.
I've had two responses to my question in the above Forum, but I can't say that I understand either of them. It would appear that the INA240 only rejects common mode PWM signals; both responses say that. I have to say that I don't understand the point if that's all they are doing. When would the PWM signal be common mode and even if it was, why would we want to reject it?
I have to say that I find the spec for this device really misleading, although once you know that the device only rejects common mode PWM signals, you can see it in the text.
Can you shed any light on this? I'll be replying to the Forum posts tomorrow, but it would be useful to have your input first if you have time.
Terry
Re: Over-current Protection Circuit
Terry
It would probably be easier to discuss this face to face but here goes; please excuse me if what I write is too simplistic, I don't have specialist knowledge of PWM signals so am going back to basics as I understand them.
Common mode rejection in a differential input circuit is where the circuit is designed to ignore signal features that appear identically at both inputs, so taking an ideal unity gain op.amp. circuit as an example: if the signal at the +ve input is a sine wave oscillating between 0V and 1V and the signal at the -ve i/p is also a sine wave perfectly in phase (and at the same frequency) with that at the +ve input but this time oscillating between 0.5V and 1.5V the output of the op.amp. would be a dc signal of -0.5V. The circuit is "ignoring" the sine wave and only reporting the differences between the signals, the 0.5V shift.
So now thinking about a PWM signal and the INA240 circuit, each end of the sense resistor will see the "same" pulsed waveform but one end will be shifted voltage wise in relation to the other by the voltage drop across the resistor, which is equivalent to the current flowing through it, the INA240's o/p should be equivalent to this voltage drop (depending on circuit gain).
I can see that sharp edged pulses would need more attention verses low frequency sine waves but what the specifics are I just don't know.
Does this help in any way?
As we discussed earlier, if the physical circuit is as per the schematic then it should work as advertised, that's also what the responders to your post say as well.
Penri
It would probably be easier to discuss this face to face but here goes; please excuse me if what I write is too simplistic, I don't have specialist knowledge of PWM signals so am going back to basics as I understand them.
Common mode rejection in a differential input circuit is where the circuit is designed to ignore signal features that appear identically at both inputs, so taking an ideal unity gain op.amp. circuit as an example: if the signal at the +ve input is a sine wave oscillating between 0V and 1V and the signal at the -ve i/p is also a sine wave perfectly in phase (and at the same frequency) with that at the +ve input but this time oscillating between 0.5V and 1.5V the output of the op.amp. would be a dc signal of -0.5V. The circuit is "ignoring" the sine wave and only reporting the differences between the signals, the 0.5V shift.
So now thinking about a PWM signal and the INA240 circuit, each end of the sense resistor will see the "same" pulsed waveform but one end will be shifted voltage wise in relation to the other by the voltage drop across the resistor, which is equivalent to the current flowing through it, the INA240's o/p should be equivalent to this voltage drop (depending on circuit gain).
I can see that sharp edged pulses would need more attention verses low frequency sine waves but what the specifics are I just don't know.
Does this help in any way?
As we discussed earlier, if the physical circuit is as per the schematic then it should work as advertised, that's also what the responders to your post say as well.
Penri
Re: Over-current Protection Circuit
Penri,
I did understand what common mode rejection was, but I never saw any reason to suppose that this would be a problem that the designers of the INA240 needed to solve. That is why I interpreted 'PWM Rejection' in the spec as removing the PWM from the output signal leaving just the DC equivalent. I suppose I assumed that the functional block labelled 'PWM Rejection' was an integrator, turning the pulses into the DC equivalent before amplification by the Sense Amplifier.
What I still don't understand is where the common mode signals come from in a typical current sensing circuit. It's clearly a problem, but I don't know why.
Would it be possible to talk to you on the phone later or maybe face-to-face at WMT?
I did understand what common mode rejection was, but I never saw any reason to suppose that this would be a problem that the designers of the INA240 needed to solve. That is why I interpreted 'PWM Rejection' in the spec as removing the PWM from the output signal leaving just the DC equivalent. I suppose I assumed that the functional block labelled 'PWM Rejection' was an integrator, turning the pulses into the DC equivalent before amplification by the Sense Amplifier.
What I still don't understand is where the common mode signals come from in a typical current sensing circuit. It's clearly a problem, but I don't know why.
Would it be possible to talk to you on the phone later or maybe face-to-face at WMT?
Terry
Re: Over-current Protection Circuit
Terry
I have a fairly busy morning this morning at WMT but we could talk a little later this afternoon, at around 16:00 onwards, I’ll call you if that’s ok.
Penri
I have a fairly busy morning this morning at WMT but we could talk a little later this afternoon, at around 16:00 onwards, I’ll call you if that’s ok.
Penri