If you are using a very low value shunt resistor, you would be best advised to handle it as a four-terminal device (at least in terms of track layout, whether or not the resistor has separate current and potential connections). You would therefore need to use either two A/D inputs to your device, or else use a differential buffer amplifier.

If you try to use a simple resistor of just tens of milliohms, with significant wiring or track resistances in series, you may find the voltage is noticeably in error, and this error will move around with temperature because the temperature coefficient of resistance of copper is relatively big.

Oh you are right. 0.1 Ohm should still be safe, but when I move to higher current I will need lower resistance where track resistance could be a problem. There are some high power MOSFETs that have a package design for cable wiring, so the power part of the circuit can all be wired.

 

What you seem to be developing is a typical mosfet driver/controller with pulse by pulse current limiting. I believe those may still be available as a single chip. They use comparators to sense an over current situation. You could use one or two comparators to sense your shunt, depending on whether you want control or just a limit. Then generate a control PWM followed by D/A converter (low-pass filter) for the comparison voltage or use a digital potentiometer, if on the low side, for that voltage. That will avoid the potential A/D issue with the PWM.

John

Edit: Look up the TPIC2101. It may still be made. http://pdf1.alldatasheet.com/datasheet-pdf/view/29139/TI/TPIC2101.html

Not sure if this is the case. I am trying to apply a specific voltage to obtain a certain current through my load, without energy losses.

Now, if indeed that's what the TPIC2101 or similar devices do, I still doubt they can meet my power requirements for the 50A version. (Even 4A is already a bit of a load).

Nonetheless, I will take a look, thank you

 

Not sure if this is the case. I am trying to apply a specific voltage to obtain a certain current through my load, without energy losses.

Now, if indeed that's what the TPIC2101 or similar devices do, I still doubt they can meet my power requirements for the 50A version. (Even 4A is already a bit of a load).

Nonetheless, I will take a look, thank you

The TPIC will handle a motor controller well in excess of 50A. I have one running that is probably on the order of 200A or more for short periods.

John

 

Oh you are right. 0.1 Ohm should still be safe, but when I move to higher current I will need lower resistance where track resistance could be a problem.

When dealing with currents as high as you're dealing with I think it would be prudent to use a commercial shunt or make one similar to a commercial model. This would exclude a shunt integral to a PCB.

 

In general, yes - a commercial shunt would make it far easier to get good results. It is not entirely impossible to arrange for monitoring tens of Amperes on a PCB, but it would require skills and resources which may not be available to a student.

Actually, I would not be too blasé about using a 100milli-ohm resistance without paying close attention to the layout, or using a 4-lead resistor. Just a few milli-ohms of stray common resistance would degrade the accuracy.

 

Yes absolutely, I think a commercial shunt is a must for the 50A+ models.

Now, I must say that for this application accuracy is not that important, we could have 10% error and still be ok. Of course accuracy will ultimately impact cost of operation so the closer to optimal, the cheaper it will be to operate.

 

The TPIC will handle a motor controller well in excess of 50A. I have one running that is probably on the order of 200A or more for short periods.

John

Oh that sounds good. I may just be able to find a supplier and run some tests.

 

or do away with the shunt and put a current transformer on your line.

 

Now, if indeed that's what the TPIC2101 or similar devices do, I still doubt they can meet my power requirements for the 50A version. (Even 4A is already a bit of a load).

Nonetheless, I will take a look, thank you

You do realize that the TPIC2101 is a controller not a power module. It will require an external pass element.

 

or do away with the shunt and put a current transformer on your line.

Hmmm, that may actually be a better idea. Thanks for the suggestion.

I need to think a bit about the signal conditioning coming off the transformer, but this will let me isolate the power circuit from the control circuit.

 

Have you considered a current sense mosfet like this:BUK7905-40AIE (http://www.nxp.com/documents/data_sheet/BUK7905-40AIE.pdf)

I am curious, what is this project for?

John

 

A classic current transformer will not give you a DC response, if this is required. Some devices which resemble current transformers will do so (e.g. Hall Effect or Fluxgate transducers).

Do you need to capture the DC term, or not?

 

You do realize that the TPIC2101 is a controller not a power module. It will require an external pass element.

Yes read the specs-sheet and it is definitely not what I am looking for. My microprocessor is not only doing this control part but a bunch of other rather complex computations, plus some UI.

Have you considered a current sense mosfet like this:BUK7905-40AIE (http://www.nxp.com/documents/data_sheet/BUK7905-40AIE.pdf)

I am curious, what is this project for?

John

I have not. How exactly do they work? I need to be able to control the current through the system by altering the voltage (PWM). This current is computed in Amps in the microprocessor and is dependent on other parameters outside of the circuit (hence the need for the microcontroller).

The application is research into electrolytic degradation of organic compounds.

A classic current transformer will not give you a DC response, if this is required. Some devices which resemble current transformers will do so (e.g. Hall Effect or Fluxgate transducers).

Do you need to capture the DC term, or not?

I just need the current through the load, independent of whether it is pulsed (PWM) or DC. My controlling parameter is actually charge in Coulombs, so that translates to Amp*seconds. The current transformer should be fine if I work with the unfiltered PWM signal. But yes if I were to filter there is always the Hall effect/magnetic flux devices.

 

Now I am thinking about power consumption.

Say I start with 20V, and a load resistance that gives me 2A.

If I need to deliver an average of 1A I would need an avg voltage of 10V. To do that with PWM I would have to use a 50% Duty cycle (half time ON, half time OFF).

If I power the load for 1 hour, that's a real ON-time of 0.5h, and my total energy consumption is

E = VIt = 20V*2A*0.5h = 20Wh

But if I had used a power supply that delivers 10V directly, to get my 1A, I would have used

10V*1A*1h = 10Wh.

So the PWM version consumes twice as much energy for this example. Is this right, or am I missing something?

In other words, since Energy is the product of Voltage, Current and Time, and the Resistance is fixed, when I double the current I need to double the voltage, so energy increases at a square rate, while the PWM just reduces ON-time at a linear rate.

 

So the PWM version consumes twice as much energy for this example. Is this right, or am I missing something?
[/QUOTE]

Huh? If your duty cycle is reduced to 25% you would be back at 10WH.

Edit: See post 51

 

Huh? If your duty cycle is reduced to 25% you would be back at 10WH.

But a 25% duty cycle does not give me 1Amp average, it gives me 0.5A

 

But a 25% duty cycle does not give me 1Amp average, it gives me 0.5A

Yes. but your question was in reference to Watts and Watt Hours, which is really what's important to you.

 

In other words, since Energy is the product of Voltage, Current and Time, and the Resistance is fixed, when I double the current I need to double the voltage, so energy increases at a square rate, while the PWM just reduces ON-time at a linear rate.

\(P = {I}^{2}R\) or \({V}^{2}/R\) yes?

 

Yes. but your question was in reference to Watts and Watt Hours, which is really what's important to you.

No, current is what I am trying to control. But it seems at the expense of higher energy expenditure.

\(P = {I}^{2}R\) or \({V}^{2}/R\) yes?

Right, that just what I implied. ( V=IR and P = VI, therefore P = I^2 R )

So I am doomed to spend much more energy if I try to regulate current this way?

 

your striking off on a tangent.

Your goal was to control current over a fixed resistor. That will generate a certain power that must be dissipated. You can choose to generate that heat in a continuous fashion, or you can pulse it in PWM fashion as twice the heat, half the time.

You've jumped back to the PWM thing when it was explained that the PWM was filtered to linearly run an amplifier.

Do you need PWM current or do you need linear current?