DC Dummy load problem

[zitman]

Hi all,

I have just built a bench power supply and thought I would build Dave's dummy load to test it.  I have bread boarded the dummy load circuit and it is not working as I anticipated.  It seems to draw a maximum of about 50mA and nothing else regardless of where I set the pot.



The bold figures are the voltages I am measuring.  When I adjust the pot the voltages at Pins 1, 2, 3 & 5 all change but the voltage at pin 6 remains roughly the same and therefore the voltage across the 1 ohm resistor doesn't appreciably change.

Any thoughts on what I am doing wrong?

Z

[kony]

Output voltage swing of second opamp (IC2B) is not large enough to open power mosfet. Look in its datasheet and search for gate treshold voltage (4V worst case, BTW !), also look on on-resistance characteristic graph.
The simpliest solution is to increase voltage powering opamps - but beware, they are not rail-to-rail input nor output, and so not sutiable for single supply operation in this case.

[alm]

Seems to me that IC2B's output is saturated against the positive rail.  The output voltage range of the 324 does not include the positive rail. I don't have Dave's design memorized, but I guess he either used a higher supply voltage, an opamp with rail-to-rail outputs or a logic level MOSFET.

[alxnik]

Seems to me that IC2B's output is saturated against the positive rail.  The output voltage range of the 324 does not include the positive rail. I don't have Dave's design memorized, but I guess he either used a higher supply voltage, an opamp with rail-to-rail outputs or a logic level MOSFET.

He used a logic level Mosfet.

@zitman

Easiest solution to experimentally tune the design is to turn the pot to the max and increase the supply voltage until you are happy with the max amperage (2Amps for dave's design). Careful with heat dissipation.

[c4757p]

He used a logic level Mosfet.

This may still be the problem. MTP3055V has a maximum gate threshold of 4V, which an LM324 powered by 5V probably can't hit. You need MTP3055VL, which has a maximum threshold of 2V. Which variant do you have?

[kxenos]

If you check the op-amp datasheet you will see that the maximum output voltage is typically Vcc - 1,5V.

[Mr Smiley]

Just power it with a pp3 or summat greater than 9v 

 

[zitman]

Thanks all,

That has sorted it I was using an LM324AN and MTP3055A.

I have replaced the 5v supply with a 9v battery and all works great.

I have run it up to 250mA.  Just need to sort a decent heatsink before I crank things up.

Looking at the data sheet for the LM324 I am unsure which value would have told me I needed more than 5V, can anyone point me at the relevant value so I can bear this in mind next time!

Again thanks everyone, I now have a DC dummy load.  Just need to get it off the breadboard and build it for real with a decent heat sink.

Cheers

Z

[M. András]

Thanks all,

That has sorted it I was using an LM324AN and MTP3055A.

I have replaced the 5v supply with a 9v battery and all works great.

I have run it up to 250mA.  Just need to sort a decent heatsink before I crank things up.

Looking at the data sheet for the LM324 I am unsure which value would have told me I needed more than 5V, can anyone point me at the relevant value so I can bear this in mind next time!

Again thanks everyone, I now have a DC dummy load.  Just need to get it off the breadboard and build it for real with a decent heat sink.

Cheers

Z

http://hu.farnell.com/h-s-marston/890sp-03000-a-100/heat-sink-300mm/dp/4106015 something like this would do?

[ahnuts72]

Thanks all,

That has sorted it I was using an LM324AN and MTP3055A.

I have replaced the 5v supply with a 9v battery and all works great.

I have run it up to 250mA.  Just need to sort a decent heatsink before I crank things up.

Looking at the data sheet for the LM324 I am unsure which value would have told me I needed more than 5V, can anyone point me at the relevant value so I can bear this in mind next time!

Again thanks everyone, I now have a DC dummy load.  Just need to get it off the breadboard and build it for real with a decent heat sink.

Cheers

Z

The lm324 wasn't the reason you needed more than 5 volts the transistor was.
The lm324 isnt a rail to rail opamp so it can only go as high on the output as supply voltage -1.5.
So you would only be able to get 3.5 volts or so out when you changed to 9 volts supply you would be able to get
7.5 volts or so out.

The transistor you where using can take up to 4 volts on the gate to turn it fully on thats why it worked after changing to 9 volts supply.

Staying at 5 volts and changing to a logic level mosfet like rfp15n05l would also have worked as the gate voltage to full on is around 2 volts max if I remember correctly.

[zitman]

@ahnuts72

OK, I think I am beginning to understand now.

One more question then, on the LM324 data sheet is it the Input Common Mode Voltage Range I should have looked at which states for a supply voltage of 5V it is the Supply voltage -1.5V?

@M. András
That is a very nice heatsink but a little out of my price range at £65.  I have some aluminium block and a milling machine so time to make some swarf I think.

Thanks

Z

[M. András]

@ahnuts72

OK, I think I am beginning to understand now.

One more question then, on the LM324 data sheet is it the Input Common Mode Voltage Range I should have looked at which states for a supply voltage of 5V it is the Supply voltage -1.5V?

@M. András
That is a very nice heatsink but a little out of my price range at £65.  I have some aluminium block and a milling machine so time to make some swarf I think.

Thanks

Z

do a little search on this forum and you will find someone who sells it at good price hint 300mm heatsink

[ahnuts72]

The lm324 wasn't the reason you needed more than 5 volts the transistor was.
The lm324 isnt a rail to rail opamp so it can only go as high on the output as supply voltage -1.5.
So you would only be able to get 3.5 volts or so out when you changed to 9 volts supply you would be able to get
7.5 volts or so out.

The transistor you where using can take up to 4 volts on the gate to turn it fully on thats why it worked after changing to 9 volts supply.

Staying at 5 volts and changing to a logic level mosfet like rfp15n05l would also have worked as the gate voltage to full on is around 2 volts max if I remember correctly.

He has a 1R CS resistor so even 1A will subtract 1Vgs. On a 5V rail this would leave him with 2.5Vgs. Depends how much current he wants out I guess?

OP try searching the forum countless dummy load posts going over every possible configuration.

Mine is setup pretty much just like his at 5 volts supply and using the mosfet I listed I can current limit my supplies at 1.4 amps which is the maximum I can go on my power supplies so don't know if it will go higher.

[kxenos]

Thanks all,

That has sorted it I was using an LM324AN and MTP3055A.

I have replaced the 5v supply with a 9v battery and all works great.

I have run it up to 250mA.  Just need to sort a decent heatsink before I crank things up.

Looking at the data sheet for the LM324 I am unsure which value would have told me I needed more than 5V, can anyone point me at the relevant value so I can bear this in mind next time!

Again thanks everyone, I now have a DC dummy load.  Just need to get it off the breadboard and build it for real with a decent heat sink.

Cheers

Z

Consider using an old CPU cooler. You can easily get less than 1C/W.

[ahnuts72]

Consider using an old CPU cooler. You can easily get less than 1C/W.

Yes its a big difference between active and passive cooling.
Using a fan will let you get away with a smaller heatsink.
I have a few pics of how mine is setup with active cooling on a smaller heatsink.
https://www.eevblog.com/forum/index.php?topic=15755.0
Read that post it will show you alot of the pitfalls I fell into that everyone helped me to see and correct on my final design.

Sent from my Nook HD

[Galaxyrise]

One more question then, on the LM324 data sheet is it the Input Common Mode Voltage Range I should have looked at which states for a supply voltage of 5V it is the Supply voltage -1.5V?

No, Input Common Mode Voltage is the voltage between In+ and GND and between In- and GND.  The datasheet parameter affecting how high the output can go is Voh, "High-level output voltage".

I have run it up to 250mA.  Just need to sort a decent heatsink before I crank things up.

Careful of the temperature of your sense resistance, too. Especially since resistors change resistance with temperature.

[Thilo78]

Digging up this old topic, but I don't think it's reasonable to make a new topic for this:

I built this dummy load as well, using a 9k/9k divider for the set voltage at the second opamp, and a IRFZ44N n-MOSFET.
My load consists of 10 resistors 10R/2W in parallel, which add up to about 1.2R.

First, I ran into the same problem with supply voltage and thus switched to a 9V battery supply.
With approx. 8 V available, the dummy load works fine and is capable of more than 1.5 A (that's the limit of my homebrew bench PSU).
I just added a fan to the flimsy aluminium heat sink that I had in my parts box.
With that I'm quite stable at 70°C at 1.25 A. Without fan I'm quickly exceeding 120°C at the MOSFET case.

Note: With the fan in parallel, I have an operating voltage of ~7.8 V DC.

But:
I wondered if the dummy load would regulate the current with variations in the tested power supply.
It does not.
If I dial in a current of 1.0 A at 10 V DC, and crank up the PSU to 15 V DC, the current rises to about 1.2~1.25 A.
(current measured with an inline multimeter in 2A range and another multimeter measuring the voltage across the load resistor)

I couldn't yet figure out, why that's happening.
It's not like the MOSFET was just acting as an adjustable resistor, but the current is not really as stable as I would have expected.

I was thinking, that maybe the power supply is too low at 7.8 V DC, but I have no second bench PSU to take that part.
So I'm trying to figure out the theory first.

Any ideas?
Or even better, advice?

Thanks!

Thilo

[Galaxyrise]

With that I'm quite stable at 70°C at 1.25 A. Without fan I'm quickly exceeding 120°C at the MOSFET case.

A note: the temperature is a function of power, not current.  Power is voltage times current, and this case it's voltage drop across the MOSFET.  If your supply is at 15V and there's 1.5V across the sense resistor, that leaves 13.5V across the MOSFET meaning it's dissipating almost 17W.  If your supply was only at 10V, then the MOSFET was only dissipating 10.6W.  Figure the temperature goes up linearly with power dissipated.

If I dial in a current of 1.0 A at 10 V DC, and crank up the PSU to 15 V DC, the current rises to about 1.2~1.25 A.
(current measured with an inline multimeter in 2A range and another multimeter measuring the voltage across the load resistor)
...
I was thinking, that maybe the power supply is too low at 7.8 V DC

A supply voltage of 7.8 will probably get you around 6V output, which will then be 4.5V Vgs at 1.25A.  That's pretty close to the gate threshold in the datasheet, but increasing the voltage on the drain would typically require a lower gate voltage to get the same current, so opamp supply voltage isn't my first guess.

Watch the opamp inputs as you turn up the supply voltage (make your measurements at the opamp itself, both positive and negative probes.)  If they stay matched, then the control loop is working and you have enough supply voltage for the opamp.  Perhaps the sense resistance is dropping due to heat, your opamp's ground voltage is rising, the set voltage is changing, or something else.  If you have an oscilloscope, check for oscillation across the sense resistor.

You've got a "good" problem here; I'm sure you'll learn from troubleshooting it

[Thilo78]

A note: the temperature is a function of power, not current.  Power is voltage times current, and this case it's voltage drop across the MOSFET.  If your supply is at 15V and there's 1.5V across the sense resistor, that leaves 13.5V across the MOSFET meaning it's dissipating almost 17W.  If your supply was only at 10V, then the MOSFET was only dissipating 10.6W.  Figure the temperature goes up linearly with power dissipated.

Yes, that's quite obvious.
It was interesting to see, that my flimsy aluminium heatsink would actually get away with close to 30 W (1.4 A at 15 V supply or 13 odd V over the MOSFET) of heat dissipated, although it would become really hot.
The next surprise was the effect of the small fan.

btw: Pictures are here:
https://www.eevblog.com/forum/beginners/dummy-load-questions-regulation-characteristics-heat-sink/

I opened a new thread because I felt like capturing your topic 
Stil, I captured your schematic. Hope you don't mind...

If you have an oscilloscope, check for oscillation across the sense resistor.

That was my problem.
All of the regulation part worked like a charm, but I ended up with serious oscillation which basically switched between full current and zero.
Adding the filtering parts as advised by TerminalJack505 solved the problem here.

btw: supplying from a 9V battery works fine. I never had the opamp hit its upper limit.

Next thing is to find a case and maybe double up the MOSFET at a later stage 
And of course: Finding a nice tool for schematics. Don't like Fritzing too much and just started with Eagle.
Let's see where this will take me.

Thilo

[idpromnut]

@Thil78: What is the power rating of your current sense resistor? If you are exceeding it, it will probably be heating up, which will increase its resistance, there for dropping a higher voltage and your op-amp will be happily attempting to set the current to match it. 10V @ 1A is about 10W, which is a lot for a non-wirewound, non-power resistor. Heck even the wire wound ones that I have that are 10W start to drift when they are dissipating close to their maximum rated power.

[Thilo78]

Temp drift is not much of a problem here.
I use 10 resistors in parallel; each is rated for 2W.
So with 1.5 A over 10 resistors they're a fair bit too solid but the array looks great. 

The most power is dissipated in the MOSFET, which drops about 14 V and carries up to 1.5 A,
So here we have the ~20 W that I sink into the dummy load.

The thermal energy in the resistor is close to nothing.

[idpromnut]

Temp drift is not much of a problem here.
I use 10 resistors in parallel; each is rated for 2W.
So with 1.5 A over 10 resistors they're a fair bit too solid but the array looks great. 

The most power is dissipated in the MOSFET, which drops about 14 V and carries up to 1.5 A,
So here we have the ~20 W that I sink into the dummy load.

The thermal energy in the resistor is close to nothing.

Bah, nevermind, I just realized that it's a 1V drop over the resistors, so no, unless they're really horrible they would only be dropping a watt or 2.