Yet another DIY Electronic Load

[pinoccio]

Btw. for me ebay is the cheapest source of components, so that is the price reference. Checked today- INA286 starts at GBP 5.19 / US $6.54 with free shipping. Mouser (or other big suppliers) might be at 2,54 € / $2.92, but adding 20€ / $30.00 shipping is unacceptable.
And we are slightly drifting away from a simple circuit too...
I'd prefer staying with the current concept for now.

Regarding precision/resolution, for a hobbyist project, the possibility to step current by 1mA is totally sufficient IMHO. Don't even need to go so close to 0, 30 is fine I guess, but about 10mA would be nice if it can be achieved reasonably easily. Don't know if that is possible with op amps.

Taking another look, can't we just omit that current sensing op amp completely and feed directly the ADC module (2 inputs are available) and sum in SW?    That would also get us back to single supply, right?.

As I already mentioned, I can go with those cheaper IXYS mosfets. If there are no objections, like they're not logic level, that could become the final version (regarding power transistors).

[JeanLeMotan]

jean, your math is still off  he is passing a maximum of 2.5A per branch (i.e mosfet). each mosfet has 2 resistors in parallel, so 1.25A per resistor. using ohms law we have P= I²*R = 1.25²*1 = 1.5625 watts 

  Forgot there are 2 branches.
In any case 1.56W per resistor is still a lot for the one you recommended and its temp will rise by about 100°C even with the 4527 package.

As I said - I'm using a 20W 0.1Ω shunt split across 4x 4527 SMD resistors and for 5A I had big issues dissipating that heat. No way I could do 5A continuous, so I dropped the max to 4A and it's still getting hot.
Here you're proposing a 0.25Ω shunt (2 resistors paralleled per branch, 2 branches paralleled) for the same 5A load - so 2.5 times more power dissipated than in my load. That will get incredibly hot!

Once you include derating due to temperature & continuous operation SMDs are just not worth it for shunts at these currents.
My advice is to go with beefier TH. Or add way more SMD series/parallel resistors to spread the heat and get a bigger surface area. Or add little cute heatsinks for each SMD resistor.

[edit] Here's the thread for my load: https://www.eevblog.com/forum/projects/jlm-electronic-load-0-30v-0-4a-1ma1ma-resolution/

[Yansi]

Do not use SMD resistors for power dissipation, unless you really must.  There is no way to dissipate much past one or two watts to a PCB, always use through hole stuff for power. Will also save you same board space and heat spreading to surrounding components.

5A over 0.1ohm, thats 2.5W which should be sort-of fine even for 3W wirewound THT resistor (when a bit of airflow is present, better to get a 4W one though). Beware, these suckers can get nasty hot, I think their rated power is getting the surface temperature sometimes well past solder melting point.

And also, I do not recommend using those plastic resin encapsulated resistors. They can't dissipate heat well. Nothing beats a proper wirewound ceramic one (unless struggling for very low inductance).

[pinoccio]

You can calm down guys  ...I never planned to use smd resistors for the shunt in the first place.
Most probably I will go with an overkill of 10 or even 25W wirewound in aluminium package each! And they will end up screwed to the heatsink too. At a price below $1/pc (the bigger ones are actually cheaper), I don't need to bother with derating and temperature issues here

[Yansi]

Wait a minute... what is your current goal?  5 or 10 amps?  I think it is quite unnecessary (and might be even contraproductive) to use such overrate shunt resistor.

It should be possible to design a current sense circuitry requiring sense voltages such low, that the overall heat dissipated won't be over 3W - which is manageable using a single THT resistor of a proper sizing.

If you want to go really precision, I'd recommend using 4-wire shunt resistors, such as from Isabellenhütte (PBV-0R01). These can be also heatsinked.

It depends on your goals - if precision is the goal or simplicity and cheapness perhaps? If you do not care about transient response behavior and loop bandwidth, then you may even use those overpriced high-side sense amplifiers - but I do not see a single reason for them. Noone is using them in DC loads (for good reasons).

[OM222O]

If you do not care about transient response behavior and loop bandwidth, then you may even use those overpriced high-side sense amplifiers - but I do not see a single reason for them. Noone is using them in DC loads (for good reasons).

I love how you call them overpriced  you clearly haven't used any chips over 2$ in any projects. some voltage refrences or even resistors can easily cost more than 10$ depending on the application. most commercial products use much more expensive differential amplifiers like I mentioned above, costing more than 5$  for example the schematic provided by john for the BK 8500. it uses the LT1013 which is about 6.6GBP from mouser  the *over priced* INA series are actually cheaper alternatives with less accuracy, but still good enough for this application 

here is a much better version of the design:

https://easyeda.com/theepicn008/electronic-load-improved
-I switched over to low side sensing (100mV/A). that is quite good but I would personally add some extra parts to improve the accuracy in the mA range. it would be a bit more complicated and you mentioned keeping things simple so I didn't do it. it requires a fet and another resistor in the feedback path of U5 to adjust the gain from x10 to x100.
-voltage sense circuit got a few improvements
-accuracy is improved by at least a factor of 10 due to much better op amps (2uV maximum offset!)
-some general clean up of the schematic

Some notes:
I changed the name of the nets: IO0 to IO3 are connected to the ADC (actually there is no IO3, I replaced it with the setpoint). DAC output is IO4.
External input / input switch had the 1k resistor in the wrong place.
I switched to 1206 for the most part. change them back to axial if you want to use those.
You don't need sil pads between the fets and the heat sink, since their drains are electrically connected anyways. if you are connecting other things such as voltage regulators, etc. be sure to use sil pads for those.
The IXYS fets would be fine, you don't have to use logic level fets. the IXTH75N15 can pass about 10A with a Vgs of 5V at around 2V Vds. it seems like a reasonable part if you can keep the case cold enough. they only specify it for case temp of 25c so I'm not sure about derating at higher temps.

with the parts you have, I'm not sure you could have gotten down to 1mA accuracy (not even 1mA resolution). It's because of your dac. with a 5V input, the resolution is about 1.2mV, which means a resolution of 1.2mA for your load 2*(0.0012mv /4)/0.5\$\Omega\$ . that is also assuming the DAC is perfect, which it's not.it has +-2 LSB INL and some DNL (I forgot sorry) which means you can be off by 4 codes, which is 4x1.2mA = almost 5mA. I think that's still reasonable for you, but let me know if you want to improve that.

this time I will be called out by Yansi for *My bullshit* and for *Use of overpriced op amps*  ... again, I'm not paid by maxim to promote their chips. you can easily buy the max chips from RS (don't use ebay for parts ... they are often clones or salvaged parts, hence the cheap prices). here is an screenshot in case I'm lying about the prices again 



also for resistors, There are some smallish 15 watt resistors that would be perfect for this application:
500m\$\Omega\$ 15 watt:https://uk.rs-online.com/web/p/panel-mount-fixed-resistors/0159714/
1\$\Omega\$ 15 watt:https://uk.rs-online.com/web/p/panel-mount-fixed-resistors/7547676/

I would personally go with the 500m\$\Omega\$ since it ends up being cheaper.
This is as simple as I could make the circuit without adding any special features. let me know if its still too complex or if you want additional features such as multiple current ranges etc.

[Yansi]

If you do not care about transient response behavior and loop bandwidth, then you may even use those overpriced high-side sense amplifiers - but I do not see a single reason for them. Noone is using them in DC loads (for good reasons).

I love how you call them overpriced  you clearly haven't used any chips over 2$ in any projects. some voltage refrences or even resistors can easily cost more than 10$ depending on the application. most commercial products use much more expensive differential amplifiers like I mentioned above, costing more than 5$  for example the schematic provided by john for the BK 8500. it uses the LT1013 which is about 6.6GBP from mouser  the *over priced* INA series are actually cheaper alternatives with less accuracy, but still good enough for this application 

Jeesus crist. The OP can't tell a difference between LM339 and 324 (nothing wrong with that, he'll learn) and you still trying to push overpriced crap. Yes, for the purpose here, it is overpriced.

Get the OP to do his first steps with cheap jellybean parts that get the work done quicker and cheaper, instead of insisting on buying completely unnecessary components.

[OM222O]

OP wants something useful at the end, besides if you look at the schematic there is no INA part. I used max4238 for every part.
If you suggest building the circuit with cheap trash, the end result will be cheap trash. besides "Good practices" are a major part of learning. nobody uses summing amplifiers for current sensing in the industry 

quite honestly I had no idea how to stabilize the current loop until a few weeks ago. every time I asked people just suggested redesign or random value components such as resistors or capacitors between x or y which wasn't helpful at all. Jay_Diddy_B posted some simulations and analysis of how the ideal situation would be and how to set the correct values to get proper phase margin and gain margin, which elements attribute more to the instability, etc. that was extremely helpful and I learned a lot. being a beginner and not knowing something isn't bad. that's why we ask questions here. giving vague / wrong advice and forcing your opinion on others on the other hand, is very bad. unless you have some constructive feedback, don't post aggressive comments such as : " YOU STILL TRYING TO PUSH OVERPRICED CRAP". so far all you have done is to tell everyone they are wrong without giving actual solutions. he provided a public schematic, so if you have any ideas, copy that and make the modifications.

you also contradict yourself by stating:

OM222O, could you please kindly stop pushing your bullshit with those high side current sense amplifiers? They don't even have enough bandwidth to begin with. They are anything but useful for designing a good stable load with good transient characteristic.

if the goal is to make something with jellybean parts, who cares about bandwidth anyways? it's not a high frequency application and in most cases transients are a non issue for a load. 100 to 200Hz BW would be plenty good enough! do you want to over complicate current sensing and increase bandwidth? be my guest!
I'm not sure what your suggestions / solutions are for this issue?

Unfortunately this aggressive comments and strong opinions seem to be getting more common here.
Learning how to do something properly is a lot more important than "just using jellybean parts that get the work done quicker and cheaper". maybe for a one off project that needs to be done ASAP and needs to be in use, getting something dirt cheap to work is a priority, but the OP seems to be more interested in learning how to design a proper load as simple as possible using existing modules.

[markce]

Now wait. This is not going to help Pinoccio.  You're both right in some way, but Pinnoccio is still at the

Btw. for me ebay is the cheapest source of components, so that is the price reference.

stage.
My approach would be to use basic proven components, from a reliable source and a simple schematic to start with. Can't be that difficult?

[Yansi]

OP wants something useful at the end, besides if you look at the schematic there is no INA part. I used max4238 for every part.
If you suggest building the circuit with cheap trash, the end result will be cheap trash. besides "Good practices" are a major part of learning. nobody uses summing amplifiers for current sensing in the industry 

quite honestly I had no idea how to stabilize the current loop until a few weeks ago. every time I asked people just suggested redesign or random value components such as resistors or capacitors between x or y which wasn't helpful at all. Jay_Diddy_B posted some simulations and analysis of how the ideal situation would be and how to set the correct values to get proper phase margin and gain margin, which elements attribute more to the instability, etc. that was extremely helpful and I learned a lot. being a beginner and not knowing something isn't bad. that's why we ask questions here. giving vague / wrong advice and forcing your opinion on others on the other hand, is very bad. unless you have some constructive feedback, don't post aggressive comments such as : " YOU STILL TRYING TO PUSH OVERPRICED CRAP". so far all you have done is to tell everyone they are wrong without giving actual solutions. he provided a public schematic, so if you have any ideas, copy that and make the modifications.

you also contradict yourself by stating:

OM222O, could you please kindly stop pushing your bullshit with those high side current sense amplifiers? They don't even have enough bandwidth to begin with. They are anything but useful for designing a good stable load with good transient characteristic.

if the goal is to make something with jellybean parts, who cares about bandwidth anyways? it's not a high frequency application and in most cases transients are a non issue for a load. 100 to 200Hz BW would be plenty good enough! do you want to over complicate current sensing and increase bandwidth? be my guest!
I'm not sure what your suggestions / solutions are for this issue?

Unfortunately this aggressive comments and strong opinions seem to be getting more common here.
Learning how to do something properly is a lot more important than "just using jellybean parts that get the work done quicker and cheaper". maybe for a one off project that needs to be done ASAP and needs to be in use, getting something dirt cheap to work is a priority, but the OP seems to be more interested in learning how to design a proper load as simple as possible using existing modules.

You are full of nonsense.  There is enough jellybean parts which one can make FAST amplifiers for cheap. Based on what do you think that only expensive stuff is the fast and good one? 

100Hz bandwidth load is not enough. For discharging batteries, maybe.

Now wait. This is not going to help Pinoccio.  You're both right in some way, but Pinnoccio is still at the

Btw. for me ebay is the cheapest source of components, so that is the price reference.

stage.
My approach would be to use basic proven components, from a reliable source and a simple schematic to start with. Can't be that difficult?

I second that. But mind you I am not the one pushing specialized components, that are of not much help to a beginner, other than cleaning his/her wallet.

[OM222O]

Now wait. This is not going to help Pinoccio.  You're both right in some way, but Pinnoccio is still at the

Btw. for me ebay is the cheapest source of components, so that is the price reference.

stage.
My approach would be to use basic proven components, from a reliable source and a simple schematic to start with. Can't be that difficult?

I actually had a look at ebay. MAX4238 goes for a minimum of 5$ there  just wow! the same part on RS Slovakia is I think 1.2 euros. even cheaper on mouser. understandably so, mouser charges shipping for low value orders but I think above 30GBP components, the shipping is free. I recommend he checks all the suppliers before assuming ebay is always the cheapest source.

You are full of nonsense.  There is enough jellybean parts which one can make FAST amplifiers for cheap. Based on what do you think that only expensive stuff is the fast and good one? 

100Hz bandwidth load is not enough. For discharging batteries, maybe.

"Based on what do you think that only expensive stuff is the fast and good one?"
based on the fact that you haven't came up with any ideas that are cheaper than my suggestions and making me seem like a shill for different companies. all of my recommendations are based on ACTUALLY USING THE PARTS in my own projects. my second suggestions was a single op amp with 3 resistors, costing about 2 euros overall. can you be please so kind as to introduce a cheaper solution?

also the use case for a hobbyist is exactly the sort of things such as discharging batteries to get their capacity or loading down a power supply to check their ripple. this is not a professional equipment that requires high bandwidth. again, I recommend you post a proper solution / some calculations / simulations that help the OP rather than calling people names. Have a nice day 

[MarkF]

HOW QUICKLY THIS BASIC CIRCUIT HAS GONE OFF THE RAILS.     
PUT IT BACK IN YOUR PANTS BOYS AND GO TO YOUR CORNER.

ONE MANS JUNK IS ANOTHER MANS TREASURE.

Btw. for me ebay is the cheapest source of components, so that is the price reference. Checked today- INA286 starts at GBP 5.19 / US $6.54 with free shipping. Mouser (or other big suppliers) might be at 2,54 € / $2.92, but adding 20€ / $30.00 shipping is unacceptable.
And we are slightly drifting away from a simple circuit too...
I'd prefer staying with the current concept for now.

Regarding precision/resolution, for a hobbyist project, the possibility to step current by 1mA is totally sufficient IMHO. Don't even need to go so close to 0, 30 is fine I guess, but about 10mA would be nice if it can be achieved reasonably easily. Don't know if that is possible with op amps.

Taking another look, can't we just omit that current sensing op amp completely and feed directly the ADC module (2 inputs are available) and sum in SW?    That would also get us back to single supply, right?.

As I already mentioned, I can go with those cheaper IXYS mosfets. If there are no objections, like they're not logic level, that could become the final version (regarding power transistors).

If you have the inputs, I think monitoring each MOSFET individually is an idea worth investigating. 
However, I would buffer the sense resistors with a voltage follower to your Arduino.

Go back to your max sense voltage of 500mV.  You would have a hard time applying your load to a 3.3V power supply with a 2.5V sense voltage. 
Stick with the 2K and 18K voltage divider for the control input and 0R100 to 0R200 for the sense resistors. 
I think a 150 mOhm 5W sense would be best.  It would give you a max 6A load current capability.  Your target (6A @ 30V) (6A @ 24V)?

I would change to the IXYS MOSFETs you mentioned.  I would look for the larger TO-247 package just for the extra heat dissipation.
I would not be putting my project at risk of getting counterfeit parts by buying the power MOSFETs from eBay. 
Pay the extra to Mouser or Digikey and get a genuine part.

The +5V supply for all your op-amps seems to be a little low to me.  I would use +12V.


Edit-  Totally missed that you will need some gain before sending the sense voltage to the ADC.
         I arbitrarily choose a gain of 5 but you might want to go as high as 10.

Here is my idea:

[Jay_Diddy_B]

Hi,

I don't want to join the discussion on the price the parts. But, I will join the conversation if I see that some of the advice given is wrong.

I will review the circuit posted here:

Link: https://easyeda.com/theepicn008/electronic-load-improved

Let us look at one of MOSFET circuits:



We have an IXYS IXTH75N15 MOSFET been driven by a 5V op-amp. We also have 0.5  source resistor.

If we look at the MOSFET datasheet:



We can see that for 5A we need about 4.5V of gate source voltage.

If we have a drain current of 5A, and therefore a source current of 5A. With a 0.5 source resistor. The source will 2.5V positive with respect to ground. If we add the 4.5V Vgs, the output of the op-amp needs to be 7V.

How can we get to 7V with a 5V supply?


Regards,

Jay_Diddy_B

[Yansi]

I have already warned about insufficient power supply voltage at #38, but it seem it got lost in OM222O's push for exclusive parts.

One way to get away with just 5V supply would be to use a bipolar transistor as the power part.  There are plentiful choices out there for rather cheap, compared to grumpy mosfets, that do not have the DC SOA fully characterized and will blow unexpectedly.

For example: TIP3055, BD249, or if one wants some speed action: 2SC5200, MJL3281 or for a very robust ones: MJL21193, etc...

(PS: avoid 2N3055 at all costs!)

[Yansi]

Just an idea for the BJT solution

[2N3055]

(PS: avoid 2N3055 at all costs!)

What's wrong with me? 
LOL!

Seriously, for this purpose good bipolars are dedicated audio types. They are expensive though..

[Jay_Diddy_B]

Hi,
As well as the load I documented in this thread:

https://www.eevblog.com/forum/projects/dynamic-electronic-load-project/

I have also build a few more. Here are a couple of examples:








This one uses the LT1635, which is a version of Bob Widlar's famous LM10. It is a combination op-amp and 200mV reference.


This one is a quad load. It uses a jumper block for setting the current range of each channel. It is intended for testing multi-rail low voltage power supplies. It use the LT1014. The LT1014 can be replaced by a LM324, but the LM324 will have larger offset voltages.








The schematic shows the common section and one of the four channels.



I have not built one with microprocessor control.

Regards,
Jay_Diddy_B

[Yansi]

(PS: avoid 2N3055 at all costs!)

What's wrong with me? 
LOL!

Seriously, for this purpose good bipolars are dedicated audio types. They are expensive though..

Lol, nothing against you, but against "manufacturers" providing today's substandard 2N3055 transistors, that simply aren't up to spec.

And yes, those MJLs I have suggested are audio amps.

[2N3055]

(PS: avoid 2N3055 at all costs!)

What's wrong with me? 
LOL!

Seriously, for this purpose good bipolars are dedicated audio types. They are expensive though..

Lol, nothing against you, but against "manufacturers" providing today's substandard 2N3055 transistors, that simply aren't up to spec.

And yes, those MJLs I have suggested are audio amps.

Not to mention old ones were diffusion types with massive crystals with massive SOA. Slow but robust. New ones are epitaxial, faster but not robust. They are different enough that sometimes are not good replacement for the old ones when repairing some circuits..

[OM222O]

john is right! I mentioned the fet being able to achieve 10A at 5V Vgs, forgetting that there will be about 1.25 source voltage  Although the current through each fet will be 2.5A, not 5, it's still pushing the limits of the gate voltage. the data sheet isn't really fine enough to tell whether 2.5A is possible at 3.75V Vgs. I also tried downloading the spice model for the IXTH75N15, but there is only a pspice model which I don't have installed. However I attached it, if somebody has it installed, they can run the simulation and report the results.

I also had a look at about 10 to 15 IXYS fets, they have very similar admittance graphs. if this doesn't work, you have to give up on linear fets. hold off on ordering them until someone can confirm 2.5A at 3.75Vgs is possible, or you can also just buy one and test it for yourself if that's within your budget.

Edit: OP initially mentioned a 5A rating, but has been vague about if he meant 5A per fet, or 5A total. 5A per fet will not work with linear fets at all with a 5v supply.
The texts were copied from the original schematic and they indicate a 10A goal, however the resistor divider used for the set points was division by 4 which indicates a 5A total rating  I made some modifications to the text to correct that.

Edit#2: the OPA188 or OPA2188 seem reasonable replacements for the MAX4238 in the load control section. they can go as high as +-18V, but it's ok to power them from 0 and 12V (assuming you have a 12V supply somewhere). They are also available on RS, only slightly more expensive.

[Yansi]

(PS: avoid 2N3055 at all costs!)

What's wrong with me? 
LOL!

Seriously, for this purpose good bipolars are dedicated audio types. They are expensive though..

Lol, nothing against you, but against "manufacturers" providing today's substandard 2N3055 transistors, that simply aren't up to spec.

And yes, those MJLs I have suggested are audio amps.

Not to mention old ones were diffusion types with massive crystals with massive SOA. Slow but robust. New ones are epitaxial, faster but not robust. They are different enough that sometimes are not good replacement for the old ones when repairing some circuits..

You sure about the diffusion type? I have heaps of locally made power transistors similar to 2N3055 that has also those huge dies (4x4 or even 5x5mm), but I think the catalogue I have clearly states the are epitaxial. 

[MarkF]

(PS: avoid 2N3055 at all costs!)

What's wrong with me? 
LOL!

Seriously, for this purpose good bipolars are dedicated audio types. They are expensive though..

Lol, nothing against you, but against "manufacturers" providing today's substandard 2N3055 transistors, that simply aren't up to spec.

And yes, those MJLs I have suggested are audio amps.

Not to mention old ones were diffusion types with massive crystals with massive SOA. Slow but robust. New ones are epitaxial, faster but not robust. They are different enough that sometimes are not good replacement for the old ones when repairing some circuits..

You sure about the diffusion type? I have heaps of locally made power transistors similar to 2N3055 that has also those huge dies (4x4 or even 5x5mm), but I think the catalogue I have clearly states the are epitaxial. 

WOW.   
WE CAN'T EVEN STAY ON TOPIC!

[2N3055]

WOW.   
WE CAN'T EVEN STAY ON TOPIC!

Sorry boss, won't happen again..

Thing is that I guess we are saying that if you want to make a bipolar transistor load good choice are audio transistors. They have good SOA and decent speed. Powerful transistors like 2N3055 are robust enough, but not fast enough to make a good load.

@ yansi I have few very old diffusion type 2N3055H made in YU.. They have better SOA. But any 2N3055 made in last 20-30 years will be EPI.

[MarkF]

WOW.   
WE CAN'T EVEN STAY ON TOPIC!

Sorry boss, won't happen again..

Thing is that I guess we are saying that if you want to make a bipolar transistor load good choice are audio transistors. They have good SOA and decent speed. Powerful transistors like 2N3055 are robust enough, but not fast enough to make a good load.

@ yansi I have few very old diffusion type 2N3055H made in YU.. They have better SOA. But any 2N3055 made in last 20-30 years will be EPI.

Not trying to be boss.
Just frustrated by all this side noise.  And I believe the OP is too.  Scan back through the topic for the last time he posted anything.

I updated my last circuit to add gain to the current sense op-amps to the Arduino.

Why don't we talk about something more relevant, like protection for the MOSFET drive op-amps and the Arduino in case the MOSFETs let out some magic smoke?
My Load doesn't have a MCU but I did use 6.2K resistors to MOSFET gate and feedback to try protect the op-amp. 
It hasn't been a problem yet.  I thought that would be too much on the gate.  But even using an external square wave for switching the load looks okay.

[Yansi]

6.2k ... that's just ridiculous. Look at the time constant of that Rg and Cg.  But, as long as you are just designing load with 100Hz bandwidth to discharge batteries, then good enough.