Building a Test Load (4-30V 0-4A)
Vložit
- čas přidán 8. 06. 2023
- Let's build a DIY test load to test USB chargers and other power supplies. It can operate at 4 to 30V supply voltage and draw an adjustable current from 0 to 4A. Warning: it can get very hot, risk of fire or burns!
Please support my channel on Patreon:
/ diodegonewild
Instagram:
/ savage_danyk - Věda a technologie
Seeing you build these fantastic projects with raw PCB stock, a Dremel, the massive soldering iron, and salvaged parts always makes me so happy. Thank you!
The right to :
Repair.
Repurpose.
Recycle.
;)
PCBMyWay?
End e ket, of kórz
YEAH! ❤
-Ferric Chloride-
*DREMEL*
Your PCB fabrication style is superb, just hack it into existence. So cool!
LOVED IT! ❤
Thanks for this useful project that almost anyone could build. Your whole collection of DIY projects is amazing and I'll never understand why your channel doesn't get more views. I like some of the well-known electronics creators and have watched them for many years, but many of them never build anything or they just build quick demos to make sparks with :(.
This channel feels like that little restaurant that's open super late that all the chefs that work in fancy places go to eat after work. I've heard most my other fav electrotubers mention that they watch Diode's channel throughout the years. :) But I agree, I wish more people would watch these videos. They're really great.
5:00 - incredible pcb manufacturing ❤
I love how you up cycle electronics so unique you and the post apocalyptic inventor are some of my favorite youtubers
Add a high value resistor between the middle of the potentiometer and gnd. This way, in case of bad contact, the op amp input is not floating but goes to zero.
Awesome danyk keep doing this you are a treasure chest of knowledge!
Another improvement to this circuit would be to add a capacitor to the noninverting input of the opamp and leave one end free. You could vary the load current at a certain frequency and by a certain amount (around the point you set with the potentiometer) with a signal generator you connect to the capacitor. It could be used to test out how fast a power supply responds to changes in the load.
Thanks a lot for all of your post.
They are very helpful.
Keep up the good work.
Salute from a fellow engineer.
fantastic PCB prototyping, and the final look is stunning! great vid and idea!
I have been thinking about building a test load, this looks very promising! I will keep a look out for the "hi-power" version also!
your ideas are so aligned with mine except you can build it
Ohhh! Thank you for this video, I really need one of these for testing power supplies and some lead acid batteries :)
Very nice project 😊
Your boards are works of art. I reckon if you nailed that to a white wall in a white room you'd sell it for thousands hahaha
2 to 3 days ago I will furiously searching for this type of current test load because I make a transformer and wants to load at it max before voltage drop.thank you for making it now I just find some components.
awesome, all the PE videos are giving me some great ideas. Thanks man
Just ad two mounting posts or screws at the back of the heat sink, so you can place the load with the heat sink fins upwards to get a better dissipation, and without accidentally shorting something on the board, the support for the pot already makes the front side "legs". Its a pretty simple and fast improvement, but obviously a fan would be much more effective.
Dany, you the man. Good video.
The stability capacitor is very important. Almost all the e-load shown on internet does not talk about the stability, because no one measures the opamp output.
If we need to increase the rating, we would like to parallel multiple MOSFETs and we would also want the dissipation to be in MOSFET and less in sense resistor. Under such a case, the sense resistor should be small (for small dissipation in resistor) and increase the gain of opamp. We have to use differential opamp. And then, the stability becomes even more critical. I tried it earlier and could not arrest the oscillation.
I was thinking the same thing when I first saw the schematic: "Hey, won't that oscillate without external compensation?" Doubly-so since the gate of the mosfet is pretty much a purely capacitive load for the op-amp, and that's usually a recipe for oscillations. Edit: Although he went to a Darlington in the end, which is probably better for the op-amp as well so it at least sees _something_ on its output.
Never imagines a pcb could be make with such ease. And here i am trying to attach a marker/spindle to a 3d printer 😂
Very nice project
Thanks for sharing
1:47 lol, sooo cooooooolllll! Nice video.
I am working on a 0-60V@ 20A test load I have it basically working but fight stability problems at certain currents and voltages. I am using 8x600w FETs and a 2x2 feet heat sink with 3 inch fins out of a 2kw amplifier.Heat management is not a problem although the room gets toasty hot after I run a 50V power supply at 20A for an hour lol.
Niiiiiice! (this word should always be pronounced with the typical up and down tonal swing, exactly the way you do it)
Of course! Niiiiyce
Awesome video
This was a pretty neat project. A while back I pulled the transistors (and a bunch of other components) out of an old surround sound amplifier that stopped working. They're pretty beefy Darlingtons, so this would be a great project for them. Might be tricky to find a heatsink huge enough for all of them, I've looked at heatsinks for other projects and they can be quite expensive. The one in the amplifier didn't have a lot of fins, so I didn't bother keeping it.
My best source of "beefy" heatsinks is the abandoned skeletons of PCs I find in other people's rubbish.
@@edgeeffect that's a good point, each one can easily dissipate 60W with the fan on (most mid-level CPUs are rated at 65W), so just drill & tap a hole to mount the transistor. If I want to do >60W I can just get a second heatsink for another transistor.
Great video. Thanks
Great video.
Weeeeeee need More videos like this 😮😮😮😮😮😮
Diodegonewild how long have you been doing electronics? I have been doing electronics for 1 and a half years. You are one of the people who inspire me to make my CZcams channel keep up the great work:)
For 30 years. He started at the age of 6.
@@vaclavtrpisovsky really wow.
@@adamselectrobench
Can be true
Because he has another channel very old
70W with no fan is too much. Nice and simple, I like it. Add maybe fuse?
Great video. Thanks 👍🇮🇪🙏🏻
I'll try this it seems interesting.
V poslední době se mně začíná líbit to, jak na pána bastlí ty pošňáky. ;)
So nice 👍👍👍👍👍
Good ! you make hot heater for winter 😌
Nice video, keep it up, thank you for sharing it :)
Awesome !...cheers.
Great job! I think i would spend a fan for the heatsink, at least at power dissipations > 40W, powered by external 12V source. I guess you could use it for 80...100W with the fan!
I gave up electronic loads that are powered from the DUT. It does weird things when the DUT is overloaded (and drops its voltage to zero) and if the circuit looks like similar to yours, then you need rail2rail opamp, which is the royal pain in the ass, so i decided to build mine with the external power supply, which opened the possibility to add a huge 230V fan to it and utilize the symetric psu for the opamp (which is OP07 in my case) and solved issues with the voltage reference (REF02 in my case).
I found using TL431 as the voltage reference to be a very bad idea. They hugely drifts with the temperature, which is a huge issue in the case of electronic load, because these things have to spread a lot of heat, which makes the TL431 drifting like a hell. The similar issue happens with linear PSUs, where this becomes even more tricky, because the power loss spread out in the form of heat depends on the output voltage and current, so the reference sometimes sits where it should and sometimes drifts like a crazy and it's not easy to figure out what's going on.
Voltage reference chips and similar circuitry should, where possible, be situated away from the components that get hot. For example, if there is a fan, they should be on the air intake side.
The opamp used here mc33171 actually works fine with input voltages at (or even slightly below) the ground level. But yes, it's something to keep an eye on when selecting the component!
Excellent.
Thanks
Thank you for your support ;)
Not surprising that an open loop opamp wanted to play high frequency tunes. As this is a DC load you could limit the frequency response to a few 100 Herts.
I think i would lift the PCB off the heatsink so as not to over stress the other components. Plus I'd go the fan route as 40W's on a small heatsink is asking a lot of convection cooling, and as you pointed out the orientation of the heatsink plays a big part, you have got to give the air a chance to get sucked in at the bottom and rise out of the top.
Surprised you dumped the cooling fan!
I love all your designs and electronics stuff. My question is, have you considered that the apparatus (your electronic load) is also eating current in parallel to the DUT (Device Under Test) such as power supply?
With Drok load you can provide external sytem 5V and in this case it will not skew the measurements as it will not use the DUT to power the load's controller and fan
0:41 Jé ty už jsi tu zátěž utavil. Tak aspoň že nějakou dobu sloužila. Jsem rád že jsem ji nevyhodil a dal jsem ti jí. 🙂 Mikro USBčkem u toho malého vypínače se dala zátěž napájet jiným adaptérem a tím to neovlivňovalo měření a po selhání zdroje to zachovalo uložené naměřené hodnoty.
I guess it's useful yet annoying sometimes to have an externally powered test load. I use one of those chinese test loads which uses regular cpu coolers that goes up to 150W, 100V or 10A max and it gets the job done very nicely, although it needs a 12V for it's own operation. I remember to have blown it's transistor one time, but just replaced and get it working again easily. It's analog controlled using coarse and fine pots but have a digital display, very handy. But when you build it, you can choose the components to match exactly what you want, and better than it, you can choose it to have longevity on mind, and anything that goes wrong you can fix with ease because you know exactly what the circuit does. The problem only relies in mass production, if you need for example more than 5, 10 units, your only choice may be to buy them.
If you were going to build more than one, you could get a PCB manufactured.
Hi Daniel, how about just mounting a slow spinning fan (under voltage) that you can manually switch on if needed? I’d bet you will achieve a much wider working range of the system and the current should stay constant in the on or off scenario. Thanks for the video!
Niiiiiice!
everydoy has bags of it ❤❤❤ that is so true 😂😂😂 just like 555s
Great
You can save a lot of effort for the voltage stabilization of the potentiometer if you operate it simply with a constant current source like the LM334, which can be adjusted by a precision trimmer. 10 µA is a very useful current to avoid self heating problems. If you want to get to lower voltages you can use a so called logic level MOSFET with a low gate threshold voltage (Vgth). But also with the IRF530 you can easily go down to 2.5 V because of its typical Vgth of 2 V. You can use the MCP6001 or MCP6002 as op amp and clamp it with a Zener diode to 6 V. According to spec it is capable to work down to voltages of 1.7 V (in reality I have operated it also with just 1.4 V).
Wonderful. I bought a larger ebay load that supposed to handle 100W but constantly kills the Fet.
Such an ...interesting accent lol good stuff man love watching your projects
I built one with additional lithium Ion battery to power the opamp (741)..
And add power resistors in the input stage to dissipate moro power
You could use cpu cooler for heatsink as they are very cheap and more capable than this heatsink. Amazing project! Probably will use it
Niiiiicceeeee
Soldering the lm317 to board converts it from to220 to dpak.
Gonna need a bigger heatsink than that for constant load. That Heatsink can handle maybe 20-30W passively.
I see designers in China will be licking their lips to make more money after watching this.. 😁
❎pcb etching
✅pcb grinding
nayyyes...
How about using boiler heater elements or a washing machine water heater and a buck converter to change the voltage. It should be much nicer for the components but also far more complicated.
cnc pcb 5:23 😂
Sorry for late 🙏🏻
NOT dodgy! 😂
What thermal imaging camera do you use please Dany?
Can you make a load tester, with a voltage monitoring plus current monitor Ver.2?
👍
Veľmi pekne zhotovená umelá záťaž, páči sa mi, že už od začiatku výroby DPS myslíš na mechanické upevnenie a takéto veci. Možno jeden taký tip na zlepšenie, zapojiť medzi bežec potenciometra a zem rezistor povedzme okolo 1 MOhmu aby, keď bežec stratí kontakt s dráhou pri otáčaní nebol vstup OZ v lufte, napätie na jeho vstupe by ale v okamihu straty kontaktu kleslo na nulu a nastavenie prúdu by sa takto skokovo zmenilo, čo by mohol byť problém, preto by možno bol lepší malý kondenzátor, povedzme 100nF až 1uF (najlepšie asi fóliový, ale myslím že by mohol byť aj keramický), ktorý by po ten okamih straty kontaktu s bežcom udržal nastavené napätie na vstupe OZ.
Wouldn't it be nicer with an old chech KD503 (in Bulgaria they sell N.O.S. for CZK54.45 and used for CZK36.3 w/o delivery) fed by a TIP110?
I have built a 300 watt load tester with lm324 but it is a pain setting up 2 volt meters and it doesn't work very well under 1 amp an its max voltage is only 30 volts. Working on smps/TV LED there is allot of times that i have to load test 48, 72 and 120 dc voltage
Danke!
Thank You.
A more powerful load gives you the possiblity to show us...
... how to parallel 2 or more Transistors in linear mode...
.. in case of available lots of same low power transistors and no power transistor in stock.
XD
Thank you ;). Yes, I think about building a bigger one with more transistors in parallel ;).
Sir, Nowadays I don't see any fire extinguisher in your work bench. Are you quitting or what..!? 😁😁😁
Will this scale to 400 watts?
I want to test an aliexpress PS I just picked up.
Can i use lm471 op amp with an external power supply?
Thanks. But, not need LM317. You can use only TL431 and series 1/4W 10KOhm resistor.
If it still has to work down to 4V, the resistor would see 4v - 2.5v = 1.5V, and a 10k resistor at 1.5V only passes 1.5v / 10 000 ohm = 0.000 15A = 0.15mA, but TL431 needs at least 1mA for proper operation. The resistor would have to be reduced to 1.5k to pass 1mA at 1.5V. But at 30V input, the resistor would see 27.5V, thus pass 27.5v / 1500 ohm = 0.018333A = 18.333 mA. Its power dissipation would be 0.018333A x 27.5v = 0.504W, so a 1/4W resistor would not be enough.
wonderful!!! :))))))))))))))))))))))))))))))))))))))))))))))))))))))
How does the current flow with 50meg ohm resistor?
its 50 milliohms, not megaohms.
@@DiodeGoneWild Thanks dude!
Recently I made this circuit for myself, I used a tip142 transistor, TL081CP Op Amp and a 22k potentiometer, but there is no current regulation, the 5v circuit constantly consumes 2.6-2.7 amperes, I tried with a 12v 2.3 amp power supply, but it reset. The transistor heats up but nothing else is warm. if I pull out the Op Amp the current stops. Would you be able to help me?
update: I fixed it by replacing TL081CP with LM358
not using lm358 😢
TIP14x has a maximum power dissipation of 125W at 25°C and 60W at 90°C. For a MOSFET in linear mode, I recommend the FDL100N50F, which has a power rating of 1200W at 90°C.
Thats nice but cost 10 times more money than the used one
Bro i tried this circuit but it only draws 10ma
If MC33171 is replaced with TDA2030?
If a television is replaced with a washing machine?
@@DiodeGoneWild underrated comment :))))))))))))))
The magic smoke will escape and the Universe will collapse.
Truly "Diode Gone Wild" style. - I got a spare Scullcom PCB for czcams.com/video/FkvvmR4_D-A/video.html dummy load design if you want it.
Môžeš kámo rozprávať ako chceš aj tak je jasné že si pepik nemyslím v zlom 😂
sir, circuit diagram plzz🙏
It's shown at 10:42
duplicating what you do not easy man! obviously you have been doing this long long time. But for somebody trying to copy what you do without having enough details( that you don't provide of course) is almost impossible. For example the way you produce the pcb is amazing but very very hard to duplicate!!
Dodgy...
The BD141 is way cheaper than an AOT240L. :)
Yep, as you ended your video, fan is required! :)
Not for 18W chargers and not for a few minutes test ;)
@@DiodeGoneWild I hate fans too, but hot power devices is way worse.
Surprised AOT240L failed at 70W, it has a good DC SOA, unlike most mosfets. It should have been fine at 70W BUT with a huge or fanned heatsink, the one you have is too small and NOT fanned, so maybe that is what killed it? ...or it was a fake?
I was also surprised that it failed despite having such a large DC FBSOA. It failed in a few seconds, so the heatsink isn't a problem. Maybe the datasheet is way too overconfident. It almost looks like they drew the DC SOA just based on the maximum power dissipation at 25°C without even testing anything for a thermal breakdown :D. Alpha and Omega Semiconductor might be sort of a low cost producer...
@@DiodeGoneWild A&O tend to over spec their SOA some, but this thing it 300W over much of the curve, so maybe it was a FAKE! :(
Needs more salt.
sir, circuit diagram plzz🙏
Are you cigán?
@@srbnat does it matter?
Its shown at 10:42
@@worldcitizenoss It does, for i am anticigánist
@@srbnat Why, what have they done to you? If you would have followed the link to his channel, you would have seen that he is from India.