Scullcom Hobby Electronics #57 - Electronic DC Load Part 9

[@juliusvalentinas]

Please continue the upgrades (quality and power handling improvement), this is the best DC load project I've seen. Documentation is also best.

[@scullcomhobbyelectronics1702]

Thanks for the comment.

[@prasoft2999]

Sir, This is one the best video series I have seen till date. Explanations are too good. In fact you have demonstrated how a product evolves from simple features to complex features and how to go about design improvements. Thank you so much.

[@scullcomhobbyelectronics1702]

Thank you for your kind comments.

[@TheFunkman]

I built mine with no problems other than one caused by me. I will add a LED that lights when the load is on. Thanks Louis!

[@theonlymudgel]

I ended up ordering all my parts from Digi-Key. Should be here this week. They supplied all but the following.. 3 resistor types from Mouser, Arduino Nano, 5x4 keypad, 2 caps and a diode from different Chinese eBay resellers. The Project case, large heatsink and fans from Australian Ebay suppliers. If it helps anyone, I can prepare a spreadsheet with all my order details. Total cost was about AUD$250. I’ve got a couple of projects already started which I’ll have to finish first.. thanks again Louis. This is going to be one awesome build.

[@scullcomhobbyelectronics1702]

Hi Michael, Looks like you are progressing well. I am sure others will take note of your offer of a spreadsheet. Regards, Louis

[@tonybell1597]

Excellent, thanks Louis..... Learning so much , all these little "gotcha's". The new PCB is superb.....

[@scullcomhobbyelectronics1702]

Thanks Tony.

[@chuckwhite2285]

I've been building a parallel project based on your initial design, but it diverged several revisions back. Nevertheless, I've been following your improvements with interest and the fix for the variations caused by the fan is very timely. I'm still testing, but I may have a cleaner solution than the heavy gauge wire spider used to beef up the ground path. Instead, I used a single heavy gauge wire to supply the ground from the power supply input directly to the emitter of the fan switching transistor. This eliminated the high current passing through the ground plane and resolved the current and voltage measurement variations. Testing will hopefully confirm it also avoids any load variations as well. For testing purposes, I rewrote the fan control routine to pulse the fans on & off at 5 second intervals regardless of temperature. Thanks for your excellent work! I've tested my version at 450 Watts for up to 1 hour while still maintaining a case temperature well below 70 degrees C and maximum current capacity is 50 Amps. Thanks again! Chuck White

[@scullcomhobbyelectronics1702]

Hi Chuck, Thanks for your comments. The additional ground wiring also improved the overall operation and calibration as well as solving the voltage fluctuations when the fan came on. I have since designed a new PCB with additional grounding. Nice to hear your version was working well. Regards, Louis

[@jprsky]

brilliant explanation, Thank You!. Indeed the best DC load project out there with such a great documentation.

[@ВасилийЧапаев-ь3ж]

Thank you! You are the GREAT engineer!

[@stompreaper]

I really enjoy this project, I'm planning on building one in the near future. Thanks for all the time you spend on this and for sharing!

[@scullcomhobbyelectronics1702]

Thanks Stephen. It is always nice to share with others. Regards, Louis

[@DavidLeeMenefee]

OK I couldn't help myself I just purchased one of your boards on Ebay(American)... PEACE as always

[@flemmingchristiansen2462]

What a pleasure to see you Back, great work as usual.

[@scullcomhobbyelectronics1702]

Thanks Flemming. Regards, Louis

[@eboy536]

I'm happy to c updates video Thank you very much dear sir

[@jameshauser1507]

Good to see you my friend. Thank you for the updates.

[@scullcomhobbyelectronics1702]

Thanks James. I have been busy of late. Regards, Louis

[@FilipBeunens]

I was looking for a while for a DIY li-on battery testing system on the internet without success, but this video series had it all... THANKS FOR THAT! The detailed explanations at each step were very instructive and fascinating, as I occasionally repair / refurbish li-on bicycle batteries I would like to use this DC load module to test the capacity of the fixed battery. These are usually 10, 13, 14s sets so to 58volt fully charged that would work I think, but could this setup discharge a pack of 16s or 20s @ 3A these are 67 & 84v when 100% charged?

[@atec9228]

Hello. First of all, thank you for such an incredible project. I have a question. The modification that prevails in the electronic load is with: R43 in 1M, R22,R23,R24,R25 1ohm 5W and R10,R12,R26,R32 in 270K??. When contrasting the scheme with the list of components I saw that although they are the same version 9.3 of 04-26-2018 there were differences. Thank you so much.

[@Moonbase59]

Great as the other parts of this series. Some points of improvement I found, nevertheless: 1. You use the Keithley to measure exact 4.096V for the reference, but trust a power supply to display correct voltage/amperage? We HAVE already seen that the power supply’s voltage differs from the Fluke readings in previous videos in this series. I’d NOT trust the power supply’s readouts for doing the calibration but use a known calibrated multimeter for that. 2. I’d strongly object to EVER having connectors on the SIDE of a unit. As can be seen in your lab, benches get stuffed with equipment and you simply won’t have the space to connect something on the sides. 3. I’d probably even put the power supply INSIDE the project case (simply HATE to have a zillion of these wall warts) and have a regular mains connector on the back. Of course it would interfere with the current heatsink + open fans on the back.I wonder how that might be improved? Maybe use a larger case with slots on top, move the heatsink inside and thus allow for fan guards on the back? 4. For the USB, I’m undecided. For control and upgrade purposes, I’d typically install a USB-B socket at the BACK of a unit, for data logging use a USB-A socket on the FRONT where one could plug in a laptop using a standard cable. (Again, mainly to enhance usability in a stuffed lab-one simply wouldn’t want to move around equipment once it has found the right place on the bench …) 5. Are you planning for an updated PCB or leave that as an exercise for the reader? ;-) Having tin leads I find okay, but the extra grounding wires somehow I don’t like. As always: Thumbs up on a great procject, and many many thanks for showing it thus clearly and understandable!

[@abeleski]

always looking forward to your videos. Thank you

[@scullcomhobbyelectronics1702]

Nice to know. Regards, Louis

[@正直誠信]

Great series of electronic load design. I have watched from part 1 to part 9. I have not watched the part 10 yet. But I have one question about the constant voltage mode. Can your design achieve constant voltage mode ?

[@scullcomhobbyelectronics1702]

Hi, I am sure the feature of Constant voltage (CV) mode could be added to this project. I think it would simply be achieved by making the electronic load sink the necessary current to keep the voltage at a preset level. This could be done in software as a new option as we already have voltage sense inputs available - if I have time will have a look at it. Feature may be useful in some battery testing operations. Best regards, Louis

[@isprithul]

1) How much power were you able to dissipate through that heatsink? 2) What are the dimensions of that heatsink? 3) Can I use IRFP250n MOSFETs? Please answer. Thanks.

[@DavidLeeMenefee]

Really nice to see you back. Thumbs up...PEACE

[@akhurash]

Great video! This is a really great design and video series.

[@scullcomhobbyelectronics1702]

Thanks.

[@freddiemortos8519]

I liked(thumbs up) it before watching your video :). As always it is superb,full of information. Unfortunately I can't build one for myself because of the lack of fund :(.

[@Dubbie99]

PM me.

[@GapRecordingsNamibia]

Hi, Thank you for this in depth video series. even years on and there is still none to surpass it. I wanted to ask for the schematics as the links to this do not work, but I navigated to your website and found even more projects. Regards.

[@randycarter2001]

I've been looking at your design. There is a problem users need to be aware of. The ground of the USB is connected to the negative input. When you are testing a grounded/earth-referenced power source there may be an undesirable current flowing through the USB cable to earth via your PC. A USB isolator can be added either internally or externally with a device like Adafruit's item 2107 USB Isolator. FYI. Don't ask me how I know.

[@scullcomhobbyelectronics1702]

Thanks for the info. If you wish you can add a USB isolator. However, so far I have not had any issues with this project. Regards, Louis

[@randycarter2001]

I've got 1 laptop with a completely fried motherboard. Had 7oVDC at over 20amp applied to USB cable. Thought it was isolated since it was running on battery, but it was hooked to a monitor which grounded it. The USB cable has insulation melted. Another PC with completely dead motherboard USB port, no power, OS doesn't see anything plugged in. Luckily the rest of the motherboard survived.

[@ciano5475]

Next time use a Bluetooth serial adapter. :)

[@mancelprince7426]

Please, define 7oVDC. I assume that it is 70 vdc or 7 vdc. Whatever it is it's out of spec for USB applications. If it is capable of delivering 20 amps it was a bomb waiting to explode. If it was 70 vdc, you're fortunate that the over voltage had a path to ground other than through you. It probably won't help but I feel your pain. I've fried a few things myself. Everyone has.

[@steenandreassen4591]

Could it be that the battery you were measuring on had the positive terminal connected to ground like on some Solar chargers, because then it makes a lot of sense since you're then shorting the battery via the USB conenction if the PC or the monitor is grounded

[@PL-VA]

Can you elaborate on the voltage sense PCB plug (4:57)? If you are to just connect the voltage sense lead to the rocker switch, what's the PCB connection for?

[@scullcomhobbyelectronics1702]

Hi Peter, You can download a wiring diagram from the link below: www.scullcom.com/DC_Load_Voltage_Sense_wiring.pdf Regards, Louis

[@stephanc7192]

An amazing project!!!

[@scullcomhobbyelectronics1702]

Thanks.

[@Jackdantte]

Passion . good quality ... thanks for your excellent videos

[@scullcomhobbyelectronics1702]

Thanks.

[@duncanjamesbarker5220]

I dont understand why the BUK956 mosfet datasheet says the absolute max power dissipation is 349W but the safe operating area in Fig.4 at DC shows limits of 100W.????

[@fellpower]

I hope you are okay. would be nice to have the schematic as a file for kicad or easyEDA. If someone did it, plz link it.

[@Arek_R.]

My recommendation of mosfet is SUP57N20 from Vishay. Could reach 200W at 15V but had soldered package directly to piece of copper(IHS from old CPU) and then this module on old AMP CPU heatsink using good MX-2 thermal paste, and still about 100*C on package(not dye) so quite on the edge. What I would do next is put 4 of them on one piece of copper, and then this module on modern CPU heatsink that uses heatpipes(around 10-20 punds with fans). Also I expected the fan to cause some issues, so I would use tiny transformer to power analog circuits with separate analog ground, then use guality meanwell 2A/12V SMPS to power the fan and digital circuits/display. Fan would be 4-pin type with PWM input, starts with 40*C at about 15% and then 100% at 100*C, also RPM monitoring for fan fault protection/cutoff in case it would get blocked by dust etc. To be honest I'm not happy with 1R resistor for balancing, could interfere with SOA of mosfets when running at low voltages@high currents. In the past I was running 10 mosfets in parallel at crazy powers and had no issues when using 0R14 resistor. But still great design, well done, wouldn't be able to design all of this myself :P

[@scullcomhobbyelectronics1702]

thanks for your comments. The SUP57N20 is likely to need high gate/source drive voltages so may not be suitable for this project. You could make changes to the gate drive voltages by using a different OP Amp operating at a higher voltage (say 12V). Regards, Louis

[@Arek_R.]

Scullcom Hobby Electronics Oh right, I had just used ordinary LM358

[@Pops180]

Any interest in switching to a normal level mosfet? Also, Littlefuse makes linear FETs in a to-247 package. Im using 12 of them in a 2kW dc load

[@prabirdebnath5197]

Awesome project - someone has shared your link recently. Wondering how i missed to watch - though not watched all part. QUICKY would like to know the meter limitation in terms of its range. What i mean what is the maximum current and voltage we can set ?

[@paultakoy7091]

You are doing an amazing job with your videos. I am looking to see if you still have PCBs available for this project as well as your other projects. If you could please reach out to me. Any information would be greatly appreciated. Thank you!

[@hansschroder4643]

Thanks, very much for your excellent video series about the DC Load! I would like to buy one or two PCB's if you still have some available.

[@eboy536]

Excuse me please sir When temperature getting high load resistance value increases, it's better results at wirewond resistance

[@tinygriffy]

as always, a pleasure to watch !

[@scullcomhobbyelectronics1702]

Thanks Nico.

[@duncanjamesbarker5220]

Fantastic videos. Best on youtube. I have some questions about the circuit related to accuracy and calibration. Firstly, concerning the 1M and 1M8 resistors to -5v used to reduce voltage to zero at the output of the buffers; is this non zero output voltage of the buffers due to to fact they are run off the ground rail, and therefor can never reach zero? and so can be fixed by running them from the -5v rail? You rightly added a differential opamp as a voltage sense but did not do the same for the current sense. The large current coming from the mosfet source going to the -ve banana jack would cause voltage drops between it and the circuit ground which is where the current sense opamp is referenced. Is this correct? Another point would be resistance temperature drift in the 0.1ohm power resistor. If its resistance changes with temperature, it would cause a error in the current measurement. In fact on video 9 you show that at 5V, 3A in, low temp, you have a error of 7mA but at 30V, 3A in you have a error of 17mA. Something i will add to my build is a input filter, including a diode from -ve straight to +ve jacks, to protect the circuitry from high voltage spikes created when testing Dc-DC converters or sparks due to connect/disconnect.

[@kalj7]

That thing about using dual rails for the op amp is a really good point. I really hope that question gets an answer!

[@FotatoPotato]

Hey guys, so I just finished building this DC Load and I re-designed the PCB to be able to handle 10A+. When I went into the 'User Set-up' menu to set the max current and power I noticed that no matter what, if I set the max current over 5A the settings would not save and it would restart the process. Also, the max power is limited to 244w and I would like to be able to go up to 300W. How can I change the code to allow me to do this? Thanks! p.s. This is a fantastic project, by far the best I have ever seen on this topic. I'm very impressed by the quality and time you put into your projects and I am very grateful for that. Thank you so much!

[@e42_channel]

You are right. The 5 amp limit is hardcoded. In version v.35 from Louis it is Line 228 in the main loop: "if(CurrentCutOff > 5){ userSetUp(); }else{..."

[@Arek_R.]

Hi there have been some changes done to Arduino IDE and now it doesn't compile, multiple errors, I tried to sort it myself but once I fix one error another shows up, even some errors with some of the libraries related to boolean, and I'm too much of a noob to figure it out myself. Could you please take a look at it? Thanks!

[@e42_channel]

Louis latest version for the firmware is v.35 / v.35b as far as i know. I rewrite the code because I modified the electronic/schematic, but the original version can be compiled with the actual arduino 1.8.10 IDE. Be sure you have all needed libraries installed; also Louis own MCP79410 timer lib.

[@mancelprince7426]

Louis, thanks for all the useful projects and the detailed explanations of their operation. It seems that you have several followers that are building your designs and I see that some of them have questions and ideas about improvements or modifications. I've been wondering if these people have a way of communicating with each other to exchange ideas about your projects other than here in the comments? If there is a blog or forum where thoughts and ideas are exchanged, I would appreciate knowing about it.

[@scullcomhobbyelectronics1702]

Hi Mancel, You can post messages on my website for each project at www.scullcom.uk

[@randycarter2001]

I just ran across something you might want to be aware of. I just bought a 20x4 LCD display with a PCF8574 interface board on the back of it from Amazon. I could not get anything to show on it. It turns out there are 3 variants for the PCF8574. The first is made by TI and has a default address of 27h. The second is made by NXP (Philips) it too has a default of 27h. The third, also made by NXP, has a part number of PCF8574A which has a default address of 3Fh. Took me 3 days of hair pulling and rewriting code just to change one number in the original program. aarrgghh!!!

[@scullcomhobbyelectronics1702]

Hi Randy, In Part 9 of my DC Load project I did give a download link in which I covered and explained this issue. It was in a zip file which included 4 pdf files, one of which covered the issue about the PCF8574 interface module. Additional files can be downloaded from the link below: www.scullcom.com/DC_Load_additional_files.zip Regards, Louis

[@brucewoods9377]

Well done. This is excellent

[@scullcomhobbyelectronics1702]

Thanks Bruce

[@bwack]

Hi Louis ! A year ago or so I said I was going to build this. Now I'm doing it :) Can I ask you about the 5W 1Ohm resistors and pn2222. Are they currentlimiting resistors in case there is a "shock" (as in hotplugging etc) to the input terminals ? Also the PN2222 that are in parallell with them, do they help load balancing the four load paths ?

[@scullcomhobbyelectronics1702]

Hi, The four 1 ohm 5 watt resistors help with the balancing and power sharing of the 4 Power Mosfets. If any one of the Power Mosfets draws more current than the others then you get a larger voltage drop across the 1 ohm in that particular Power Mosfet source return path which will then cause that mosfet to have a lower gate/source drive voltage and so reduces its drain/source current. The four PN2222 transistors are there to provide some protection for the Power Mosfets - if any one mosfet draws excessive current then the associated PN2222 transistor conducts and reduces the drive voltage for that Mosfet and so stops it going in to thermal runaway. I have found in practice since I increased the four source return resistors from 0.1 ohm (original design) to a 1 ohm we have now, you could now leave the four PN2222 transistors out of the PCB if you wish. Regards, Louis

[@abeleski]

So is there a spec sheet for the current part 9 DC load? Describing accuracy, resolution , power rating, features etc? I think it would be a nice summary to add. I do not see that anywhere in any videos or the scullcom website.

[@scullcomhobbyelectronics1702]

If I have time I will look at this later. Regards, Louis

[@Ken-wu2bf]

Great videos! One question / suggestion. Instead of trim pots can the current and voltage readings be cal'd digitally? I've used that technique with success when I worked for a sensor manufacturer. I know you packed a ton of functionality in this Arduino, I'm not sure if you have the room?

[@scullcomhobbyelectronics1702]

Hi Ken, Yes you could do it electronically but we have very little memory left and you would also need additional circuitry as well. Regards, Louis

[@theonlymudgel]

How timely were these last 2 videos. I had just completed locating all the parts (I’m in Australia) when you release video 9 and 10 with some changes just in time for me to include them in my build the first time round. When I finish I’ll send you my list of part numbers, as I got the majority of them from Digi-Key. Farnells who are Element 14 in Australia are just far too expensive. I’ll be about a month before I can start as a few Ebay parts will take that long coming from China. Again Louis, thanks so much for your great educational videos. I’m sure I’ll enjoy the build even more than I have the many hours of video that you’ve provided.

[@scullcomhobbyelectronics1702]

Hi Michael, Nice to hear your progress with the project. Delivery of parts from China can take more time but they are much cheaper. Regards, Louis

[@steveh4044]

Hi Louis, thank you for your great videos and clear explanations. Even though this project is a bit beyond my skill level at this stage, I still enjoy watching and learn a lot. Would you mind giving a quick explanation of how you chose the value of R10? I understand the concept of Q6 and the need for the increase in R10 as a result of the change to R22, but how do you determine the value to get just the right amount of adjustment to the gate voltage of the mosfet? Thanks again.

[@scullcomhobbyelectronics1702]

Thanks Steve for your comment/question. Calculating the correct resistor value for R10 (and the other emitter resistors for Q6, Q7, Q8 and Q9) was a bit tricky. I found in practice each power mosfet gate/source threshold voltage was slightly different even though they are the same type. Therefore the mosfet with the lowest gate/source threshold voltage would start conducting first and get hotter sooner than the other three. I therefore did some initially testing - I ran the DC Load set at 3 amp with a 30V supply without any of the transistors Q6, Q7, Q8 and Q9 installed. I then check for the Mosfet which was getting the hottest and drawing more current than the other three. I then fitted the PN2222A transistor for that mosfet which was drawing the most current and for the emitter resistor (for example R10) I used a 100K resistor in series with a 500K preset. I then adjusted the preset until that mosfet was supplying the same current as the other three - so giving some active feedback to reduce the gate/source voltage on that mosfet (you can check for similar voltages across the 1 ohm Mosfet source resistors when comparing the currents through each mosfet) and found that the effective emitter resistance worked out to be around 270K for best results. I tested this method several times using different sets of 4 mosfets and got a similar results. I therefore selected a standard value resistor of 270K and used that in all the emitter return resistors and and fitted all the transistors Q6, Q7, Q8 and Q9, I found it worked fairly well, so I stuck with that. Sorry, nothing scientific here, just a case of some tedious practical testing for best results. Regards, Louis

[@steveh4044]

Scullcom Hobby Electronics Thank you for your detailed response. That makes sense and further demonstrates how much time and effort you have put into this project. Thanks again.

[@no5x937]

Most of your problems would be solved by making a 4 layer board with 2oz copper clad ground plane on inner layer 2 and 2oz Cu clad power plane on layer 3. Layer 1 component layer for THT components and Layer 4 solder layer for SMT components.

[@juliusvalentinas]

The LOAD and SENSE wires are rather thin, and come as 2 wires to connect to PCB. That is rather strangle. You should have 4 thick wires going into PCB. And then have a relay on it for switching within close proximity. Then have switch on front panel control relay.

[@scullcomhobbyelectronics1702]

Hi Julius, The Load wires are heavy gauge wire going to the PCB. The voltage sense wires carry no current so can be low gauge wire. Regards, Louis

[@juliusvalentinas]

I thought the arduino code would run on better quality then the rest robotdyn WIFI NodeM ESP8266, 32M flash, CH340G and that even arduino IDE can be used ? With 32M flash these is no shortage of memory.

[@juliusvalentinas]

kzbin.info/www/bejne/fJzZp6CIacR3Y6M I see same gage wire going to 4 posts on front panel and the switch. Two wires then go to PCB. Are these sense wires? Then where are the LOAD wires in this shot?

[@scullcomhobbyelectronics1702]

Hi Julius, What you are seeing in that part of the video is the sense wires going to the sense switch. If you do not use the sense input leads you need to switch the sense input to connect to the load input terminals instead. There is no current effectively in any of these wires, as no load current is flowing through these wires, so you can use small gauge wires. That screen shot does not show the Load wires which connect from the Load Input terminals to the Power Mosfets - these wires are heavy gauge to handle the load current. The Load wires from the PCB to the Load input terminals were soldered on later when the PCB was installed in the case. Regards, Louis

[@scullcomhobbyelectronics1702]

Hi Julius, I have not tried a robotdyn WIFI NodeM ESP8266 so far. Do you have knowledge of using this module, if so what is your experience? Have you used it together with an Arduino Nano? Regards, Louis

[@abeleski]

Is it possible in the future to update the PCB to incorporate some of the changes re grounding etc that you did in this video? I wouldn't mind buying another one. Some may disagree about changing components etc as bugs and new features are added but I think its a price I personally am prepared to pay as I am learning so much following this project.

[@scullcomhobbyelectronics1702]

I did have a look at re doing the PCB with a heavier gauge of copper tracks (say 3 oz) but the cost of the PCB would be nearly 5 times more, so I think it is more cost effective to add some copper wire in some areas. I may do an update of the PCB files in the future to add a few minor updates and if I do I will make the PCB gerber files available. Regards, Louis

[@abeleski]

That would be great and much appreciated. I have a bit of OCD and the copper wires although do exactly what they need to do just "bothers" me :) Thank you once again. Love your projects and have learned a lot from them.

[@scullcomhobbyelectronics1702]

I have just had a few new PCB's made where I have made some improvements and added additional grounding. I have made the new PCB (version 9.3) gerber files available, download link below: www.scullcom.com/DC_Load_ver9_3_gerber_files.zip Regards, Louis

[@abeleski]

Scullcom Hobby Electronics thats so kind of you. Are you selling the new pcbs? I will check your page and see if you are. Thank you sooo much. Love your projects. Alex

[@scullcomhobbyelectronics1702]

I have had a few of the new PCB's made. If you post a comment/request on my website www.scullcom.uk I will then be able to email you with details. Regards, Louis

[@deathblowhere]

Hi, Louis! Could to tell please power consumption of the device? I see it's 15-18VDC, but what amperage? I've got a transformer with two 18VAC @ 0.2A winding and wonder if it'll be capable to supply enough juice. Thank you for your time and effort! This load is a true diy BK Precision 8500 killer xD Regards, Phil.

[@fillipedelgado]

Amazing! Congrats!

[@TCBEperformance]

Hi Louis, do you think if a 3oz copper and a full ground plain at least on one side on the PCB will help the problem of the grounding issues? I know it’s a challenge some times to do when routing tracks with a full ground plane.

[@scullcomhobbyelectronics1702]

Hi, I did look in to 3oz copper but I found the cost increased 5 times. I have done an update PCB design with additional grounding but not fully tested it yet. This will be version 9.3. I have had a few PCB's made if anyone is interested. I have also made the PCB (version 9.3) gerber files available download on my website at www.scullcom.uk so you can have PCB made youself - the download link is also below: www.scullcom.com/DC_Load_ver9_3_gerber_files.zip The version 9.3 schematic, component layout and updated parts list can also be downloaded from the link below: www.scullcom.com/DC_Load_ver9_3_files.zip Regards, Louis

[@TCBEperformance]

Yes, thanks for your reply, I think the only way to get them made cheaper is from china I had some made years ago with great prices, much cheaper than in Australia around 4 times less. After that I constantly get emails from various china manufactures asking me if I need boards made..haha. Kind Regards. joe

[@nitinkshirsagar3729]

Sir , 30V 30Amo Design available?

[@jonny393]

Just ordered all the parts...wish me luck :D

[@scullcomhobbyelectronics1702]

Hope you enjoy the project. Regards, Louis

[@jonny393]

Right after the joy of soldering everything on the pcb (9.2) I ran into an issue. The -5 and -2.5V ref voltage was not present at the probe points. I measured +0.55V on the -5V point. Im pretty sure all the parts are in the correct place, do You have any idea what could cause this?

[@jonny393]

the Lm431 is working outside the circuit, i tested that

[@tkarlmann]

What did the parts cost you? How did your project work out?

[@Chen-zd6rx]

I am very like your all video

[@jeffersonpimentamelo]

Iam Very interested in see the control circuit analysis by a control theory way, stability and design a proper compensator, is that possible? Thanks!

[@scullcomhobbyelectronics1702]

I suppose you could use a Spice Model to analyse the design if you wish.

[@jeffersonpimentamelo]

How about matlab and spice? :D

[@pippopappo3526]

Excuse me and can I buy the printed circuit board? where is it? thank you

[@electronic7979]

Excellent video

[@scullcomhobbyelectronics1702]

Thanks.

[@jameshauser1507]

Hello Louis. I was wondering is the current leakage due to the extra load of the fans?

[@scullcomhobbyelectronics1702]

Hi James, No it was not caused by the fans. Just needed improved grounding. Regards, Louis

[@jameshauser1507]

Oh yes got it. I watched the video again afterwards.

[@FlyingShotsman]

I'm curious... With three 6-1/2 digit Keithley meters on your bench, why did you not use them to measure the output voltage and current from your Tenma power supply during calibration? They'd have been much more reliably accurate than the meters built into the power supply, and would have had higher resolution than the device being calibrated, instead of less. Very enjoyable series, Louis. I always look forward to your next video. Thanks!

[@recolaq]

This is hobbyist level. No need to show of expensive equipment since most people don't have them available.

[@scullcomhobbyelectronics1702]

Yes you could use more accurate test meters if you wish and that is always there as an option. But as we are only looking at accuracy to with in a few milliamps or millivolts the power supply readings should be OK for the hobbyist and I decided to keep it as simple as possible on this occasion. I am always conscious of the fact that most hobbyist do not have access to expensive meters. Nice to know you enjoy this project. Regards, Louis

[@schluderjupp]

Scullcom Hobby Electronics thank you so much for doing just that. 😀 If we weren't Hobbyists, we wouldn't build this and just buy a Electronic Load for many hundreds of $/€/£. Besides building things is so much more enjoyable than buying a finished product. This reminds me of when my father built a Heathkit multimeter/oscilloscope/television as part of a electrics course in the military.

[@romandavydov8684]

Guys I am trying to download the archived files for the project. However links under description are not working Could somebody please share the archives Thank you!

[@FilipBeunens]

Copy the link and open it in another window, that worked for me

[@henrikworg350]

Is there any way to calculate the needed size of the heatsink? ... especially for noningeniers?🙈

[@scullcomhobbyelectronics1702]

Hi Henrik, I did show some brief details of thermal calculations in Part 2 of this project: kzbin.info/www/bejne/jGe1gpp6YtKigKM The info starts 13.10 minutes in on that video. A good explanation is also to be found on a Hackaday You Tube video at: kzbin.info/www/bejne/fmfPnqdtdtyAnJI Regards, Louis

[@henrikworg350]

Hi Louis, thank you and sorry for asking when you did show it. I was looking all videos yesterday in preparation of building this nice project. After that i was looking for a different cooling and found this: www.fischerelektronik.de/web_fischer/en_GB/heatsinks/D02/Miniature%20cooling%20aggregates/PR/LAM4K_/index.xhtml So i will try to calculate. Thanks again for your answer and help. Best Regards Henrik

[@scullcomhobbyelectronics1702]

Hi Henrik, No problem. Glad to hear found the info. Regards, Louis

[@henrikworg350]

Do you think this LAM4K is enough?

[@scullcomhobbyelectronics1702]

Hi Henrik, Sorry for delay, just seen your last message. The LAM4K looking like a nice heat sink with integral fan. It would appear it comes in different lengths but not clear on the datasheet - the longest is only 150mm, would be better if it was longer/larger. It should be OK for up to 90watt (30V at 3 amp) but it may be a good idea to have the fan running all the time. Anyway it is worth a try. I could not see a price on the link you gave. Regards, Louis

[@CliveChamberlain946]

Doesn't it sound a bit desperate to change ballasts a whole order of magnitude? Testing lithium cells down to a 2v cutoff can never exceed ~5a due to RLmin of >0.35* ohms. At 2v using 0.33 ohms would allow >10a and 0.25 ohms >12amps (* provided another binding post is mounted to bypass even more resistance offered by D5 and F1). I would never go by the name Alice, but now I guess I have to see Part 10... ;-)

[@scullcomhobbyelectronics1702]

Hi Cliff, I see where you are coming from but so far I have had no issues testing LiPo batteries. I do set the current to 1 amp or lower so any voltage drop is small. If you wanted to replace D5 with a shorting link that is always possible if you wish. I tested the DC Load with various values of Mosfet source ballasting resistors and found that a 1 ohm was a nice compromise to give good power sharing and operation of the DC Load overall. The DC Load seems to work just fine. Regards, Louis

[@RodrigoSilva-ic5vw]

great project

[@dapreygerard8981]

Hello I want to use an oled screen but as the memory of the program is limited, I would like to replace the Arduino Nano by a RobotDyn Mega Pro. the advantage is that it is compact (38x54 mm) and has a memory of 256 KB, an SRAM of 8 KB and a EEPROM of 4 KB. I'm going to make an adaptation PCB to plug in instead of the Nano. I think there is no particular trap, just pin connection for pin. Thank you for your opinion and thank you again for this great achievement

[@scullcomhobbyelectronics1702]

Hi Daprey, An adapter PCB to plug in is a good idea. I have not used the RobotDyn Mega Pro yet but I can see no reason why it should not work. You should be able to port the current code to the RobotDyn Mega Pro. Not only would it give you more memory but you could also add extra functions. Regards, Louis

[@sandipucelj9194]

Hello, is there any BOM for your project? With it will be easy complete all materials :) I look at all you files but not find that :\. BR Sandi

[@scullcomhobbyelectronics1702]

Hi Sandi, It was in one of the zip files but below is also a direct link for just the latest Parts List: www.scullcom.com/DC_Load_Parts_List_Part9_2updated4.pdf It was originally included the following zip file with other documents: www.scullcom.com/DC_Load_update_files.zip Which was provided as a download link in Part 9 of this project. Regards, Louis

[@sandipucelj9194]

Thank you for helping me out :).

[@rcfunshop1]

Hi, I have made a new PCB with recent updates version 9.2 with 2oz copper and 5x4 keypad and partly with SMD resistors and capacitors and full grounding . If there is interests, I can give some. I can send pictures and Gerber Files. At this point I would like to thank Luis for his great project.

[@kalj7]

That would be very cool! I'm definitely interested!

[@georgelivanes2832]

Hi Louis, Excellent project Asked you before about PCB availability, but this time having trouble, trying to contact you using your websites contact info. Missed out on your original PCB offering. Am I too late late to buy a PCB in your latest offering (Can’t find anything on ebay this time) Cheers George

[@scullcomhobbyelectronics1702]

Hi George, Simply leave a comment on my website and then I will have access to your email address to contact you directly (I am the only one that can see your email address) go to: www.scullcom.uk/design-build-an-electronic-dc-load-part-1/ and scroll down to the bottom where you can leave a comment. Or send me your email address in privates messages and I will contact you. Regards, Louis

[@cverity]

Wow, eBay PCBs are already gone. Are you planning on selling more? Thanks.

[@scullcomhobbyelectronics1702]

Hi Cory, I have a few spare PCB's left. If you would like one leave a comment on my web site www.scullcom.uk I will then have your email and can contact you by email to arrange. Regards, Louis

[@arduino8506]

Anyone knows where to find a suitable heat sink please for this DC load? 250*90*30mm is no longer available.

[@e42_channel]

I build my own version of Louis DC Load and used a Fischer Heatsink (200x75x40) and mounted two fans on it. Works well for up to 100W - 120W. The type is SK 47/75 ... farnell: 4621888. There is also a 200x100x40 version.

[@arduino8506]

@@e42_channel Thank you for your suggestion. I'll look into this. I currently have 150x90x30 and its way too small to handle 120W power. I was thinking about getting a Fischer Force Cooled Heat Sink (Farnell: 1222518), but its out of stock, way too expensive and I am not sure its would be enough for high power. What's the temperature is like on yours when you go to 30V x 4A (120W) please? Is it consistent? Thank you

[@e42_channel]

@@arduino8506 About 50 to 60 °C. I would prefer and recommend your solution. I think the cooling channel is better than my build. You can find this type of cooling solution in some commercial electronic loads also. But at the time I was searching for a Heat Sink the pcb was ready and the case had arrived. BTW: I also use FreeCad to build and check if everything fits well mechanically (Pictures: c.1und1.de/@752881489212872893/0IUL5HqNStaHnXjgkHw7jg )

[@arduino8506]

everything42 Installed the Fischer cooling aggregate. What a difference! However, I did notice that at high currents shunt resistors are getting very hot (around 80-90C). Maybe I should install another fan browsing onto resistors

[@e42_channel]

@@arduino8506 Good decision to use the cooling aggregate. My shunt resistor (0.1 Ohm / 50W) gets only "warm". How much current do you draw? The Power dissipation @ 5A is only I²R = 2.5W. What do you mean with shunt resistor(S!)?

[@happyhippr]

Awesome stuff

[@scullcomhobbyelectronics1702]

Thanks.

[@TheOcela]

Hi Luis! Recently started to follow your developments. I liked the voltage calibrator and current calibrator! Now I want to repeat the electronic load. Question. Why do not you want to do a software calibration instead of a mechanical one? Just adjust the indicator's readings to the actual voltage and memorize it in memory. I also developed my electronic load for 100W with software calibration, but only the current setting was mechanical, resistors vrtp.ru/index.php?act=categories&CODE=article&article=3662

[@scullcomhobbyelectronics1702]

Hi Andrew, The calibration required may vary across the whole range so a fixed mechanical calibration would not cater for that. I tried to check your link but did not see your article. Regards, Louis

[@billigerfusel]

This is some high quality shit

[@scullcomhobbyelectronics1702]

We try -:)

[@miguelsbpgj]

Hi Scully, This is about Mosfet RFP12N10L, but first : Thanks for your detailed and brilliant videos... We are missing new content from you (although, from your other answers, we know you are busy with other things)... Even your videos are very informative, I need some help from you... The 3rd mosfet of your choice, is the Fairchild (actually ON Semiconductor) RFP12N10L : At Electronic DC Load Part 2 video, you stated 30V 3A (90W) for the project, fitting inside the BUK956R1-100E safe operating area. Now, for the RFP12N10L, even it´s maximum power dissipation is rated only as 60W, I can still fit the same 30V 3A (90W) inside its safe operating area (as you can see from the image on the link bellow). ibb.co/zmzNn8s Being RFP12N10L Pd max 60W, I sholdn´t be able to fit 90W inside its SOA graph. I´m I wrong ? Do you thing RFP12N10L safe operating area graph could be erroneous ? Really lots of thanks in advance

[@vitolaudicina92]

Ciao, Si può utilizzare un display TFT i2c, su eBay ho trovato questo 252096499341. Io vorrei costruire il tuo progetto, ma vorrei aumentare la potenza per usare MOSFET con VGS da 20v cosa devo fare? Complimenti per il progetto. Saluti Vito

[@scullcomhobbyelectronics1702]

Hi Vito, I have just translated your message to English which reads: "You can use an i2c TFT display, on eBay I found this 252096499341. I would like to build your project, but I would like to increase the power to use MOSFET with VGS 20v what should I do? Congratulations on the project." My answer is as follows: The VGS of 20V is quite high. However, you could consider using IRFP250NPBF for the 4 Power Mosets, which operate at higher VGS voltages. However, as these Power Mosfets are not Logic Level Mosfets and will require a higher gate/source drive voltage, you would need to make some circuit design changes. To achieve this you could replacing the U7 (Quad Ultra Low Voltage Offset OP Amp) AD8630 (which can only operate to a maximum of +6V Supply Voltage) with a AD8664 which can operate at a higher supply voltages, you could then feed it with the +12 supply (this would then give you higher VGS drive voltages - up to around 12V). As the pin layout are identical to the it is a drop in replacement for the AD8630. If you use my DC Load PCB then you would need to cut the track from pin 4 of U7 (+V pin) and then connect Pin 4 of U7 to the +12V supply with a wire. This should then give you the required gate drive voltage for a IRFP250NPBF. I have not tested this option yet myself but it should work. The IRFP250NPBF is a TO247 package compared to a TO-220 package of the BUK956R1-100E, which would need to be taken in to account. If you are looking to have a maximum of say about 25 Amp then I would suggest using a lower value sense resistor for R17 would be better. If you replace R17 with say a 0.02ohm 10 watt 1% resistor then at 25A the voltage across it would be 0.5 volt. The change in voltage then fed via buffer U7D to the ADC (U6) wound need to be compensated for as you are now using a lower sense resistor, this could be done by setting the gain of channel 2 of the MCP3426 in software to a gain of 8 (instead of the current setting of x4) and also making some changes to the code where the calculation is made for the current reading on the LCD. With regards the resistor divider R6, R14 and RV4. Try changing R6 to say a 27K and R14 to a 1K and RV4 to a 1K preset, to ensure you have sufficient range. Hopefully the above will give you some ideas. Regards, Louis