Proper old school engineer, The correct use of "Pre-set" get a big thumbs up from me. Love the long videos and detailed explanation. Keep up the good work....
Following this series, you can see a product evolve. Good old style engeneering indeed: We have a need. We find a (rather crude) first solution and constantly refine it. AND constantly revise bad design decisions (like the 0.2V "correction" for the DAC mid-value programming). THIS is what I love, and of course you very clearly explaining your thoughts and designs on the whiteboard. Everyone starting out in the beautiful realm of electronics should watch this series! Thanks for taking the time and sharing this with us!
One of those times when a simple thumbs up cannot express what my thoughts are regarding this project. As has been mentioned in other comments, it started as a simple idea, but the additions and changes are welcome and extremely beneficial to the basic design without over complicating the design requirements. An incredible journey so far, presented and explained at a level and pace that anyone can follow along and hopefully learn from. Incredible work, explanation where needed and presentation. The next on my to-do list... Karl
That is - hands down - one of the best projects on TH-cam, embedded in one of TH-cam's "bestest" video-series. You, Sir, have a gift of marrying your superlative know-how with the flow of ideas from the community. I cannot even emphasize any particular thing, as the whole series is just on a stellar level. Though I do admit, that I occasionally pause your videos to enjoy the schematics that you hand-draw so beautifully - but then again, I'm a loonie .. LOL Marvellous work!
After watching all of the video's I should make this project! After the improvements Louis and the new PCB, it should perform exactly as designed! I also love how you explained the improvements and or suggestions to use other parts to make it better! You are also open to inputs from your viewers, others should learn from you! Thank you again Louis!
Great improvements, all down to the spot! I can't tell you how anxious I am to see this wonderful project come to its final version and build one myself. One thought crossed my mind is to change the display with a dot-matrix one. That way, the most important info can be shown with bigger fonts, and a better utilization on the display can be achieved. 128x64 or a little bigger ones does not cost much more than 20x4 character displays and only a small change in the software is enough to display everything better. Thanks again for this project and for your kind heart for sharing everything about it.
Thanks for your comments. A different display could be an option. I am looking at one or two other options. But need to fully test the new PCB on all functions. Regards, Louis
Hey Mr J!! Where do content producers like you get the time to watch heavy stuff like this? Never expected to find you here (but then you are welcome anywhere!)
LOL That's a funny one :) I don't consider myself as a very active "content creator" as you say, but I'm on TH-cam each and every day. Mostly at night, I deprive myself of sleep, that's how I manage to watch these long videos :D Maybe you saw BigClive's "So much TH-cam. (No time for sleep.)", same thing ROFL
Hey Mr. J, congrats for passing half a million views! Keep up the natural personality (like the time that fly made you mad when working on something and you put it on the video..) ha! Move over Electroboom!
+Pomona Bill I don't think he meant something bad, it was said in a humorous fashion. But yeah, each to their own. I like Mehdi's videos myself and I'm not trying to snatch a piece of his cake. I guess Cliff just found my videos funny enough to compare to Mehdi's.
This truly is one of the best electronics diy series on the tube. Thanks for the hard work and hours you have put in to make this series! I was considering getting a commercial dc load from a brand manufacturer but this project has advanced to a stage that makes me consider this as very interesting alternative. Perhaps something to ponder about that current offset topic is going with 4 terminal sense resistor and INA on +/- supply. This would leave the D opamp free for low output impedance offset adjustment. Perhaps it will complicate the project too much but it would deliver very solid sense implementation and a capability to tackle wider current range with improved precision and lower drift as the internals get heated during operation. This would also open up the capability to do 2 automatic ranges by arduino controlled shorting of the INA gain resistor for 10x and 1x amplification (2A/1mA and 20A/10mA range), for example.
Very nice work. Thank you very much for all of your time and efforts in making this wonderful piece of tool that everyone desires for who are doing powersupply stuffs.
akhurash, I absolutely agree with you. The problem is most of the people nowadays are more interested on channels which people buy stuff from China and play around with modules, sensors, etc. or assemble some kits. Nothing wrong with that, but sometimes I feel we are missing the essence of electronics. This channel is a rarity and it deserves to be one of the most viewed channels about electronics on TH-cam. I will be here giving all support I can. Love Scullcom hard work!!!
Robledo Einstein Yup. As a hobbyist and engineer I enjoy these types of videos. Building things from basic components is great to learn and understand electronics.
Excellent progress with the project. I have also made several improvements but I never got to creating a proper schematic diagram. That is until you released this latest Kicad package and included the source files. I have never done anything in Kicad so I needed a jump start and having a design that is not very far from my own was a great jump start. I have only done the schematic. I have additional plans for enhancements I want to try before I make a PCB. I have replaced both DAC and ADC with some faster/better ones. I have added hardware debounce of the rotary. I have replaced the fan by a Sanyo-Denki 4 pin type that can be speed controlled by a 25 kHz PWM signal. It is a BEAST and the load can now run with 250 W continuously with that single MOSFET in the SOT227 housing. Amazing. But at full load the noise is massive. I have completely rewritten the software. I have changed the D key so it works as a menu key and moved battery check, transient, calibration, and user settings under this menu. No more pressing two buttons. And I now store the calibration in EEPROM. Because I replaced the ADC and DAC, my software is no longer compatible with the original design but it it not difficult to merge the functions that control ADC and DAC from the Louis software into mine. It is a real interesting project and it is fun to see it develop.
And with the danger of TH-cam holding back my post as spam. Here is the URL again to my version of the load just updated today. Incl new schematic and new code. www.lavrsen.dk/foswiki/bin/view/Kenneth/ElectronicLoad
Hi. I now have an updated version of my design that also implements constant voltage. I have not updated my website yet. I need to update the Kicad files. But the principle is simple. I have added two analog switches on the inputs of the op amp that drives the MOSFET. When I change the mode to CV the DAC changes to controlling the inverting input and the non-inverting input is connected to the output of the opamp that monitors the voltage. And then it is simple code to add the constant voltage setting. I put that feature in the D menu. It would be easy to implement same in Louis' design. Very easy. I have to work a little on the bias zero circuit which drifts 1-2 mA now. Not a big deal. It just annoys me ;-)
I have now completed the next step of adding constant voltage mode to the load. This cannot be added as software because you just cannot regulate the current fast enough in a software loop. So I have added to analog switches so that I can run either in the old current loop mode where the op-amp that drives the FET compares the shunt voltage with the DAC voltage. For constant voltage you have to move the DAC voltage to the inverting input of the op-amp and feed the load voltage to the non-inverting. And it happens to be that the voltages going into the ADC for the voltage measurement and the control voltage from the DAC are in the same range so it is actually a very simple exercise. I have added the constant voltage mode under my "D-menu" I found that you have to be very careful with cleaning flux away from the 4,7 Mohms resistor used for nulling the DC current. I was fighting drifting zero for hours until I realized it was the crappy chinese flux I had used for the rework adding the constant voltage circuit. After multiple cleanings with IPA it is stable again. I have uploaded both PDF, Kikad project, and my least code to my wiki page (linked in my own comment earlier in this thread)
@@KennethLavrsen Hi Kenneth! Thank you your commitment and share to extend and modify this project. As far as I see you are the only one who provides downloadable materials for this project up until today, 22.20.2021. Good work!
Hi, you are the best, please continue making stuff which is affordable to poor electronics hobbyists.... I have a question for you, what is your specialization in electronics, power electronics???
Hello Louis and everyone, just a comment i thought to mention, not sure if this was brought up before, but the I2C pull-ups are also implemented in the PCF8754 LCD control board. So if so these should be removed otherwise you will get 2.2k ohms on the data & clock lines. I came across it by accident when I bought some MCP4725 on a module from ebay which have pull-ups already on the board, this made me think to check the PCF8754 module.
Thanks for your comments. The correct value for a pull-up resistor is an interesting subject and may be worth discussing in more details on one of my videos. In short, the maximum pull-up resistance is limited by the I2C bus capacitance. If the pull-up resistors values are too high, the I2C line may not rise to a logical high before it is pulled low (due to the RC time constant). The speed of the I2C bus can be increased by lowering the value of the pull-up resistors, within limits which depends on the Vcc Supply to the circuit - in our case the supply voltage is +5V. At +5V you can normally reduce the pull-up resistor value to as low as about 1.8K. The trade off is that a lower value resistor means the circuit uses slightly more power but gives you a higher speed. So if you happen to have two sets of pull-up resistors then as long as the overall value does not go below 1.8K you should be OK. Pull-up resistors on a PCF8754 LCD control board are usually either 10K or 4.7K, so there should not be an issue. Hope my explanation is of help. Regards, Louis
Thank you for your kind explanation, so I suppose only if you have lots of module in parallel on the system and goes below 1.8K then its best to remove some resistors. Yes I understand your explanation the lower the resistance the faster the lines are at their logic level especially if you have the fast I2 communication switched on. Kind Regards, joe
Just discovered your fascinating channel and web site Louis, thanks for all the great videos. I have decided to tackle building the DC load but the Mosfets you chose are no longer manufactured, can you suggest an alternative to the BUK956R1-100E?
I wish I'd run across this video sooner. Your concern about the -5v power source is noted. It doesn't matter what it is, 4.1v for example, it just has to be stable. You're only using it to bias the op-amps to operate in the required range. As for I2C, I've always used 2.2K or 2.7K pull-up resistors and never had any problems. I've seen circuits with pull-ups as low as 1K. Why anyone wants to cut the current in the bus to such low levels is beyond me.
Thanks for your comments. Yes agree the main thing is that the negative voltage source is stable. With regards the pull-up resistors on the I2C bus using a 2.2K or 2.7K pull-up resistor will allow a faster speed on the bus. I think the lowest value you can go down to is 1.8K. The only drawback of lower value resistors is it draws more current but this should not really matter if your power supply can handle it. On the plus side you can run it at a higher clock speed. The normal default I2C clock speed for the Arduino is 100KHz but you can take it up to 400KHz when you use lower value pull-up resistors. As the LCD display module also has pull-up resistors then adding additional pull-up resistors as in this DC Load circuit we are reducing the overall pull-up resistor value and so allowing a higher bus speed, which is set in software. Regards, Louis
Happy New Year, Louis for part 8 (I do hope you are still advancing the project) I would like to suggest to realize a data interface to collect the measurement data, especially for the battery testing use case. The pins TX/RX (D0,D1) for the Nano in your projekt are still unused. With a USB to Serial adapter based on a FT232R USB UART IC it wouldn't be too difficult to build a USB interface in the DCLoad. There are readymade adapters available on ebay. I guess you know it... Feedbackloop 's (a youtuber) has shown in episode #10 a way to create a graph based on data collected from a dc load. The SW is available on github. Pretty cool... What do you say? ;)
Hello, I write from Brazil. I have now reached your channel on youtube and I was very happy with the quantity and quality of the content, so I come first of all to thank and congratulate you for your work! I'm going to build your project, it was great. Sorry for my question, but what is the biggest voltage and power of the load?
In solving the issue with the DAC starting at half output voltage, you can just write the EEPROM default setting to be 0. I'm not sure I understand how adding that turn-on bias circuit is the simplest solution when there's a software fix, granted it works the very first time without having to power it up once and command a default setting, that does not seem like a necessary benefit. I also wonder if, rather than trying to bias the input of the opamp on startup, it would make more sense to have a switched positive supply to that opamp. This is common for automotive motor drivers where there's a logic input to shut down the gate drive.
Why not just store 0 in the EEPROM of the DAC and do away with the the whole 2V business? 4K7 pull-up resistors on the I2C lines are pretty much standard in the hobby robotics world. They usually place it at the end of the bus. Thanks for the KiCad files and all the hard work you keep pouring into this project. Its a perfect project for me and I am about half way there. It's such a useful project, unlike blinking LEDs and fidget spinners which are all the rage.
Thanks. As I said on the video I will look in to the possibilities of reprogramming the default setting in the eeprom to see if that can be easily done. Regards, Louis
This is a most interesting project, and becoming quite sophisticated. A couple of things jumped out at me while watching though:- The protection diodes in parallel. Although they are on the same die, there may still be enough difference in forward voltage drop to cause one diode to hog all the current. Maybe a very low value ballast resistor in each anode leg could be added. Though I suppose even just one diode at 40A is sufficient... I'm not sure if the drop across the diode(s) (and any added resistor(s)) is of any significance in this application, but if so, or in any case, maybe have a reversed diode across the load terminals (via the fuse) as further protection against reverse polarity. The need for null current trim. With op-amps that have 1uV offset voltage, I was surprised to see this was needed. Could this trouble be caused by a single op-amp trying to balance the behavior of the 4 mosfet circuits? And might separate op-amps per mosfet (as someone else suggested) help in this regard? (When I finally do get around to building something along these lines, I will certainly be looking into that...) Thanks for all your work on this!
Thanks for your comments. With regards the dual power schottky diode. As the two diodes are on the same silicon die they are thermally coupled so it is unlikely to be an issue. Also both diodes are rated at 40 Amp so I do not think it will be a problem. With regards the offset null adjustment. This requirement was found to be needed for either a single Mosfet or 4 Mosfets output stage. Although we are using an OP Amp with a 1uV offset, other factors may come in to play such as input impedance differences between the + and - inputs of the OP Amp. Also the capacitive feedback loop (R8 & C19) was have an effect. DC noise may also be an issue. It could even be the layout of the PCB tracks. Regards, Louis
Nice. Only thing I do not like too much is the offset nulling technique on the ADC inputs. Why not to power those buffers from -5V rail to reach true zero on their output instead?
if my memory is correct, the DAC has an EEPROM where you can programm the default start up voltage....so you dont have to fiddle arround by offsetting the inverting input of U7B by any value....simple tell the DAC it should output 0V at startup. Take a look at the datasheet how to accomplish that edit: i was to fast, you already mentioned that :)
Regarding the offset issue with those two op amps, wouldn't the required resistor values depend on the individual op amp? I'm surprised that you don't have a trimmer in there for those. And regarding the speeding up of the I2C bus, you don't say why you want to do this, or why the speed issue has become a problem. What prompted this?
Hi Louis just a suggestion it may be that when the encoder switch gets some wear it might switch bounce a bit and a by pass capacitor on the A B inputs should clear this up, a better way is to also add a series resistor and then the cap to D4& D2.
Fantastic Project Presentation! An avid follower... Got most parts, and almost finished my own boards However I am using negative photoresist and could do with a negative version of this new pcb. Would be greatly appreciated.
Thanks. I have made available an alternative PCB image for you which can be downloaded from the link below. It is in a zip file. Remember when printing to set printer to Actual Size. www.scullcom.com/DC_Load_PCB_v7_1_positive.zip Regards, Louis
Hi Louis, i'm in the middle of building this and looking for a power supply for this beast. I think you had mentioned in one of the videos a 15v? But i forget exactly, what voltage and current rating would you recommend to use? Thanks once again for the great videos, you're awesome. -Vic
Dear Louis Really like the project! Well done and thank you. Learnt a lot! Those surface mount components are very small. Isn't it possible to het through hole components? Welcome back! Hope everything went well! Best wishes Stephan
Hi Stephan, Thanks for your comments. I may have a look at a possible option which uses standard 8 pin DIP IC's in place of the surface mount ones. The SOT23-6 packages are very small and can be difficult to solder by hand. Regards, Louis
Hi Louis, Excellent work with this new version and thank you very much for publishing it. I also appreciate your clarity, simplicity and hardworking to make your project and video too. However, I tried your fan submodule and found it is a not as effective and requires a near high PWM value (>200) else it creates humming noise. It added a kind of DAC (RC using 1K resistor+plus 47uf Cap) followed by a buffer (a opamp unity gain) before biasing to base of the TIP122 and it worked well. I saw your Kicad version of file and its good. However, may I suggest you to use your DipTrace (www.diptrace.com) EDA instead. Its very easy to use, complete, instant 3d viewer, power autorouter, rich libraries etc. I like this verses earlier I was using Kicad. It is also available as free version for non-commercial use( I am using this). Very effect and prompt support by the product team and professionally used. They have a demo/video for a quick learning. This will help you to create a dual side PCB with ease.
Thanks for the comments and your suggestion regarding the fan. This may vary depending on what type of fan you use. Also thanks for the info on Dip Trace, I was not aware of this software. Regards, Louis
Most purchased DC-Loads require calibration on 1A and 10A to work precisely. However sourcing precise 10A in hobby land is difficult. Any help is appreciated.
Thanks, excellent job. I appreciate the amount of effort that goes into making these series of videos. Is there a way you could put a copy of the gerber files with a circuit board manufacturer so we could order one direct?
Thanks Joseph for your comments. I have made available my KiCad files for this part 7 in the link below. If you have KiCad installed (which is free software) you should be able to generate gerber files from the files. www.scullcom.com/DC_Load_v7_1_KiCad_files.zip Regards, Louis
Just one other point. If you are thinking of having a PCB manufacture then it may be better to design a double sided PCB which would avoid all the links as well as been smaller. My PCB designs are intended for the hobbyist to easily make themselves.
Hello Louis, excellent job,especially the new pcb is a good move, I waited since episode 2, now I will make my first selfmade pcb. Concerning the Kicad files. (I use Kicad version 4.0.5.). The programm complains that it cannot find two librarys, "special" and " DC_Load_Voltage_Sense-cache". Both libs do not sound like real libs, I am a total newbie with Kicad, is it my setup or are libs missing? All the best to you, U.
Hi Link 15V DC Power Supply HKL-5M15 AC DC Switch power supply module $5 they work very nice, choose 5v 12v 15v 🧐 so glad i found these videos 🥰a real project 🥳🍝☕
At 22:39 why going to -5v with 47k? This means that the transistor is always on! And there's no threshold effect and the -5V rail is maybe not strong enough to handle the current from multiple opamp outputs . Why not just using a npn to ground to limit to 6 or 7A per mosfet? To be achieve a better balance a solution could be to have a oa to drive each mosfet individually but that would increase parts and cost.
Thanks for your comments. I will have a further look at this part of the circuit. You could use a separate OP Amp to drive each Mosfet but that would also complicate the circuit and be more costly. The main thing is it seems to work quite well at the moment but I will look at any further improvements. Regards, Louis
Absolutely no problem Louis ! Ive got a couple of questions regarding modifying the build for my purposes, mainly setting up the unit for a maximum of 40V @ 10A, Would it be possible to pick your brain on this please Louis ? Best, Steve
Hi Steve, For a maximum of 40V @ 10A (400 watt) you would need to use higher rated Mosfets. There are a number of Linear L2 Mosfets from IXYS SEMICONDUCTOR such as the IXTX90N25L2 (cost about £20 each) which may be suitable (you may only need to use one). You would need a very good heatsink with fan. Also the gate drive voltage for these would be higher so the circuit would need to be modified to accommodate that.
I like your videos, they are very educational the best on youtube.Can you please design an capacitor meter using only discrete ic's with 7 segment display. I have been trying to make one without success.Thank you.
Hi :) I have ebike about 4kw on brushless dc motor. I want to make my controller stronger and i am just worried about my mosfets. They are 12 (4on each phase) I was thinking if its ok to double them (on new small board) and add more capacitors. Just simply solder wire from base of exisiting mosfet to base new mosfet. Do you think new mosfets can have some delay in time or you see other issues? I have space for make small board with mosfets on side my controller box
Does anyone happen to have an extra one of these pcbs I could purchase? I'm not setup to make my own and really want to give this a go. I'd be happy to pay for the board and shipping!
Good day sir! Indeed a beautiful project and very educational. I have one question tough. Since you provided for free the schematic and the PDF of the layout I suppose that your project does not have a copy right. I ask this question because if possible I would want the Gerber files. I want to build this project. I will not change, sell, rent, or any other activities from which I will benefit in any way.
Mitroi---I believe the gerber files are in the downloads. I plan on ordering a few boards myself to build. If you beat me to it I will be happy to buy some boards from you.
Flip-Flop Hi thanks for your comment. I have been busy with other work lately but I am now just putting a new project together and hope to film it in the next week. Regards, Louis
you said its dual power schottky diode. As the two diodes are on the same silicon die they are thermally coupled so it is unlikely to be an issue. Also both diodes are rated at 40 Amp so I do not think it will be a problem...there is a voltage drop in the diode, so when we sense the voltage (vout = Vout - Vd),sense voltage have drop. how to avoid this drop?
As we are testing a Power Supply the important point is to sense the voltage directly at the power supply under current load testing. Having a voltage drop across the dual power schottky diode will make no difference to the current loading we are applying to the Power Supply so it is not an issue.
Very nice job, excellent! I have also the question like rb: Your schematic (7.1) shows Q10 as a NPN 2N3904 while the parts list refers to a PNP 2N3906. I think the scematics leads, but can you clarify please.
Quick question for clarification. Your v7 schematic shows Q10 as a 2N3904 NPN while the parts list refers to a PNP 2N3906. I can't the PNP on the schematic and since the parts list doesn't have schematic codes I'm not sure which one is correct. I'm leaning towards the schematic as the master. Can you clarify? Thanks
This looks like a very feasible design. I have only one problem with it, there is no cutoff relay. When you're doing battery testing, you want at least 2 to cutoff methods. If for any reason the MOSFET's fail to turn off there is no way to stop the test. Batteries don't care if the alarm has sounded and are never off. Invariably you'll get distracted and fail to stop the test before you leave. The result is fire or at least a ruined battery.
I built this unit and the 5V regulator gets very warm indeed.I am going to add a second, smaller dedicated fan for the linear regulators. I couldn't find the original mosfets at a reasonable price, so I used IRLB4030PBF-ND, from DigiKey www.infineon.com/dgdl/irlb4030pbf.pdf?fileId=5546d462533600a4015356604640258d My load ON/OFF switch has an issues, if I just push the button once, the "OFF/ON" flickers back and forth, so I have to push it again to make it stay "ON". I will try a couple different switches/buttons to see if I can get it to turn ON without double-clicking. It may be a latching/ bounce issue, or other mistake I made. The original sketch limits the current to 4.000A, and a maximum of 50W's. I adjusted it up to 100W's / 5A's. I have a massive heat sink from an old power inverter with dual fans, and so far so good, at 24VDC. I'll scale up the input for higher voltages and try to get it to accept up to 60V ( in order to test a fully charged LI-ION 48V battery (on CC mode, not Battery mode, as battery mode is currently limited to just 1A). I want to scale it up to at least 300W at 10A's, like most commercially available models, but noticed I have to adjust more than just the software to do so...I'll check back once I have done all the mods. Cost of construction, with parts from digikey, $28 for (5) the PCB from JLCPCB, $138. Not a cheap build but well worth it if I can make to handle more than 50W ( note: the parts can easily handle the power/ current, but need to mod the build a bit and tweak the code). Code is written very well and laid out so any DIY'er can tweak it to their needs. Additional mods I am going to make: 3.5" LCD, custom 3D printed case, & DIY hand-wound, dual tap transformer for 15 & 7 VDC inputs (reduce heat at the 7805)Thanks again Mr. Louis, this was a very fun project.
Thanks for sharing the information on your build, interesting. If you are looking at going up to 300W then it may be worth considering using a Linear Power Mosfet such as the IXFN200N10P manufactured by IXYS SEMICONDUCTOR, they do a range of Linear Power Mosfets but they are not cheap. However, you may only need to use one. The gate drive circuit for these mosfets would need to be modified to provide the required drive voltages. I may have a look at this as an update to my design. Good luck with your own project. Regards, Louis
Thank you Sir. The credit is most certainly yours. I have been playing with different ideas and may even go to the TO-247 package ( I have some that I salvaged to test the idea). I will certainly check out the IXYS devices and have read a bit about their purpose made line. I moved the temp sensor to the middle of the heatsink and changed the ratio of the DAC voltage-divider. Pushed up to 20A briefly. Still more tests to run but most of the time I am just testing (a single) DIY 18650 battery bank at 4.20V so its still "low power," just higher current. I am pleasantly surprised how well the unit behaves. The next step is get the box printed and 3.5" LCD installed and add an SD card for data logging.
I noticed that when the fan kicks on at 30*C, the voltage across R6 goes up by 150mV which get sent to the gate of the FETs. I am not sure if this is a result of my modifications. Maybe something to visit in part 8? Much love.
Some types of fan can cause issues. Have you tried different types of cooling fan. You could try increasing the value of C16 to a 1uF to give some extra filtering on the drive voltage. Regards, Louis
I was just wondering if you checked out the output while the LOAD was attached before you turned on the "electronic DC load circuit." Obviously you could scope out the output while turning on the device to see if there is a current pulse....PEACE
ElectronicWizard That was what was happening before I made the modification. With the additional modification there was no current surge at switch on. Sorry I thought you question was relating to before the modification. The mod seems to work very well.
I did not check your design PCB size, but 100x100mm x10psc costs 2USD from china while 100x150mm cost 24USD, quite a good reason to fit to smaller size and perhaps use ribbon cables or other connectors to stack them etc. I checked PCB at jlcpcb.com/quote
Eric Groslevin Hi Eric, I have been busy for the last month with family business so had lack of spare time, but will be filming a new project soon and hope to upload it in the next week. Regards, Louis.
Thanks Maurice for your comment. Yes I have thought of data logging. I have been playing with an application which reads the serial data when using the DC Load in Battery Capacity testing mode and got it to draw a graph in real time on a PC screen. I used Electron which is a framework for creating native applications with web technologies like JavaScript, HTML, and CSS. This still needs further development. If you are interested then check out the Electron website at electron.atom.io/ Regards, Louis
Hi Louis, I have a question regarding the mosfet type. I have seen a Commercial DC load "Maynuo M9711"using the IRFP250 mosfet. Do you think this can be implemented in your project for higher power use with out too many complications? I understand the RDSon spec is10V @ the gate and is not a logic gate voltage as the one you use. Have you looked at this mosfet before? Thank you for your help. Regards joe
Hi Joe, The IRFP250 mosfet is not a logic level mosfet so it would not work in my design without some modification to the gate drive voltage for the mosfet. But it is possible to modify. I have not looked at this mosfet type sofar. Regards, Louis
Yes the Array 3711A may be using the IRFP250 mosfets. May be worth looking at using them and making some modifications to the gate drive circuit. Food for thought. Regards, Louis
This is only to say "Hi" and "good work" in all your videos. I hope you are doing good and, if you want, we can enjoy new videos soon. And I like "World of Genesis" too ! ;) But is difficult to read in that format. Don't have you PDF files ? Thanks anyway !
Thanks for your kind comment. The "World of Genesis" is my youngest sons work which I simply put on my website for him. Not sure what you meant about the PDF files? Regards, Louis
Oh ! So, please, send him my "Good work !" :) About the PDF thing, I mean I can't download (I think) "World of Genesis" in PDF to read more comfortable, i need to read in the webpage and is a bit difficult (zoom, navigation, page change, etc.). Thanks anyway and I'm glad you are back ;) Cheers from Spain!
Ah now I understand. When you hover with you mouse arrow on the comics page you will see a double arrow appear at the edge of the page in the middle,clicking on the double arrow then turns the pages. I will pass on your regards to him and mention you have been looking at his comics. Regards, Louis
Sofar I have tested this version with the 4 Mosfets with heatsink and fan and it seems to work without issues upto about 100 watt. Adding a much larger heatsink with a larger fan we should be able to achieve 150 watt. i may make further improvements later.
Hi Colin, Due to some security updates I think the site stopped working for a short time. All should be back to normal now. I have just checked and all the download links seem to work OK. Regards, Louis
You could use the CR (Constant Resistance Mode) but this is not the ideal solution for Testing LED Drivers. The best solution would be to provide a special LED Mode which would use a combination of CV (Constant Voltage) and CR (Constant Resistance) to closely mimic the actual behavior of an LED. I do not have a LED Mode on this project at the moment but it is something to look at for a future upgrade. Regards, Louis
If you have to buy all the parts then the basic cost is around £75. If you then want to add the same project box and large heatsink which I used then that would add an additional £40. If you shop around you can get many of the parts much cheaper on eBay from suppliers in China, but then you have to be prepared to wait a longer time for delivery. The total actual cost I spent was around £70 as I already had some basic components in my workshop. Regards, Louis
Scullcom Hobby Electronics thanks very much Louis. I will prob get all the bits and make it my self. Only thing I can’t do due to not having the bits is the making of the board. Many thanks. Steve
To read private messages that you've received on TH-cam, go to your Messages page. You can get there by clicking your account icon in the top right-hand corner then select Creator Studio. In the left-hand menu under "Community", click Messages.
Really like you channel thanks. However, Windows just warned that your web site is unsafe and on my iPad, the address in the browser went from the proper one to another to another to another to another without even showing anything.
Brilliant, been following you since the beginning of this project - have all the components now bought so the timing of this update is great - Going to try and produce a double sided PCB for it. I know other have said it but a big thank you for your effort in both the design and producing these videos. You have taught me lots !!
Hello Louis, i really enjoyed watching this nice project to evolve - thank you for your efforts! Though my own build of this still only runs "half" on a breadboard (waiting for parts :-), i already played around a bit with the software. Since at one point i want to try adding a nice(r) TFT, i made some changes to free up RAM and pins. Perhaps you may have a look and see, whether you want to make use of one or another modification as well: github.com/jlegen/Scullcom_DC_Load
I eill pay csn u make me a pwm dim led like hurricane access control LED lighting controller and allow Sunset I just want something that will give me sunset and sunrise even if it's a timer situation scenario slowly Dempsey If that's too hard I would like to by or give you the parts to make a power supply that has a digital readout that is just cheaper then the normal power supply thinking about using a computer power supply
Hi Jonny, I believe someone has already designed an LED controller with sunset and sunrise using an Arduino. I would recommend you have a look at what they did. Direct link to their details is below: reefuae.com/forum/viewtopic.php?f=13&t=1513&start=0 I may do a power supply project in the future but at the moment I am working on a number of other projects. Regards, Louis
Proper old school engineer, The correct use of "Pre-set" get a big thumbs up from me. Love the long videos and detailed explanation. Keep up the good work....
Thanks Mike for your comments.
Following this series, you can see a product evolve. Good old style engeneering indeed: We have a need. We find a (rather crude) first solution and constantly refine it. AND constantly revise bad design decisions (like the 0.2V "correction" for the DAC mid-value programming). THIS is what I love, and of course you very clearly explaining your thoughts and designs on the whiteboard.
Everyone starting out in the beautiful realm of electronics should watch this series!
Thanks for taking the time and sharing this with us!
Everyone deserves a great teacher like you. I wish you good health and great ideas.
Many thanks.
One of those times when a simple thumbs up cannot express what my thoughts are regarding this project. As has been mentioned in other comments, it started as a simple idea, but the additions and changes are welcome and extremely beneficial to the basic design without over complicating the design requirements. An incredible journey so far, presented and explained at a level and pace that anyone can follow along and hopefully learn from. Incredible work, explanation where needed and presentation. The next on my to-do list... Karl
Many thanks Karl for your kind comments.
Regards,
Louis
That is - hands down - one of the best projects on TH-cam, embedded in one of TH-cam's "bestest" video-series. You, Sir, have a gift of marrying your superlative know-how with the flow of ideas from the community. I cannot even emphasize any particular thing, as the whole series is just on a stellar level. Though I do admit, that I occasionally pause your videos to enjoy the schematics that you hand-draw so beautifully - but then again, I'm a loonie .. LOL Marvellous work!
Thanks for your comments, much appreciated.
Regards,
Louis
Wish there were more projects on TH-cam like this! Awesome quality and pace of explanation
Thank you.
After watching all of the video's I should make this project! After the improvements Louis and the new PCB, it should perform exactly as designed! I also love how you explained the improvements and or suggestions to use other parts to make it better! You are also open to inputs from your viewers, others should learn from you! Thank you again Louis!
Thanks Tony.
Great improvements, all down to the spot! I can't tell you how anxious I am to see this wonderful project come to its final version and build one myself. One thought crossed my mind is to change the display with a dot-matrix one. That way, the most important info can be shown with bigger fonts, and a better utilization on the display can be achieved. 128x64 or a little bigger ones does not cost much more than 20x4 character displays and only a small change in the software is enough to display everything better. Thanks again for this project and for your kind heart for sharing everything about it.
Thanks for your comments. A different display could be an option. I am looking at one or two other options. But need to fully test the new PCB on all functions.
Regards,
Louis
Hurray, you yourself found the effect of seperate emitter resistors on the gatedrivers. Chapeau!
Beautiful job, sir! It started as a simple project and now it is a rather sophisticated device :)
Hey Mr J!! Where do content producers like you get the time to watch heavy stuff like this? Never expected to find you here (but then you are welcome anywhere!)
LOL That's a funny one :)
I don't consider myself as a very active "content creator" as you say, but I'm on TH-cam each and every day. Mostly at night, I deprive myself of sleep, that's how I manage to watch these long videos :D Maybe you saw BigClive's "So much TH-cam. (No time for sleep.)", same thing ROFL
Thanks.
Hey Mr. J, congrats for passing half a million views! Keep up the natural personality (like the time that fly made you mad when working on something and you put it on the video..) ha! Move over Electroboom!
+Pomona Bill I don't think he meant something bad, it was said in a humorous fashion. But yeah, each to their own. I like Mehdi's videos myself and I'm not trying to snatch a piece of his cake. I guess Cliff just found my videos funny enough to compare to Mehdi's.
This truly is one of the best electronics diy series on the tube. Thanks for the hard work and hours you have put in to make this series! I was considering getting a commercial dc load from a brand manufacturer but this project has advanced to a stage that makes me consider this as very interesting alternative. Perhaps something to ponder about that current offset topic is going with 4 terminal sense resistor and INA on +/- supply. This would leave the D opamp free for low output impedance offset adjustment. Perhaps it will complicate the project too much but it would deliver very solid sense implementation and a capability to tackle wider current range with improved precision and lower drift as the internals get heated during operation. This would also open up the capability to do 2 automatic ranges by arduino controlled shorting of the INA gain resistor for 10x and 1x amplification (2A/1mA and 20A/10mA range), for example.
Thanks for your interesting comment and suggestion.
Very nice work. Thank you very much for all of your time and efforts in making this wonderful piece of tool that everyone desires for who are doing powersupply stuffs.
Many thanks for your kind comments.
Wow awesome work! You deserve much more subscribers.
Thank you.
akhurash, I absolutely agree with you. The problem is most of the people nowadays are more interested on channels which people buy stuff from China and play around with modules, sensors, etc. or assemble some kits. Nothing wrong with that, but sometimes I feel we are missing the essence of electronics. This channel is a rarity and it deserves to be one of the most viewed channels about electronics on TH-cam. I will be here giving all support I can. Love Scullcom hard work!!!
Robledo Einstein Yup. As a hobbyist and engineer I enjoy these types of videos. Building things from basic components is great to learn and understand electronics.
Excellent progress with the project.
I have also made several improvements but I never got to creating a proper schematic diagram. That is until you released this latest Kicad package and included the source files. I have never done anything in Kicad so I needed a jump start and having a design that is not very far from my own was a great jump start. I have only done the schematic. I have additional plans for enhancements I want to try before I make a PCB.
I have replaced both DAC and ADC with some faster/better ones. I have added hardware debounce of the rotary. I have replaced the fan by a Sanyo-Denki 4 pin type that can be speed controlled by a 25 kHz PWM signal. It is a BEAST and the load can now run with 250 W continuously with that single MOSFET in the SOT227 housing. Amazing. But at full load the noise is massive.
I have completely rewritten the software. I have changed the D key so it works as a menu key and moved battery check, transient, calibration, and user settings under this menu. No more pressing two buttons. And I now store the calibration in EEPROM. Because I replaced the ADC and DAC, my software is no longer compatible with the original design but it it not difficult to merge the functions that control ADC and DAC from the Louis software into mine.
It is a real interesting project and it is fun to see it develop.
And with the danger of TH-cam holding back my post as spam. Here is the URL again to my version of the load just updated today. Incl new schematic and new code. www.lavrsen.dk/foswiki/bin/view/Kenneth/ElectronicLoad
Thanks Kenneth for the update and info. I had been considering menu screens myself as an option.
Regards,
Louis
Hi. I now have an updated version of my design that also implements constant voltage. I have not updated my website yet. I need to update the Kicad files. But the principle is simple. I have added two analog switches on the inputs of the op amp that drives the MOSFET. When I change the mode to CV the DAC changes to controlling the inverting input and the non-inverting input is connected to the output of the opamp that monitors the voltage. And then it is simple code to add the constant voltage setting. I put that feature in the D menu. It would be easy to implement same in Louis' design. Very easy. I have to work a little on the bias zero circuit which drifts 1-2 mA now. Not a big deal. It just annoys me ;-)
I have now completed the next step of adding constant voltage mode to the load. This cannot be added as software because you just cannot regulate the current fast enough in a software loop.
So I have added to analog switches so that I can run either in the old current loop mode where the op-amp that drives the FET compares the shunt voltage with the DAC voltage. For constant voltage you have to move the DAC voltage to the inverting input of the op-amp and feed the load voltage to the non-inverting. And it happens to be that the voltages going into the ADC for the voltage measurement and the control voltage from the DAC are in the same range so it is actually a very simple exercise.
I have added the constant voltage mode under my "D-menu"
I found that you have to be very careful with cleaning flux away from the 4,7 Mohms resistor used for nulling the DC current. I was fighting drifting zero for hours until I realized it was the crappy chinese flux I had used for the rework adding the constant voltage circuit. After multiple cleanings with IPA it is stable again.
I have uploaded both PDF, Kikad project, and my least code to my wiki page (linked in my own comment earlier in this thread)
@@KennethLavrsen Hi Kenneth! Thank you your commitment and share to extend and modify this project. As far as I see you are the only one who provides downloadable materials for this project up until today, 22.20.2021. Good work!
Amazing and useful project. I'm following this series of videos. God job.
Thank you.
Great work Louis, thanks for the hard work, this is shaping up to be a cracking piece of bench test gear.... Keep it up!
Thanks Tony.
Hi, you are the best, please continue making stuff which is affordable to poor electronics hobbyists....
I have a question for you, what is your specialization in electronics, power electronics???
Hello Louis and everyone, just a comment i thought to mention, not sure if this was brought up before, but the I2C pull-ups are also implemented in the PCF8754 LCD control board. So if so these should be removed otherwise you will get 2.2k ohms on the data & clock lines. I came across it by accident when I bought some MCP4725 on a module from ebay which have pull-ups already on the board, this made me think to check the PCF8754 module.
Thanks for your comments. The correct value for a pull-up resistor is an interesting subject and may be worth discussing in more details on one of my videos.
In short, the maximum pull-up resistance is limited by the I2C bus capacitance. If the pull-up resistors values are too high, the I2C line may not rise to a logical high before it is pulled low (due to the RC time constant).
The speed of the I2C bus can be increased by lowering the value of the pull-up resistors, within limits which depends on the Vcc Supply to the circuit - in our case the supply voltage is +5V. At +5V you can normally reduce the pull-up resistor value to as low as about 1.8K. The trade off is that a lower value resistor means the circuit uses slightly more power but gives you a higher speed.
So if you happen to have two sets of pull-up resistors then as long as the overall value does not go below 1.8K you should be OK. Pull-up resistors on a PCF8754 LCD control board are usually either 10K or 4.7K, so there should not be an issue.
Hope my explanation is of help.
Regards,
Louis
Thank you for your kind explanation, so I suppose only if you have lots of module in parallel on the system and goes below 1.8K then its best to remove some resistors. Yes I understand your explanation the lower the resistance the faster the lines are at their logic level especially if you have the fast I2 communication switched on.
Kind Regards,
joe
Good job sir, thanks for sharing your knowledge.
Thanks. Your welcome.
Just discovered your fascinating channel and web site Louis, thanks for all the great videos. I have decided to tackle building the DC load but the Mosfets you chose are no longer manufactured, can you suggest an alternative to the BUK956R1-100E?
I wish I'd run across this video sooner. Your concern about the -5v power source is noted. It doesn't matter what it is, 4.1v for example, it just has to be stable. You're only using it to bias the op-amps to operate in the required range. As for I2C, I've always used 2.2K or 2.7K pull-up resistors and never had any problems. I've seen circuits with pull-ups as low as 1K. Why anyone wants to cut the current in the bus to such low levels is beyond me.
Thanks for your comments. Yes agree the main thing is that the negative voltage source is stable. With regards the pull-up resistors on the I2C bus using a 2.2K or 2.7K pull-up resistor will allow a faster speed on the bus. I think the lowest value you can go down to is 1.8K. The only drawback of lower value resistors is it draws more current but this should not really matter if your power supply can handle it. On the plus side you can run it at a higher clock speed. The normal default I2C clock speed for the Arduino is 100KHz but you can take it up to 400KHz when you use lower value pull-up resistors. As the LCD display module also has pull-up resistors then adding additional pull-up resistors as in this DC Load circuit we are reducing the overall pull-up resistor value and so allowing a higher bus speed, which is set in software.
Regards,
Louis
It's very useful and helpful project video ever seen
Happy New Year, Louis
for part 8 (I do hope you are still advancing the project) I would like to suggest to realize a data interface to collect the measurement data, especially for the battery testing use case. The pins TX/RX (D0,D1) for the Nano in your projekt are still unused. With a USB to Serial adapter based on a FT232R USB UART IC it wouldn't be too difficult to build a USB interface in the DCLoad. There are readymade adapters available on ebay. I guess you know it...
Feedbackloop 's (a youtuber) has shown in episode #10 a way to create a graph based on data collected from a dc load. The SW is available on github. Pretty cool...
What do you say? ;)
Hello, I write from Brazil. I have now reached your channel on youtube and I was very happy with the quantity and quality of the content, so I come first of all to thank and congratulate you for your work! I'm going to build your project, it was great. Sorry for my question, but what is the biggest voltage and power of the load?
In solving the issue with the DAC starting at half output voltage, you can just write the EEPROM default setting to be 0. I'm not sure I understand how adding that turn-on bias circuit is the simplest solution when there's a software fix, granted it works the very first time without having to power it up once and command a default setting, that does not seem like a necessary benefit.
I also wonder if, rather than trying to bias the input of the opamp on startup, it would make more sense to have a switched positive supply to that opamp. This is common for automotive motor drivers where there's a logic input to shut down the gate drive.
Why not just store 0 in the EEPROM of the DAC and do away with the the whole 2V business?
4K7 pull-up resistors on the I2C lines are pretty much standard in the hobby robotics world. They usually place it at the end of the bus.
Thanks for the KiCad files and all the hard work you keep pouring into this project. Its a perfect project for me and I am about half way there. It's such a useful project, unlike blinking LEDs and fidget spinners which are all the rage.
Thanks. As I said on the video I will look in to the possibilities of reprogramming the default setting in the eeprom to see if that can be easily done.
Regards,
Louis
Yes. I will Program the EEPROM on mine. Less hardware to go bad that way.
Very cool device! Thank you!
This is a most interesting project, and becoming quite sophisticated. A couple of things jumped out at me while watching though:-
The protection diodes in parallel. Although they are on the same die, there may still be enough difference in forward voltage drop to cause one diode to hog all the current. Maybe a very low value ballast resistor in each anode leg could be added. Though I suppose even just one diode at 40A is sufficient...
I'm not sure if the drop across the diode(s) (and any added resistor(s)) is of any significance in this application, but if so, or in any case, maybe have a reversed diode across the load terminals (via the fuse) as further protection against reverse polarity.
The need for null current trim. With op-amps that have 1uV offset voltage, I was surprised to see this was needed. Could this trouble be caused by a single op-amp trying to balance the behavior of the 4 mosfet circuits? And might separate op-amps per mosfet (as someone else suggested) help in this regard? (When I finally do get around to building something along these lines, I will certainly be looking into that...)
Thanks for all your work on this!
Thanks for your comments.
With regards the dual power schottky diode. As the two diodes are on the same silicon die they are thermally coupled so it is unlikely to be an issue. Also both diodes are rated at 40 Amp so I do not think it will be a problem.
With regards the offset null adjustment. This requirement was found to be needed for either a single Mosfet or 4 Mosfets output stage. Although we are using an OP Amp with a 1uV offset, other factors may come in to play such as input impedance differences between the + and - inputs of the OP Amp. Also the capacitive feedback loop (R8 & C19) was have an effect. DC noise may also be an issue. It could even be the layout of the PCB tracks.
Regards,
Louis
You do very nice clear videos, good job.
Thanks.
Nice. Only thing I do not like too much is the offset nulling technique on the ADC inputs. Why not to power those buffers from -5V rail to reach true zero on their output instead?
Thanks for the comment. That is another option.
Excellent job sir! Thanks!
Thank you.
if my memory is correct, the DAC has an EEPROM where you can programm the default start up voltage....so you dont have to fiddle arround by offsetting the inverting input of U7B by any value....simple tell the DAC it should output 0V at startup. Take a look at the datasheet how to accomplish that
edit: i was to fast, you already mentioned that :)
Thanks Ray. Yes as you say I have already mentioned that as an option.
Thank you, amazing contribution!
Regarding the offset issue with those two op amps, wouldn't the required resistor values depend on the individual op amp? I'm surprised that you don't have a trimmer in there for those. And regarding the speeding up of the I2C bus, you don't say why you want to do this, or why the speed issue has become a problem. What prompted this?
Hi Louis just a suggestion it may be that when the encoder switch gets some wear it might switch bounce a bit and a by pass capacitor on the A B inputs should clear this up, a better way is to also add a series resistor and then the cap to D4& D2.
Thanks for the suggestion. Regards, Louis.
Another cracking video thanks sir!
Thanks Simon.
Fantastic Project Presentation! An avid follower...
Got most parts, and almost finished my own boards
However I am using negative photoresist and could do with
a negative version of this new pcb. Would be greatly appreciated.
Thanks. I have made available an alternative PCB image for you which can be downloaded from the link below. It is in a zip file. Remember when printing to set printer to Actual Size.
www.scullcom.com/DC_Load_PCB_v7_1_positive.zip
Regards,
Louis
Thank You so much for such a quick response! Greatly appreciate the amount of effort put into this project..
Hi Louis, i'm in the middle of building this and looking for a power supply for this beast. I think you had mentioned in one of the videos a 15v? But i forget exactly, what voltage and current rating would you recommend to use? Thanks once again for the great videos, you're awesome.
-Vic
Hi Vic, The 15v power adapter I am using is 1 Amp rated, but it should also be OK with a lower current rating say of 0.5 Amp. Regards, Louis
Dear Louis
Really like the project! Well done and thank you.
Learnt a lot!
Those surface mount components are very small.
Isn't it possible to het through hole components?
Welcome back!
Hope everything went well!
Best wishes
Stephan
Hi Stephan,
Thanks for your comments. I may have a look at a possible option which uses standard 8 pin DIP IC's in place of the surface mount ones. The SOT23-6 packages are very small and can be difficult to solder by hand.
Regards,
Louis
Hi Louis,
Excellent work with this new version and thank you very much for publishing it. I also appreciate your clarity, simplicity and hardworking to make your project and video too.
However, I tried your fan submodule and found it is a not as effective and requires a near high PWM value (>200) else it creates humming noise. It added a kind of DAC (RC using 1K resistor+plus 47uf Cap) followed by a buffer (a opamp unity gain) before biasing to base of the TIP122 and it worked well.
I saw your Kicad version of file and its good. However, may I suggest you to use your DipTrace (www.diptrace.com) EDA instead. Its very easy to use, complete, instant 3d viewer, power autorouter, rich libraries etc. I like this verses earlier I was using Kicad. It is also available as free version for non-commercial use( I am using this). Very effect and prompt support by the product team and professionally used. They have a demo/video for a quick learning. This will help you to create a dual side PCB with ease.
Thanks for the comments and your suggestion regarding the fan. This may vary depending on what type of fan you use.
Also thanks for the info on Dip Trace, I was not aware of this software.
Regards,
Louis
Thank you very much, excellent video.
Thanks.
Great work.
Thank you.
Thanks.
Most purchased DC-Loads require calibration on 1A and 10A to work precisely. However sourcing precise 10A in hobby land is difficult. Any help is appreciated.
Thanks, excellent job. I appreciate the amount of effort that goes into making these series of videos. Is there a way you could put a copy of the gerber files with a circuit board manufacturer so we could order one direct?
Thanks Joseph for your comments. I have made available my KiCad files for this part 7 in the link below. If you have KiCad installed (which is free software) you should be able to generate gerber files from the files.
www.scullcom.com/DC_Load_v7_1_KiCad_files.zip
Regards,
Louis
Just one other point. If you are thinking of having a PCB manufacture then it may be better to design a double sided PCB which would avoid all the links as well as been smaller. My PCB designs are intended for the hobbyist to easily make themselves.
Hello Louis,
excellent job,especially the new pcb is a good move, I waited since episode 2, now I will make my first selfmade pcb.
Concerning the Kicad files. (I use Kicad version 4.0.5.). The programm complains that it cannot find two librarys, "special" and " DC_Load_Voltage_Sense-cache". Both libs do not sound like real libs,
I am a total newbie with Kicad, is it my setup or are libs missing?
All the best to you,
U.
Hi Link 15V DC Power Supply HKL-5M15 AC DC Switch power supply module $5 they work very nice, choose 5v 12v 15v 🧐 so glad i found these videos 🥰a real project 🥳🍝☕
At 22:39 why going to -5v with 47k? This means that the transistor is always on! And there's no threshold effect and the -5V rail is maybe not strong enough to handle the current from multiple opamp outputs . Why not just using a npn to ground to limit to 6 or 7A per mosfet? To be achieve a better balance a solution could be to have a oa to drive each mosfet individually but that would increase parts and cost.
Thanks for your comments. I will have a further look at this part of the circuit. You could use a separate OP Amp to drive each Mosfet but that would also complicate the circuit and be more costly. The main thing is it seems to work quite well at the moment but I will look at any further improvements.
Regards,
Louis
Fantastic work !!
Thank you.
Absolutely no problem Louis !
Ive got a couple of questions regarding modifying the build for my purposes, mainly setting up the unit for a maximum of 40V @ 10A, Would it be possible to pick your brain on this please Louis ?
Best,
Steve
Hi Steve,
For a maximum of 40V @ 10A (400 watt) you would need to use higher rated Mosfets. There are a number of Linear L2 Mosfets from IXYS SEMICONDUCTOR such as the IXTX90N25L2 (cost about £20 each) which may be suitable (you may only need to use one). You would need a very good heatsink with fan. Also the gate drive voltage for these would be higher so the circuit would need to be modified to accommodate that.
I like your videos, they are very educational the best on youtube.Can you please design an capacitor meter using only discrete ic's with 7 segment display. I have been trying to make one without success.Thank you.
Hi :) I have ebike about 4kw on brushless dc motor. I want to make my controller stronger and i am just worried about my mosfets. They are 12 (4on each phase) I was thinking if its ok to double them (on new small board) and add more capacitors. Just simply solder wire from base of exisiting mosfet to base new mosfet. Do you think new mosfets can have some delay in time or you see other issues? I have space for make small board with mosfets on side my controller box
Top-notch!
Sir please make a video on designing and building a DC power supply unit.
Thanks again for another good video / tutorial....PEACE
Thank you.
Does anyone happen to have an extra one of these pcbs I could purchase? I'm not setup to make my own and really want to give this a go. I'd be happy to pay for the board and shipping!
Good day sir! Indeed a beautiful project and very educational. I have one question tough. Since you provided for free the schematic and the PDF of the layout I suppose that your project does not have a copy right. I ask this question because if possible I would want the Gerber files. I want to build this project. I will not change, sell, rent, or any other activities from which I will benefit in any way.
Mitroi---I believe the gerber files are in the downloads. I plan on ordering a few boards myself to build. If you beat me to it I will be happy to buy some boards from you.
Good night my friend, how are you?
How much will we have new videos?
Michel -Rio de Janeiro - Brazil
Flip-Flop Hi thanks for your comment. I have been busy with other work lately but I am now just putting a new project together and hope to film it in the next week. Regards, Louis
Another nice video.
Thanks.
you said its dual power schottky diode. As the two diodes are on the same silicon die they are thermally coupled so it is unlikely to be an issue. Also both diodes are rated at 40 Amp so I do not think it will be a problem...there is a voltage drop in the diode, so when we sense the voltage (vout = Vout - Vd),sense voltage have drop. how to avoid this drop?
As we are testing a Power Supply the important point is to sense the voltage directly at the power supply under current load testing. Having a voltage drop across the dual power schottky diode will make no difference to the current loading we are applying to the Power Supply so it is not an issue.
Very nice job, excellent!
I have also the question like rb: Your schematic (7.1) shows Q10 as a NPN 2N3904 while the parts list refers to a PNP 2N3906.
I think the scematics leads, but can you clarify please.
Sorry about that it should read 2N3904 in the parts list - silly error by me.
Regards,
Louis
Hi Louis,
That's what I expected, thanks a lot.
Btw, I love the professional way of your channel, please continue...
Greetings from Germany,
Wolfgang
Quick question for clarification. Your v7 schematic shows Q10 as a 2N3904 NPN while the parts list refers to a PNP 2N3906. I can't the PNP on the schematic and since the parts list doesn't have schematic codes I'm not sure which one is correct. I'm leaning towards the schematic as the master. Can you clarify? Thanks
Sorry my error in the parts list. It should read 2N3904.
This looks like a very feasible design. I have only one problem with it, there is no cutoff relay. When you're doing battery testing, you want at least 2 to cutoff methods. If for any reason the MOSFET's fail to turn off there is no way to stop the test. Batteries don't care if the alarm has sounded and are never off. Invariably you'll get distracted and fail to stop the test before you leave. The result is fire or at least a ruined battery.
Before you leave-----??? Why would you test any thing and then walk away from it??
I built this unit and the 5V regulator gets very warm indeed.I am going to add a second, smaller dedicated fan for the linear regulators. I couldn't find the original mosfets at a reasonable price, so I used IRLB4030PBF-ND, from DigiKey www.infineon.com/dgdl/irlb4030pbf.pdf?fileId=5546d462533600a4015356604640258d
My load ON/OFF switch has an issues, if I just push the button once, the "OFF/ON" flickers back and forth, so I have to push it again to make it stay "ON". I will try a couple different switches/buttons to see if I can get it to turn ON without double-clicking. It may be a latching/ bounce issue, or other mistake I made. The original sketch limits the current to 4.000A, and a maximum of 50W's. I adjusted it up to 100W's / 5A's. I have a massive heat sink from an old power inverter with dual fans, and so far so good, at 24VDC. I'll scale up the input for higher voltages and try to get it to accept up to 60V ( in order to test a fully charged LI-ION 48V battery (on CC mode, not Battery mode, as battery mode is currently limited to just 1A). I want to scale it up to at least 300W at 10A's, like most commercially available models, but noticed I have to adjust more than just the software to do so...I'll check back once I have done all the mods. Cost of construction, with parts from digikey, $28 for (5) the PCB from JLCPCB, $138. Not a cheap build but well worth it if I can make to handle more than 50W ( note: the parts can easily handle the power/ current, but need to mod the build a bit and tweak the code). Code is written very well and laid out so any DIY'er can tweak it to their needs. Additional mods I am going to make: 3.5" LCD, custom 3D printed case, & DIY hand-wound, dual tap transformer for 15 & 7 VDC inputs (reduce heat at the 7805)Thanks again Mr. Louis, this was a very fun project.
Thanks for sharing the information on your build, interesting. If you are looking at going up to 300W then it may be worth considering using a Linear Power Mosfet such as the IXFN200N10P manufactured by IXYS SEMICONDUCTOR, they do a range of Linear Power Mosfets but they are not cheap. However, you may only need to use one. The gate drive circuit for these mosfets would need to be modified to provide the required drive voltages. I may have a look at this as an update to my design. Good luck with your own project. Regards, Louis
Thank you Sir. The credit is most certainly yours. I have been playing with different ideas and may even go to the TO-247 package ( I have some that I salvaged to test the idea). I will certainly check out the IXYS devices and have read a bit about their purpose made line. I moved the temp sensor to the middle of the heatsink and changed the ratio of the DAC voltage-divider. Pushed up to 20A briefly. Still more tests to run but most of the time I am just testing (a single) DIY 18650 battery bank at 4.20V so its still "low power," just higher current. I am pleasantly surprised how well the unit behaves. The next step is get the box printed and 3.5" LCD installed and add an SD card for data logging.
I noticed that when the fan kicks on at 30*C, the voltage across R6 goes up by 150mV which get sent to the gate of the FETs. I am not sure if this is a result of my modifications. Maybe something to visit in part 8? Much love.
Some types of fan can cause issues. Have you tried different types of cooling fan. You could try increasing the value of C16 to a 1uF to give some extra filtering on the drive voltage. Regards, Louis
I was just wondering if you checked out the output while the LOAD was attached before you turned on the "electronic DC load circuit." Obviously you could scope out the output while turning on the device to see if there is a current pulse....PEACE
Yes I did and got a current surge for just a second or two and then it faded away.
That's no good is it. Wasn't your modified circuit supposed to prevent that?....PEACE
ElectronicWizard That was what was happening before I made the modification. With the additional modification there was no current surge at switch on. Sorry I thought you question was relating to before the modification. The mod seems to work very well.
OK got it. I would have been nice to have seen the before and after wave forms.....PEACE
I did not check your design PCB size, but 100x100mm x10psc costs 2USD from china while 100x150mm cost 24USD, quite a good reason to fit to smaller size and perhaps use ribbon cables or other connectors to stack them etc. I checked PCB at jlcpcb.com/quote
Where are youu, miss your videos.
Eric Groslevin Hi Eric, I have been busy for the last month with family business so had lack of spare time, but will be filming a new project soon and hope to upload it in the next week. Regards, Louis.
Have you thought about making a isolated serial interface and then make a nice PC Software for logging and controlling?
Thanks Maurice for your comment. Yes I have thought of data logging. I have been playing with an application which reads the serial data when using the DC Load in Battery Capacity testing mode and got it to draw a graph in real time on a PC screen. I used Electron which is a framework for creating native applications with web technologies like JavaScript, HTML, and CSS. This still needs further development. If you are interested then check out the Electron website at electron.atom.io/
Regards,
Louis
Have you published your test code? A public git repository would make it possible that your viewers can develop the software...
Hi Louis, I have a question regarding the mosfet type. I have seen a Commercial DC load "Maynuo M9711"using the IRFP250 mosfet. Do you think this can be implemented in your project for higher power use with out too many complications? I understand the RDSon spec is10V @ the gate and is not a logic gate voltage as the one you use. Have you looked at this mosfet before? Thank you for your help. Regards joe
Hi Joe,
The IRFP250 mosfet is not a logic level mosfet so it would not work in my design without some modification to the gate drive voltage for the mosfet. But it is possible to modify. I have not looked at this mosfet type sofar. Regards,
Louis
Thanks for your response much appreciated. Just a guess and could be wrong, but your Array 3711A could be using this mosfet.
Yes the Array 3711A may be using the IRFP250 mosfets. May be worth looking at using them and making some modifications to the gate drive circuit. Food for thought. Regards, Louis
Sorry for the Late reply, yes I thought so thanks Louis. Regards, joe
ISL7660 is a nice IC, can we make the circuit by ourselves?
Hi,
You could make a DC to DC converter with an NE555 and some components if you wish. But it would be cheaper to use the ICL7660.
Regards,
Louis
This is only to say "Hi" and "good work" in all your videos. I hope you are doing good and, if you want, we can enjoy new videos soon. And I like "World of Genesis" too ! ;) But is difficult to read in that format. Don't have you PDF files ? Thanks anyway !
Thanks for your kind comment. The "World of Genesis" is my youngest sons work which I simply put on my website for him. Not sure what you meant about the PDF files? Regards, Louis
Oh ! So, please, send him my "Good work !" :) About the PDF thing, I mean I can't download (I think) "World of Genesis" in PDF to read more comfortable, i need to read in the webpage and is a bit difficult (zoom, navigation, page change, etc.). Thanks anyway and I'm glad you are back ;) Cheers from Spain!
Ah now I understand. When you hover with you mouse arrow on the comics page you will see a double arrow appear at the edge of the page in the middle,clicking on the double arrow then turns the pages. I will pass on your regards to him and mention you have been looking at his comics. Regards, Louis
How many Watts does this version allow to dissipate, is increasing wattage as simple as adding more mosfets?? PC logging connection?
Sofar I have tested this version with the 4 Mosfets with heatsink and fan and it seems to work without issues upto about 100 watt. Adding a much larger heatsink with a larger fan we should be able to achieve 150 watt. i may make further improvements later.
PC logging connection would be best appreciated I think.
Why when I click the links in the description,do no files exist or are displayed ?
Hi Colin, Due to some security updates I think the site stopped working for a short time. All should be back to normal now. I have just checked and all the download links seem to work OK. Regards, Louis
sir, Do you have CR-LED mode on this device? if it possible please share with me
You could use the CR (Constant Resistance Mode) but this is not the ideal solution for Testing LED Drivers. The best solution would be to provide a special LED Mode which would use a combination of CV (Constant Voltage) and CR (Constant Resistance) to closely mimic the actual behavior of an LED. I do not have a LED Mode on this project at the moment but it is something to look at for a future upgrade.
Regards,
Louis
Thanks for your replay...Once you upgrade the device with the feature of LED MODE its more useful for me...
Roughly how much did it cost for all parts?
If you have to buy all the parts then the basic cost is around £75. If you then want to add the same project box and large heatsink which I used then that would add an additional £40.
If you shop around you can get many of the parts much cheaper on eBay from suppliers in China, but then you have to be prepared to wait a longer time for delivery.
The total actual cost I spent was around £70 as I already had some basic components in my workshop.
Regards, Louis
Scullcom Hobby Electronics thanks very much Louis. I will prob get all the bits and make it my self. Only thing I can’t do due to not having the bits is the making of the board.
Many thanks.
Steve
Hi Steve, I may be able to help. I have sent you a message in the private messages section of you tube hope you get it. Regards, Louis
Scullcom Hobby Electronics hi Louis. I’m not sure where the private massage part is on TH-cam. Maybe it’s not on the app?? Steve
To read private messages that you've received on TH-cam, go to your Messages page. You can get there by clicking your account icon in the top right-hand corner then select Creator Studio. In the left-hand menu under "Community", click Messages.
Really like you channel thanks.
However, Windows just warned that your web site is unsafe and on my iPad, the address in the browser went from the proper one to another to another to another to another without even showing anything.
Thanks for the info. I just checked my website this morning and it seems OK now. So not sure what happened. I will check further. Regards, Louis
Brilliant, been following you since the beginning of this project - have all the components now bought so the timing of this update is great - Going to try and produce a double sided PCB for it.
I know other have said it but a big thank you for your effort in both the design and producing these videos. You have taught me lots !!
Thank you. A double sided PCB would be a nice improvement.
Hello Louis, i really enjoyed watching this nice project to evolve - thank you for your efforts! Though my own build of this still only runs "half" on a breadboard (waiting for parts :-), i already played around a bit with the software. Since at one point i want to try adding a nice(r) TFT, i made some changes to free up RAM and pins. Perhaps you may have a look and see, whether you want to make use of one or another modification as well: github.com/jlegen/Scullcom_DC_Load
Hi,
Thanks for your comments and your software update. I will have a look at it.
Many thanks.
Louis
Big Master pwm car battery charger normal and deep cycle pwm mptt charger make ...
I eill pay csn u make me a pwm dim led like hurricane access control LED lighting controller and allow Sunset I just want something that will give me sunset and sunrise even if it's a timer situation scenario slowly Dempsey
If that's too hard I would like to by or give you the parts to make a power supply that has a digital readout that is just cheaper then the normal power supply thinking about using a computer power supply
Hi Jonny,
I believe someone has already designed an LED controller with sunset and sunrise using an Arduino. I would recommend you have a look at what they did. Direct link to their details is below:
reefuae.com/forum/viewtopic.php?f=13&t=1513&start=0
I may do a power supply project in the future but at the moment I am working on a number of other projects.
Regards,
Louis