This project goes from excellent to fabulous. I wish I had a electronics teacher like you in my college back then. I wait every new video with great expectation.

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks for your kind comment. Regards, Louis

Another very nice update, Louis. You've developed the most usable, and useful, DIY load I've ever seen. Well done! Just one comment about MOSFET selection. You spent a couple of minutes discussing Rds(on), but that's not really germane to this application. Rds(on) is characterized with the MOSFET turned hard on (often at Vgs=10V), not when it's used in its linear region. In a DC load, we purposely operate the MOSFET at a gate voltage that results in a drain-source resistance greater than its minimum value, making Rds(on) irrelevant. The more important thing to look at in the data sheet is the safe operating area (SOA) graph. This relates Vds to drain current, showing the power dissipation capability of the device over time. For a DC load like we're discussing here, your application should fall within the DC region of the SOA graph If you Google "linear MOSFET," the first non-ad result is an Infineon application note that explains this concept in greater detail. Several manufacturers offer devices that are tailored for linear mode operation, most notably IXYS (see article further down in the Google results).

@scullcomhobbyelectronics1702

7 жыл бұрын

Thank you. As you say the Safe Operating Area is the most important data to look at for this type of operation. I did cover that in detail in part 2. What I should have said in this video is that I compared the SOA graph for each of the mosfets as well as looking at the headline data. I have had a look at some of the Linear Mosfets from IXYS Semiconductor which may be an option for the future. They need higher drive voltages and so the circuit would need some changes, they are also very expensive (usually around £12 each). I have found lower cost one from IXYS for about £6 which is the IXTP15N50L2. I have ordered a couple and will have a play with them. Thanks again for your comments and interest. Regards, Louis

Great set of Fluke DMM...a 85, 87 (first series ?) and newest Fluke 87V...great project for a great electronic engineer like you Louis.

First I want to say thanks for a great project. It has been a great inspiration for my own build of an electronic load Some differences I did I used a single FET - IXTN60N50L2. It is a device made for this kind of linear applications with a high forward bias safe operating area. It is in a SOT227. The kind of big monster package that can make small children cry of fear. :-) It requires a gate drive above 5 V so I added an additional OPAMP. I used LMC7101 because I had it lying around and it can run both inputs all the way down to ground. I use the 12V DC as power supply for the LMC7101. I wanted a higher power rating so I use a 0.01 ohm 1% four terminal resistor so I can run up to at least 20 A. As heat sink I used a heat sink Fischer Elektronik LA-6-150 Conrad Electronic P/N 166111 which is 0.3 K/W with a 60x60 mm fan mounted at the end. With this single FET and this heat sink I can draw 140 W permanently with the heatsink at 82-83 deg C. I set the limit at 85 deg C. The IXTN60N50L2 mounting surface is isolated from the silicon so you do not waste precious K/W from isolation mica. I can run the load at 13.8 V / 15 A and probably also at 20 A for several minutes - enough time to do load measurements on a power supply. You have made some provisions in software to eliminate offset. This offset is actually the result of input offset of the AD8630 and the problem for it to pull output completely to ground. It is only 1-4 mV we talk about but this is enough to prevent the load to go properly to zero A on the FET. I made two solutions. First - quite many rail to rail CMOS based opamps can go to perfect zero if you put a resistor on the output of the opamp to a negative voltage. By experiment I found that putting 220 k resistors on the outputs to -5V from the ICL7660 made the AD8630 go perfectly to ground. I still had problem with the D/A MCP4725 which also has an output offset. You cannot compensate for this in software because you still end up with a current flowing in the FET. My solution was to make a -2.45 V voltage with an LM431/3.9K resistor to the -5V V. And then I put a 10K multiturn potentiometer between the 4.096V and the -2.45 V. Both these voltages will be very stable over temperature. The wiper from the 10k potentiometer is connected via a 4.7 Mohms resistor to the inverting input on the opamp that drives the FET. This way I can null the offset from both opamp and DAC to perfect zero. For the voltage sense I used OPA334AIDBVTG-4 because I happended to have one around. It is an opamp with very low input offset. I run it on single 5V DC supply. To get the output to perfect zero I put a 2k resistor to the -5V from the ICL7660. That gives me a perfect zero volt on the output. It is actually the data sheet for the OPA334 that gave the hint of putting a pull down resistor to a negative rail. For the OPA334 they recommend 2k. For the AD8630 I experiented my way to the 220 k. I connect the OPA334 output directly to the MCP3426 ADC. The AD8630 just added offset and is not really needed when you use the OPA334 with single supply and the pull down to negative trick. For the fan control I chose to use a FET 2N7000 instead of the BJT. I had a big problem with the startup current. The first second the opamp that drives the FET turns on the FET 100% putting an almost short on the DUT. To control the FET during start I made these changes. I moved the temperature sensor from A0 to A6 on the Arduino Nano. A6 can only be an input. A0 can be used as output. So the LM35 on A6 frees up A0 to switch force the load off. The A0 is connected to a 10k pull up to 5V. And also to a 10k to the base of an NPN MMBT3804 NPN which has emitter to ground. The connector has 22k to +5V. And the collector is also connected via 10k to the base of an MMBT3806 PNP. The PMP has emitter to 5V and its collector is connected via 100k to the inverting input of the opamp that drives the FET. The result is that when the Arduino powers on the A0 default to input. The pullup turns on the NPN. The NPN turns on the PMP and the PMP pulls the inverting input higher than the non-inverting input can ever get turning the FET totally off. In the software I set the A0 high and change it to an output and in the code that sets the D/A to set the current I change A0 to LOW when the current is anything alse than 0 and the load is turned on. Last thing There is a bug in the software fanControl function. You have two conditions. One when temperature is below tempMin and one when the temperature is between tempMin and tempMax. However there is a valid termperature interval above tempMax that that is ignorred so the display stops updating the temperature when you hit tempMax. I added a if (temp > tempMax) { section that just sets the fanspeed to 255. And I moved the 4 lines updated the LCD to after so they are always executed and temperature is always updated. I hope this long comment is a help for you Louis and for anyone building along. Kenneth

@javierpallalorden

7 жыл бұрын

Hi Kenneth, you should make a video with your modifications, some are very useful and I would like to see it in practice. Thank you for sharing!

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks Kenneth for your detailed description of your changes to this project. I am sure others following my videos will benefit from the details of your version. Using a Linear Power Mosfet will give you the additional power, as these mosfets specially designed for linear mode operation so they are very suitable for DC loads. However, they are very expensive and required additional drive circuitry as you point out. The IXTN60N50L2 you are using costs about £24. I have been playing with the IXTP15N50L2 which is cheaper at about £5 but lower power rating but can still give me about 250 watt or more. Trimming out the OP Amp input offset voltages does improve calibration I have been testing its use on the OP Amp feeding the Power Mosfet similar to what you have been doing this will remove the need for software calibration factor which I had in my Arduino code. Thanks for pointing out the bug in the fanControl function, I will make the change in the next software update. I am currently having another look at the fanControl as there are some other issues which can effect the load voltage reading slightly on the LCD readout when the fan turns on. I will discuss this on my next video update. I am just finishing the trigger input circuity and will also cover that in my next video. Thanks again for sharing in detail the changes you have made. Best regards, Louis

I subscribe to a lot of electronics channels. This is the best one on youtube !

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks.

@victorkaplansky

7 жыл бұрын

scott how - Definitely!

@satishkumar-gg5fm

3 жыл бұрын

Are you all right???

Thank you for the excellent videos. Your detailled explanation proves to be very valuable to me! Keep up the good work 👌

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks for your comments.

Part 5! Thank you Louis, excellent update, my own has now made it into a large case, looking forward to upgrading the fet drive and code... Superb.....

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks Tony.

Fantastic! You just amaze me with more and more improvements! Thanks again! Love your channel.

@scullcomhobbyelectronics1702

6 жыл бұрын

Thanks Tony. Regards, Louis

this is a wonderful hobbyist with engineering procedures project evolution 😃so many projects are kind-of close and need the last details improved on 🧐thanks for the extra mile 🥳👌

Enjoying this series! Tip: make ‘0’ the Exit in ALL menu’s; now it changes over time when more options are added….

Thank you so much for this. This is definitely on my To-Do list as well on my list of videos to translate so that Brazilians can have a bit of your wisdom.

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks Bruno.

Thank you very much for the update, I'm learning a lot from your channel.

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks.

I like the way how you explain everything VERY detailed. Nice videos!

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks.

@Roy_Tellason

4 жыл бұрын

@@scullcomhobbyelectronics1702 Yeah, but I could do with a little less repetition, as there are only so many hours in a day and hearing the same thing as what I've heard a bit ago isn't that terribly productive...

Very impressive project, thanks for sharing it!.

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks

A couple of more small changes I forgot. The MCP4725 A0 input. I changed the pull up so it goes to VDD (4.096) instead of 5V. I was in doubt if it was OK to pull up to a higher voltage than the IC is running at. I tied A0 directly to the VDD pin with no resistor. On the original design the MCP4725 is connected to AD8650 unity gain buffer and the the voltage is divided by 9k/1k high precision resistors. I needed more reduction because I design for 20 A on a 0.01 ohms shunt. And I did not have the 0.1 % resistors to go with it. So I used a good 27k 1% film resistor to the non-inverting input of the opamp that drives the FET and a 2k multi-turn potentiometer to ground. This way I can set the current accurately with the potentiometer. So to calibrate the circuit you first set the output current with the 2k at a high current value using a good amp meter to verify the current. Then you turn off the load and set the zero amps on the other 10k multiturn potentiometer that I described in previous post (the one that goes from -2.45 to +4.096). You turn it so current starts to flow and then turn back till you reach zero. Then you have perfect bios elimination. And then you go back and set the current setting of the 2k potentiometer to correct the current to accurate reading. The AD8630 that measures the current over the shunt is still unity gain in my design. I just increased the gain to maximum in the MCP3426 and that works OK. Again the output of this opamp is pulled down to -5V from the 7660 via 220k to get it to perfect zero volts. With that I never have to use the zero offset settings on the software. I have perfect zero reading on both voltage and current. Kenneth

@KennethLavrsen

7 жыл бұрын

I will try and put either some video or download material online. I need to make a clean schematic of my changes. Right now it is spread as marked up schematic and notebook pages. And I have plans for making a real PCB for it. I built it all on small breakout PCBs. 10 to 12 of them. I also have an idea to add an additional constant voltage loop for testing chargers. That is the only real pro feature missing now. It does everything else.

@scullcomhobbyelectronics1702

7 жыл бұрын

Hi Kenneth, Thanks for the additional info. I have found that as long as you take A0 of MCP4725 high via a limiting resistor it is OK to connect to either 5V or 4.096V, it does not matter. You mentioned in the last paragraph connecting the output of the OP Amp to the -5V via a 220K (to eliminate the input offset error) . Was this connected to pin 14 of the AD8630 ? and did you add any means of adjustment? Regards, Louis

@KennethLavrsen

7 жыл бұрын

Hi again. If we go through the op amps one by one. U7A could benefit from a 220k pull down to negative but since I already have 4 mV offset from the DAC I had to compensate U7B with my additional circuit on the U7B inverting input. In my case the U7B is a different OPAMP supplied from 12 V because of the different FET. But same nulling circuit and load off circuit could be used in your original design. There is no point in pulling the output of U7B to negative because with any FET the gate will be operating far from zero. The issue is the millivolts offset on the inputs of U7B which accumulates from U5, U7A, and U7B. There is a need to have a way to compensate this. In my case the problem is 10 times bigger because my shunt is only 0.01. But even with 0.1 ohms a 4 mV offset you cannot set the load current below 40 mA which is annoying and cannot be compensated in software because you cannot send a negative voltage from the DAC. The places the 220k to negative will make wonders in the original design and also in mine is U7C and U7D. In both cases this enables these single supply opamps to go to true zero when current is 0 and/or load voltage is zero. You have a zero bias adjustment on the OPA277 that helps on the voltage measurement. The opamp I use (OPA334 because I did not have the OPA277 available) for voltage runs single supply (it is a max 5.5 V opamp so I could not run dual supply) and has no bias adjustment pins. But with the pull down to negative on the OPA334 and the extremely low input offset of the OPA334 I got an offset error far below 1 LSB on the ADC. So all in all in your original design I would try the 220k to neg 5 on U7C and U7D. This resistor does not eliminate input offset errors. It helps the opamp output stage when running single supply to pull all the way to true zero ground. I bypassed U7C and connected the OPA334 directly to the ADC. I can do that because the OPA334 is single supplied so I do not need to worry about negative voltage on the ADC input. In your design it is smart to use the last U7C at unity gain to make sure the ADC only gets positive voltage. But the AD8630 cannot pull its output all the way to ground without a little help. And since you have the -5V available the trick with the 220k to -5V will probably reduce any offset error you have.

Nice revisions Louis. You have put a lot of work into this project for which I for one thank you......................................Berni

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks Berni.

Excellent, Great update Best regards Great videos

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks Cesar.

Tip my hat to you. You read my mind, I was thinking on transient mode. Two things that can be added on transient mode are total time for running, so after a set amount of time the unit is switched off. The other thing is to add a SD card reader to store the voltage, time and amps while you are running in transient mode. In that way you may be able to record anomalies on the power supply under test that relate to changes on the load been applied. All your projects are excellent,

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks for your suggestions. I am sure there are many other features that could be added. Hopefully, it will encourage others to take my project and build on it. Regards, Louis

Finally the right man who uses the Arduino as it must be used! Not just blinking a stupid LED with Arduino or connecting fucking cheap Chinese modules to Arduino!!!! REALLY THANK YOU FROM MY DEEP HEART!

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks Tom.

Very Nice! you get a BIG THUMBS up from me, it is starting to look like a commercial grade Electronics load. One feature that I think would be very nice would be a logging facility with the ability to down load it to a PC and import the data to your favorite spread sheet application and make graphs. Thanks for sharing and keep up the great work!

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks Michael. Data logging may be an option for later. I am sure there a lots of improvements which could be made. Regards, Louis

@juliusvalentinas

7 жыл бұрын

Yes data logging would make this the best load, you just need to build :)

This is awesome. Thank you very much for sharing your knowledge with us. Your explanation is easy to understand structured in stages especially for people like me with more basic knowledge in electronics. For my next project I was going to build myself simple Bench power supply but now it will have to wait. I will be building this DC load project first. And who knows by the time i finish it maybe your next project will be DIY Bench PSU 😉 which can designed and tested with the help of this project.

@scullcomhobbyelectronics1702

7 жыл бұрын

Glad you found it helpful. A Power Supply project may be a idea for later.

@Hasitier

7 жыл бұрын

Bay Van I like the idea of a bench PSU project. Maybe also with keypad input along with rotary encoder. At the end it does not sink but source current. So it should be very similar in some points. Thanks for your great work. I really like your teaching style. I can easily follow as an electronic hobbyist.

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks Micheal for your comments. Regards, Louis

This is one of the nicest projects around. I wish the damn parts would get here. I think China uses turtles to deliver orders! I think I would go the "Menu" route instead of having to remember all these (unmarked) keys. For example, holding the rotary switch in for 3 seconds, will clear to zero and beep. these little code mods are easy to do on my own. Thanks again for a fantastic and a very useful test piece.

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks. Yes China can some times take a life time for items to arrive but the prices are much lower :) Adding more menu options could be a way forward. Regards, Louis

Well done. Two suggestions: 1) On the user setup function it would be nice if the default settings were displayed before an adjustment is made. 2) Have a quick reference sheet that can be taped to the top showing the combination of buttons to press to access each feature. "Like" clicked!

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks for your suggestions.

Thanks so much for this project! I have wanted to build a DC Load for some time and in fact had even previously acquired the enclosure you ultimately used with the intention of building my own design. Several of my amateur radio associates are building the same variant, so it's become a fun group project. While doing some testing this evening, I discovered a minor bug in the code. In the fan control routine, the LCD print statements need to come out of the IF - THEN sections. As it is now, the display stops updating at 50 C. The OVERTEMPERATURE still trips at 60 C, but the display continues to read 50 until it trips out. Thanks again! Chuck AI4WU

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks Chuck for your comments. I will have a look at the code for the fan routine which you pointed out. Good luck with your own project of your DC Load. Regards, Louis

Nice update. The mosfet arrangement is definitely a brilliant improvement! However, the software has gotten quite complex considering the button push combinations required for the various new modes. There should be an easier way to toggle between the different modes like using the rotary encoder initially to select the mode, then after that the encoder can be used to set the various values.

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks Sebastian. I agree the software is getting quite complex but this is a result of the way I have been adding the additional features while maintaining compatibility as the project progresses. Regards, Louis

@elchino835

4 жыл бұрын

@@scullcomhobbyelectronics1702 are the fets matched?. if they are not they are not going to dissipate the same power and the one with lowest gate threshold is going to die...and then the next ..and the next one and so on and so forth

good job master

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks.

Very good! Maybe the pulse shaper could be omitted to keep the trigger input able to receive analog signal? This could then steer the load according to the external analog value for some feedback loop purpose. Also instead of pushing two buttons simultaneously a short push / long push would seem to be more sensible to me. As for those mosfes, I wolud recommend to stick with at least 100V ones. Not only they can withstand some transient spikes, but higher voltage mosfets also tend to behave better in linear region. It is starting to be a very feature-rich project. Using faster converters and controller would render it into a semi-professional tool.

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks Medvidek for your comments. As you suggest there are a number of options for the trigger input. I will try an cover this in the final part of this project. The short push/long push may also be an option. The limiting factor with this project is the Arduino and as you say a faster processor would certainly improve its operation and accuracy. Regards, Louis

Thank you so much

@scullcomhobbyelectronics1702

6 жыл бұрын

Thanks.

Louis, Thanks for your recent comment (in "Part 4") about "I Set" not agreeing with the actual current. I've caught up with Part 5, and now have issues with the "Sense" voltage reading not with the voltage of my programmable power supply. (which is spot-on) I've tried adjusting the multiplication factor in the code calculating the "actual voltage" (50.4, I believe) and if I get the setting correct at, lets say, 10 volts, it goes off, below and above 10 V. In other words, non-linear. I wonder if you could discuss this calibration issue in the next video ? As always, great work and thank you for the learning experience!

@scullcomhobbyelectronics1702

7 жыл бұрын

Hi Phil, I will do some further work on calibration and see what I can do to improve things. Regards, Louis

Hi Louis, another option would be to replace the 12V fan and use a 5V instead. You can then use two 18650 Lithium Ion batteries in series with a small DC-DC converter to power the whole setup. I am guessing that it will draw 500mA-600mA so you could easily run it for several hours continuously.

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks Javier for the idea. The total current draw is only 70mA maximum when the fan is off. When the fan kicks in at 26°C the current draw goes up to a maximum of 220mA. So it would run a long lime on two 18650 Lithium Ion batteries. Regards, Louis

this is a amazing project! So much detailed information, love it! Just wondering if more mosfets in paralell would introduce new issues, let´s say 8 in paralell to handle about 450watts. A better route may be to use fewer and higher power rating mosfets?

99% of everything you say is over my head, but I'm still entirely fascinated.

@scullcomhobbyelectronics1702

6 жыл бұрын

;-)

Super!

It might be worthwhile considering changing over to a 4 by 5 switch matrix. This could add four buttons for just one I/O pin. Buttons to consider for the four are: Load On/Off, Zero readings, Trigger (via an opto-isolator or a relay), and one other. It may be worthwhile keeping the rotary encoder's button out of the matrix if you wish to continue using it as a 'shift' key although, personally, I'd suggest dropping that part of the User Interface. There's a typo on the schematic, "Sense" is written as "Sence".

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks Brain for your comments. I have considered a larger matrix keypad. I found one option on eBay which would give 20 keys as part of a front panel with a window for a LCD display. www.ebay.co.uk/itm/191693869222?ssPageName=STRK:MESINDXX:IT&_trksid=p3984.m1436.l2649 I may order one to see what is is like. Thanks for pointing out the typo error on the schematic. Regards, Louis

@Brian_Of_Melbourne

7 жыл бұрын

Interesting keypad. I reckon that there's enough of a window to see the 20 by 4 LCD through. Nice to have 'CLR' and 'ENT' buttons, they will go well with the current UI. A bit of a pity the rest don't, although '+/-' makes a good '.' replacement. I assume that you've seen this goo.gl/NJKbW8 on AliExpress.

@scullcomhobbyelectronics1702

7 жыл бұрын

Brian of Romsey Yes I have already ordered a few of these 5x4 keypads for another project I have in mind.

Hello Louis, I very much like the idea of transient mode, however as others have mentioned, the keypad combinations are becoming less intuitive. Perhaps the letter keys can be mapped to menu lists, with sub menus as the numbers (ie: "A" key displays menu "A", where you can select "1 = CC", "2 = CP" etc...). I would also like to suggest adding a BNC external input, allowing the user to switch between "internal" DAC driven load, or an "external" signal from a waveform generator to give the user unlimited transient waveform options. Obviously some input protection circuitry would be required to clamp the voltage on the op amp current sense input, but not a huge mod to the unit.Keep up the good work... Brett

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks Brett for the suggestions. Adding menus may be an option. With regards Transient Mode I was thinking of adding a List Mode which would allow the input of a range of current limits. As this project has progressed it has take on a life of its own and sometimes it is difficult to know when to stop. Hopefully, it will encourage others like yourself to add to my project to fit your own needs. Regards, Louis

Good work. (y) One tip: 31:28 Continuous/Toggle/Pulse, title show only "Tansient Mode" , add to software title like "Tansient Mode (TC)". That each sheet being would be see mode (CC/TC/CP/CR/BC).

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks. That could be simply added in software if you feel it needs it. Regards, Louis

Never use wire wound resistors. You are getting and inductor and resistor in one package causing an over damped ringing during high to low and low to high current transitions.

Hi Louis! Thanks for your professional videos! How would you rank the CSD18502KCS FET ?

@scullcomhobbyelectronics1702

7 жыл бұрын

Hi Phil, I have had a look at the datasheet. I do not think this is a Logic Level Mosfet. It needs a higher gate/source drive voltage. Also the maximum drain to source voltage is 40v. The Junction-to-Case Thermal Resistance of 0.6°C/W is also higher than the one I used so it will get hotter. At 25°C the maximum current at 30v would be 2 amp but this will reduce by at least 50% when operating as a DC load. Therefore, I would not use this Mosfet in this project. Regards, Louis

Wow,very impressive. Do you take suggestions? Maybe revisit your voltmeter and milliohm meters, combine them and add additional functionality to make a bench multimeter? That is something I'd love to make for myself.

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks for your comments. Nice idea but it may need a re-think on the design first. Regards, Louis

+Scullcom Hobby Electronics Since there's no pvt msgs so I'll post my topic here and so all can learn or maybe help! I'm having a problem with a precision rectifier circuit and it gave me a real headache. I had started for 3 months with a small energy meter project (for 230VAC). Everything went well except the AC to DC converter part! I tried a lot of circuits (which the Internet is full with) and got luck with some circuits (one of them is linked below), but, the problem is that the circuit isn't linear ( I'm not sure if I'm describing it well). In other words, I couldn't find a relationship between, VAC-in, VDC-out or VPP! I searched a lot, but as I said got no luck at all. My goal is to convert that ACV ( which between 15 - 550 mV) to some DCV so the micro-controller can read it. Hope I described my problem well and hope YOU HAVE TIME FOR SOME KIDS' QUESTIONS lol Since you are an analog man more than a digital so I guess you can help me and everybody reads this comment also is welcome. Every help or thought is very appreciated. ================================== Here's the link of my schematic sound.whsites.net/appnotes/an001-f8.gif ================================== BTW, I'm using ADC (ADS1115) which is 16-bit ADC IC so I don't care about amplification and I'm not using op-amps for that purpose. So, if there's another way to convert AC to DC precisely I'll be soooo thankful!

@scullcomhobbyelectronics1702

7 жыл бұрын

Hi Tom, One option may be to use a precision RMS-to-DC converters IC such as the AD8436 or the LTC1966. Have a look at their datasheets. This may solve your problem. Regards, Louis

@tommyallehamn9543

7 жыл бұрын

Hi Louis, Thank you for your reply! I checked both datasheets and I think the LTC1966 has easier applications circuit and more accrued. THANKS AGAIN :) Regards, Tom

Hello In TT mode, on first trigger, it looked like it went to 2A first and then right away to 0.5A. Is that so or not? for the next time, this transient modes, can you take out scope and show how current goes, what is the rise/fall time, overshoot if any?

@scullcomhobbyelectronics1702

7 жыл бұрын

In the demo of this feature I simply used a push button to ground to simulate a trigger pulse and what you saw was due to the key bounce on the push button. As I said it really needs a pulse shaping/reduction circuit on the trigger input and I plan to add one during finalizing this project. We can have a look with a scope next time but you really need the trigger input circuit there first. Regards, Louis

Do you have any intention on respinning the PCB now that the components have been changed so much? Good work. Once things are finalized, this will be a project that I plan on building. It looks pretty capable.

@scullcomhobbyelectronics1702

7 жыл бұрын

Hi Jonathan, Once the project is finalised I may design a new PCB and make the artwork available as a download. Regards, Louis

Nice improvements. Did you have a chance to look at the latest PCB? I was waiting for this update to then have a go at PCB for the heat sink area.

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks Rob. I have been busy of late but hope to get back the your PCB layout soon. Regards, Louis

@robbruno8246

7 жыл бұрын

Thanks. No problem.

Instead of : if (BatteryLife lcd.print("000"); } if (BatteryLife >= 10 && BatteryLife = 100 && BatteryLife

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks for your suggestion. Coding is not one of my strongest skill sets. Regards, Louis

@electronicsmokeUK

7 жыл бұрын

You are welcome, and thanks for the great videos.

Louis, The maximum current that I can draw is 4.1 amps. I checked the output at the divide by 10 resistors and the voltage corresponds to .41 V. If I step the current any higher, the load cuts off, if I back the current down, the load comes on. The limits are set high enough. Any ideas ?

@scullcomhobbyelectronics1702

7 жыл бұрын

Hi Phil, The reason the maximum current limit is 4.1 Amp is due to the fact that I added a 4.096V reference to the DAC to improve accuracy. If you want to go higher than that then you could go back to the original circuit design but would loose some of the benefits of the change. In future we could look at other ways of reaching higher currents. Regards, Louis

It appears that the diff-amp and the negative supply are missing from the Schematic, v5.

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks. I am working on an update for this project and will redraw the whole schematic.

I was thinking why not when the unit powers up with no load have it automatically zero the amp and watt instead of it being manual ?.. Just a thought.

@scullcomhobbyelectronics1702

7 жыл бұрын

In one of my software versions I did make this adjustment automatic at switch-on but then decided to allow it as an option with a key press. One of the reasons was that if a load is connected at switch-on then it would zero the reading to that voltage making it read zero instead of the load voltage. If you prefer an automatic zero as an option it is very easy to do with small changes to the software.

One more modification. In several places I run into trouble with the display because I run beyond 99.99W. and 9.999 A. So in several places I had to put in modifications to the display of Amps and Watts so that when I have more than 10A I reduce the decimals to 2 and when I reach more than 100W I reduce the decimals to 1. This way the display stays on the second line.

@scullcomhobbyelectronics1702

7 жыл бұрын

Hi Kenneth, You will notice in my code I automatically switched from 3 decimal places to 2 decimal places if the voltage reading reaches 10V or above for the same reason you found with the watts and amps (LCD display movement). As I was limiting the current and power in my project that was not an issue for those readings. I may make the same change to the watts and amps in my code as I did for volts so that the code is future proof for higher power ratings which could be added later if the project is upgraded. Regards, Louis

@KennethLavrsen

7 жыл бұрын

Hi. I have since this comment continued testing and found another 10+ places where I had to adjust things for the higher current and power that I build for. I also found some bugs in the constant resistance code. I will make my code available on my website in 1-2 days. I just want to walk through all features before I do so. I still have not tested any transient or battery discharge features at all.

Is there a PCB layout for v5?

@scullcomhobbyelectronics1702

7 жыл бұрын

Hi, I may produce a new PCB artwork later when the project is complete. However, one of my followers Rob Bruno has already designed a new PCB using my design but we have not tested it yet. It should be OK and uses surface mount components. If you are interested it is available from: www.oshpark.com/shared_projects/tXUxCSVa You can also download the circuit from the link below: www.dropbox.com/s/orwivgbt23gl158/ElectronicDCLoad1.pdf?dl=0 Regards, Louis

Hi i am using this SD6109 ic in my smps project how can i make a variable duty cycle in this chip is there any idea Thanks

@scullcomhobbyelectronics1702

7 жыл бұрын

Hi, At first glance it does not look as if you can vary the duty cycle of the PWM in the chip. There is a pin (pin 16) which is the "error amplifier output and the input of the PWM comparator" you could try playing with this pin. Have a look at this pin on an oscilloscope to see what the signal is like. You may be able to apply a variable voltage or external signal on this pin, which may alter the duty cycle but I am not familiar with this chip so this is just a guess.

@youawa7dsa133

7 жыл бұрын

Thanks

22:05 "All the options we had before..." Nope! Lead-acid isn't there any more!

Noooo! By mounting the MOSFets so close together you undo the self-balancing by the positive TC! On the other hand, the Drain resistors are really necessary an you could estimate the correct value using the expected maximum dissipated power, the expected max junction temperature and the characteristics. All multi-transistor electronic load manufacturers do it like that!

@freda5344

6 жыл бұрын

um, by spacing them further apart, the positive TC effect will allow each to operate differently depending of each individual parameter characteristic by that tiny amount, and and the temperature gradient of the heatsink effect. The closer the mosfet cases are to each other, the tighter the temp difference will be between them - my opinion is that this IS superior....but but yes its harder to wire up!sigh! Re drain resistors, the required calculated low value power resistors, they are freaking expensive! better to just use a safer area of dissipation of the mosfets. Something you just can't do with BJT types.

@fabimre

6 жыл бұрын

Fred Z, I think you're wrong. The whole idea is that the positive TC of the MOSFet limits its current compared to the others when it gets HOTTER then the others. So putting them closer makes their temperature more equal, therefore negating the self-balancing of the circuit. Best is spacing the apart, so each feels it's own teperature more than the total average and adjusts it's internal resistance accordingly by it's negative TC. P. S. Drain resistors are cheaper than MOSFETs!

Hi Louis! Thank you for this nice project, I haven't seen anything more functional published so far.This will become my next project after I finish my function generator from Qex magazine for my little lab www.arrl.org/files/file/QEX_Next_Issue/May-Jun_2013/Fernandes_QEX_5_13.pdf. I just desperately need more power dissipation, so I am thinking of adapting the project for higher power and use IXTQ170N10P mosfets. Do you see any problems with that?

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks Vytautas, Interesting function generator project. The IXTQ170N10P is a higher power Mosfet but I think it may need higher gate/source drive voltages. The Junction-to-Case Thermal Resistance of 0.21°C/W is very good so it should run cooler. I have not tested this one on my project as yet. It may need some changes to the gate/source drive voltage. But worth a try. Regards, Louis

@vytautassl8379

7 жыл бұрын

Thank you. I will certainly give it a shot. By the way, a method that I use for fixing output power transistors in my audio projects, for example, is bolting them down to the heatsink through one 3 mm aluminum plate which I put on top of all devices. This way, all transistors are pressed to the heatsink very uniformly, and the plate serves as an additional heat transfer/distribution path. This is a method I had learned from Elliott Sound webpage sound.whsites.net/projects-0.htm#msc

I decided to make a quick and dirty wiki page of my version of this project. I put a photo of the front and some quick snaps of handwritten schematics that shows what I changed. They are not complete. You have to take the original schematic. I did not change display, keyboard, rotary, Arduino Nano etc. Decoupling is also all the same. I also put a version of the modified code. I changed quite a lot already and I went through all the code and fixed the indentation to make the code easier to read. Here is the link. www.lavrsen.dk/foswiki/bin/view/Kenneth/ElectronicLoad

@KennethLavrsen

7 жыл бұрын

Update. I bought a much more powerful 60x60 cm fan. It is a PABST fan with double the capacity of the low noise fan I had used initially. The result is impressive. At a heat sink temperature of 82-84 deg C I can now load 230 W permanently. Before it could hold 140 W. My initial goal was 200 W. The downside is noise. The new fan sounds like a Harrier fighter plane is taking off in my lab. So I had to refine the fan control. The fan starts and push lots of air already at a PWM setting of 20! And enought to use the load at 4 A / 12 V with very little noise. When I want to test a big power supply I can live with the noise. While doing this I found a new problem with the original design. I could not make the PWM of the fan work at all. It turns out that an Arduino Nano cannot PWM output D2. D3 is an interrupt pin for the rotary. D3 can also work as interrupt. So it was a simple matter of swapping the rotary interrupt and the fan control in the design. The new version of code and corrected schematic is on the same link I gave above. www.lavrsen.dk/foswiki/bin/view/Kenneth/ElectronicLoad

@KennethLavrsen

7 жыл бұрын

I wonder if my this thread is marked as spam KZread because of the URLs. I can only see them when I am logged in with my private Google account. I am at work now and wanted to see if i needed to reply to any questions and I could not see my updates so I assume noone can except Louis and myself. My links are just pointing to a wikipage with 3 photos and my updated code but I guess the KZread algorithms cannot see what is spam and what is ham.

@scullcomhobbyelectronics1702

7 жыл бұрын

Kenneth, For some reason Google put your message in the spam folder. Just found it. Regards, Louis

@drankenkorps

7 жыл бұрын

I and we would like to build your version of load..I try to find your private email but didnt. We will make in kikad pcb and share .. Your version have some different components.. A little cooperation will be super..if you wish. jeanlucdemarc@gmail.com

@KennethLavrsen

7 жыл бұрын

Jean Luc De Marc, I am still improving the load and still the schematic is handwritten. I have not added the latest update to my wiki page yet but I will in a few days The last improvements have been - I replaced the 12-bit D/A converter by a 16-bit. Still I2C based and very easy to fit into the software. So now I can set the current in mA steps even though I have max 20A - I have been playing with the fan control. It is difficult to PWM voltage control a fan. They tend to suddenly speed up in one step. I have ordered a Sanyo Denki fan which is a 12 V 4-wire fan controlled by a 25-kHz PWM signal. I am still waiting for it to arrive. I have altered code that runs the PWM at the needed 25 kHz. For now I have added a buffer with two BJTs to drive a power FET. Better but I still cannot get the fan I have now to run medium speeds. But it cools very well at low and I can run my 250 W at high - I have done a lot of changes in the code to get more accuracy. There was a lot of float to interger convertion that worried me so I tried to do integer math when possible - I found the code slow to respond to key presses and the reason is that the A/D converter is pretty slow. It takes 80% of the loop time. I want to try a different A/D which on the paper should be able to run faster. I want to see if that is at a cost of any precision. With the existing A/D, the time is actually spent on two reads: Current and voltage. I won a lot of usability simply by letting the loop read voltage and current every second time. This way the loop services the buttons twice as fast. That is actually quite OK. - Last, the rotary encoder code needed a major rewrite. The code Louis uses comes from another KZreadr and it has the problem that if the rotary is stopping between two positions (happens easily), the code halts because the interrupts are firering all the time (LOW). I rewrote the code so the interrupt happens edge triggered. I did this in combination with a proper hardware debouncing. I will soon update my wiki page. My email is kenneth at lavrsen dot dk

Lost your reasoning for looking at Rdson values. The MOSFETS are operating in the active region, not in the ON region. Consider revising.

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks for your comment. If you look at the transfer characteristics of this particular Power Mosfet you will see that at just over 3 volts VGS it crosses over and a such RDSon will have an effect. For most of the time when using a Power Mosfet in a DC Load you are operating in the saturation region and the amount of current is dependent on VGS. For different levels of VGS and junction temperature the Mosfet RDSon will change slightly. As more current flows through the Mosfet the temperature rises.