Scullcom Hobby Electronics #46 - Electronic DC Load Part 2
Вставка
- Опубліковано 6 жов 2024
- In part 2 of this project we make a number of improvement and additions. The project now includes a battery capacity discharge function for LiPo batteries. Also the display layout has been upgraded. The software has also been updated and a new Arduino library has been added. Below are links to download the latest software (version 2.0), new library for the RTC timer function (zip file), RTC small PCB artwork and circuit (zip file), detailed parts list and an updated circuit diagram with notes for the main PCB:
www.scullcom.co...
www.scullcom.co...
www.scullcom.co...
www.scullcom.co...
www.scullcom.co...
An engineer like yourself should have a much bigger audience.The in depth electrical and thermal design explanation is amazing.
keep up the great work i am learning a lot from your videos !!!
Thanks for your kind comments.
You could use a STM32F103C8t6 Arduino, also called Blue pill, instead of the Arduino NANO. Because -
1) It costs the same as Arduino Nano!
2) It has built in RTC
3) It has built in 12 bit ADC
4) it was a very fast, 72MHz frequency and 16 bit PWM timer, so you could use the PWM at high frequency and then convert it to DC voltage using Low Pass Filter which will act just like an DAC, but with higher resolution.
5) It probably has a big EEPROM inside.
I don't know if it's the right part to mention this.
Thanks for the whole series.
Thanks. I have been looking at the STM32F103C8t6 and may use them later. I ordered a few from eBay and will have a play with them. Regards, Louis
Great project, Louis. Very well diagrammed and explained, as always. Well done!
One useful feature in electronic loads that you might care to add is the facility for remote sensing of input voltage. Particularly for battery capacity testing, it's important to sample voltage *at the cell* rather than at the end of a test lead carrying significant current. Calculated cell capacity can have substantial error if the voltage drop in the test leads is not accounted for. A small SPDT switch on the front panel could select between internal or remote sense.
There have been a couple of comments about whether MOSFETs can be used in the linear mode and about selecting the right MOSFET based on Rds(on). Your discussion of safe operating area (SOA) hinted at the answers, but for viewers who are new to linear regulator design it should be stated plainly that Rds(on) is *not* relevant to MOSFET use in a linear application. Rds(on) is an expression of the device's resistance *only* at a *specific* gate voltage (usually turned fully on).
In a linear application, the gate drive circuit will purposely throttle the MOSFET, creating whatever drain-source resistance is required to achieve the desired output. Because of the way many MOSFETs are made (multiple parallel transistor "cells" on a single die), they're subject to hot spots that can lead to thermal runaway. As you discussed in the video, this is why you must pay close attention to the forward bias SOA and design your circuit to stay *well* within its boundaries.
This is one niche application where Rds(on) is irrelevant and it makes choosing the right MOSFET difficult, as you can't do a parametric search on SOA. One way to find what you need is to sort by max power dissipation (Pd) and then slog through the data sheet SOA diagrams for parts in the Pd range that suits your application.
For deeper explanations of this subject, viewers can read the following application note by Fairchild Semiconductor and article from Power Electronics Technology:
Fairchild AN-4161: preview.tinyurl.com/hepm8fz
PET Article: preview.tinyurl.com/z7s6u4q
[The links above will show a preview of the full URLs at TinyURL.com, where you can choose whether or not to click through.]
For some reason Google put you comment in the "Likely Spam" box, maybe due to the web page links you added so I had to allow it manually, hence the delay in seeing it in the comments.
Thanks for your details comments which I am sure will be of interest to the viewers. I will have a look at the remote sense issue you raised. Good point about focusing to much on the Mosfet ON resistance - thats why I decided to cover the Safe Operating Area in my video. Thanks for the links which will be helpful to others watching. Regards, Louis
Hello, I just came across your blog. I wish I would have found it sooner. I am in my upper 50s, electrical engineer by field and am getting back into electronics again. Building, testing and creating. Thank you for the great explanations and the incredible content of your videos. Please, keep up the great work!
You are inspiring me. I build a variable resistor project with optocouplers and resistors and now I want to switch to mosfet ohmic region!
thanks for the very detailed tutorial. very few channels build electronics projects like this.. inspired
Thanks
Thank you for your effort. Exellent as allways. Big thanks and thumbs up.
Thanks for the thumbs up.
Good job with the updates! I really like the way you added the RTC chip. I2C is great for that kind of mods: Just two pins for power and two for data and one can add "unlimited" amount of peripherals without redesigning the main board.
Thanks for your comments. I have some more ideas for further enhancements using the I2C bus. Hope to show them in Part 3.
RTC - good idea.
Last time I was using internal clock from arduino for capacity measurements I had very unprecise results...
Thanks Arek. I have found the external RTC option always to be the best. Regards, Louis
Best explanation on dc loads. Thank you for uploading these videos!
Amazing and brilliant work. Sharing your knowledge willingly shoes that you are a humble and well educated person. God bless you.
Thank you soo much for the superb and in depth explanation
Cheers!!!
Thank you.
Maye a footprint for another parallel mosfet(s) and the their current-ballancing source resistor(s) could be added to the PCB design to handle higher currents. Also the power tracks could be done somewhat beefier as there is a lot of unused space on the PCB, but not a big deal for
Thanks for your comments. I am looking at adding additional Mosfet's in parallel. We could even take the Mosfet and sense resistor off the PCB and use heavy gauge wire.
I would also be interested in increasing the current capacity by running parallel Mosfets, and believe taking them and the sense resistors off the PCB would be the way to go. I look forward to your future videos, great information regarding the engineering design process.
Thanks Dale.
While finishing your millivoltmeter I am following this project with great interest.
Despite having several Chinese electronic loads I want to build this one because seems promising.
Great job as usual and thanks for your hard effort to bring another brilliant design.
Thanks for your comments.
Excellent build and the use of software and hardward.
Thank you.
Concerning the BC function you may, if not already mentioned here, set a CV discharge when you are getting closer to 3V. When reducing the discharge current, the voltage will rise a bit. You could set a cut off current then, e.g. 10mA or 5mA. This will give a more accurate mAh measurment.
The BC function sets a constant current which is maintained as the battery voltage reduces.
Thank you very much for all your videos, they are just brilliant!
Thank you for sharing your professional knowledge with everyone, not much professionals actually do that.
Thanks for your comments.
Always nice to know that you've uploaded another video. I might give this one a go! As I've said before, I think your videos are a tremendous resource and I know that you are making a positive impact on up-and-coming engineers as well as seasoned veterans alike. Keep them coming! :-)
Thanks.
Very nice video again.
Would be nice to include an external voltage reference to increase te accuracy of the readings.
Thanks for your comment. In Part 3 of the project I will be looking a calibration. I have already been testing a 4.096V reference IC to improve accuracy of voltage steps of the DAC (i.e. 1mV steps).
Hi Louis, the MCP4725 takes the Vdd as the reference voltage for the DAC, so wouldn't it be better to take 5.000V (AD586) as a reference voltage instead of 4.096V. Or is there something I overlooked?
You and Phil sound like you are from the same city. Looking forward to part 3.
I do not know where Phil is. Working on Part 3 at the moment.
great video. But i think the transistor is not appropriate for anything above 25 watts according to my experiment with these sort of mosfets. The mosfet might die very soon. May be it is a good idea to use multiple mosfets in parallel. Anyways great project. Keep it up
It would be nice to be modular and add on additional MOSFET nodes to allow higher currents over all. I'd like to go beyond 3A for some things.
Thanks for your comments, yes it is a good idea as you say to add multiple Mosfet's. So far I have tested it up to 50 watts with a suitable heat sink and it seems to be OK. As we progress with this project it is my plan to add additional Mosfet's in parallel to increase the power capability. Regards, Louis
Excellent video series. You say that the LM35 still requires an isolating sil pad. Why is that? I would have thought that the pad would give you incorrect temperature readings (even if only marginal) since it is more of a thermal insulator than the mica pad or just thermal paste alone. Or, if the reason is electrical isolation, isn't that defeated by the metal screw holding it against the heatsink. Finally, I thought you were not supposed to use heatsink compound with silicon pads, and that the mechanical pressure alone would maintain adequate thermal contact. Thanks for the clarification!
Hi Adam,
The GND pin on the LM35 is also connected to the metal body of the LM35. Initially on this project I bolted the Power Mosfets to the heatsink without any isolating pads so the Heatsink was at the drain voltage which could be up to 30V. I did this to improve the heat transfer from the Power Mosfets and so I had to isolate the body of the LM35 from the heatsink. On later versions of this project I used Mica isolating pads on the power mosfets and so you could then remove the isolating pad on the LM35 if you wish.
With regards using heatsink compound with silicon pads. Normally you can use silicon pads without any heatsink compound. However, silicon pads are only normally intended for single use, but I sometimes do clean them and reuse them, and in that case I add the heatsink compound as well to ensure good heat transfer over the whole surface.
Regards,
Louis
Great project! I just didnt understand why would there be a problem if a higher voltage was applied to the mosfet and it was connected to the heatsink only with thermal paste and the heatsink was isolated of everything else in the circuit.
Great video as always. I appreciate the well explained code, design considerations...and you giving your time of course. Thanks
Thanks. The work on the code will continue. I plan to make a number of other changes and improvements, after testing my ideas I will cover them in a future video.
thanks for your contribution of the rtc timer library! appreciate it!
great project!
Thanks. I plan to make some additional libraries available as we progress.
Thank you. The capacity meter is most helpful. Not sure how you only have sub 11,000 subscribers as you should have loads more as your channel is very informative. All the best for subscriber growth in 2017.
Thanks George. My subscribers seem to grow slowly probably due to the fact that I am not very proactive in promoting my UA-cam channel on other social media platforms. Thanks for your best wishes of subscriber growth in 2017. Regards, Louis
Awesome project! Just what i was looking for about a month atm
Hi Louis!
My transparencies are not dark enough. Could you tell us what Lazer
printer and materials that you are using? Your transparency traces look
real black, I've been stacking two transparency's to obtain the
necessary blackness, which is a real pain.
Thanks for your update, very professional as usual!
Hi Phil,
I do not use a Laser Printer. I use JetStar Inkjet Film for Epson Printers which are made by Mega Electronics Limited. I purchase them from Rapid Electronics (their order code number is 39-0780) - direct link below:
www.rapidonline.com/mega-electronics-100-070-jetstar-standard-ver-ii-a4-pack-of-10-39-0780
They come in packs of 10 A4 sheets for £9.88.
I find they print darker if you set your inkjet printer setting for Premium Gloss Paper or High Gloss Paper.
If you still have problems message me in Private and I will see if I can help. Regards, Louis
Great project. May i ask if you with ever make a digital controlled power supply?
Thanks for your comment. A digital controlled power supply may be an interesting project for the future. I will have a look at that after I have finished some of my other projects.
It would be great if it can connect to serial port of the computer (USB-> TTL).
Agree!!! It would be an awesome project.
Thanks for the very good explanation of thermal resistance. High quality video as always.
Thanks Markus.
gee... wished i seen 🤓 these cool videos during the development era for up to date feedback🙄 anyway, thanks a lot and the best 🤩🍹
Thank you sir for this project
Especially battery capacity measurement function is awesome
Thanks. Your welcome.
Thank you for such a nice project! You are always thinking ahead and I love that also!
Thank you :)
There is a possible improvement in the circuit. If there is no zener diode in parallel with the current sense resistor, it might fry the opamp when the mosfet dies to create a short circuit between drain and source in which case almost all of the input voltage will appear across the sense resistor and it might damage the opamp. Therefore add a zener diode in parallel with the current sense resistor.
Thanks for your comments. Yes we could add some protection in the event of the Mosfet going short circuit. This is something I will have a look at. Using a zener diode may not be the best way, since if you have a high voltage, high current source this could also damage the zener diode which in turn could resulting in still damaging the OP Amp.
Usually, the inputs have ESD protection diodes to both VCC and GND. If you limit the current through it to 5mA, you are fine.
For that, R16 needs to be increased. If not, a schottky diode to ground and VCC after R16 (U7D/pin-12) should do the job.
Great project, looking forward to part three.
Thanks. Hope to complete Part 3 soon.
Hello, I really liked your project. I have the following question, can it be done for a range up to 100v and 20A to work, the idea is to discharge battery packs up to 24s.
I like this project, and I'll be following it with interest. Your change of MOSFET makes a lot of sense, heat sinking is always an issue in these designs. Perhaps you could consider parallel devices for increasing the power? It increase the cost and complexity, so perhaps only a consideration for people wishing to take if further? I've used the IRFP260 MOSFET in some of my projects that required constant power loads. The gate drive requirement is higher, but it's very stable over temperature and it's easy to use them in parallel.
Keep up the great work! Looking forward to the next video.
Thanks for your comments. I have been testing a number of Power Mosfets. So far I have found the BUK956R1-100E the best one to date. I am looking at putting two in parallel. I am also looking at the IRL640 which is capable of higher drain/source voltages up to 200V, but have not tested it yet (have a few on order).
I think the BUK956R1-100E is a very good choice given the design requirements you've chosen. I like the IRFP260 because it has an isolated mounting hole. It removes the electrical isolation issues associated with direct heatsinking, plus the low RDSon is not bad either! Keep trying new devices as you never know what you'll turn up! Thanks for your reply to my comments, I know it must be difficult to answer everybody who writes to you.
The isolated mounting hole of the IRFP260 does make it easier to isolate it from the heat sink and make good contact. I have been test a wide range of Power Mosfets including the one you mentioned (IRFP260) but to date I still keep coming back to the BUK956R1-100E. Many of the others are not true logic level Mosfets and do not work very well with the restricted gate/source voltage.
The BUK956R1-100E is a nice device, but it's a fairly unique part. It's already been noted by other viewers that the part is now marked as End of Life on the NXP website. www.nxp.com/products/discretes-and-logic/mosfets/automotive-mosfets/n-channel-trenchmos-logic-level-fet:BUK956R1-100E?tab=Buy_Parametric_Tab
I did find a few others with fairly dog linear characteristics but the Vgs range is a bit too high for the design in it's present form. One such device was the Fairchild FDP075N15A_F102 www.fairchildsemi.com/datasheets/FD/FDP075N15A.pdf Perhaps you might consider modifying the design use a wider Vgs range it's would increase your possible choices of MOSFET.
Despite the chosen MOSFETs EOL status it still seems to be available from a few sources.
I think the Lipo testing is a good feature. Would it be possible to combine features from your waveform generator software and introduce a limited form of pulse load testing?
Thanks for your detailed comments. I am currently working on a number of improvements and additional features for Part 3. I have consider a Pulse Load option and may decide to add that later. I hope to get Part 3 of the project finished shortly which has involved addition work on the software. Regards, Louis
Thanks again for your big work to make the project running 👍😀
Thanks, enjoy.
Thank You for your effort making all this.
Amazing as always, thank you so much for sharing a bit of your knowledge.
Thanks Bruno.
Thanks for your informative and challenging projects
Thanks.
Great video series. I'm curious about what Lithium battery you used for the BC test? A typical Li-ion 18650 battery today has 2,000maH (or higher) rating from 100% charge down to 20% discharge level. But your test only showed 51maH.
Thank you for your projects and keep up the good work !
Thank you.
Great video. Where can one find out the thermal resistance values for silicon pad and mic etc. ?
In case someone want to use Precision RTC Real Time Clock Module (ZS-042) with chip DS3231M will face the address collision in this DC-Load project.
I2C address 0x68 is assign to the DS3231 and cannot be changed. Even if the RTC board (ZS-042) contain 3 pair for solder joints for address selection unfortunately this change address only for EEPROM memory. In project DC-Load is used the A/D MCP3426A0-E/SN as dual 16-Bit Sigma-Delta ADCs which occupy same I2C address 0x68. Solution is to use HW I2C protocol for all slave devices as it is, exclude DS3231 which is connected to:
SDA to pin A2, This pin has a 10K pullup resistor to Vin
SCL to pin A1, This pin has a 10K pullup resistor to Vin
This two pins are used in function as "Software I2C". Thanks to Mr. Henning Karlsen which developed library: DS3231. Then you to slightly change the code:
- text which have to be replaced or commented.
+ add a line into DC-Load code
-//#include //Scullcom Hobby Electronics library www.scullcom.com/MCP79410Timer-master.zip
-//#include "RTClib.h" //DS3231M timer declaration
+#include
+DS3231 rtc(A1, A2); // Init the DS3231 using the software interface
+unsigned long timestart; //unix time when the function timestart() was executed.
+byte timerstatus = 2; //1 - running, 2 - stopped
+unsigned long elapsedtime; //current unix time
+unsigned long timedelta; // number of seconds the timer is running
+unsigned long timestopped; // unix time where the time was stopped.
+unsigned long stopSeconds = 0; // number of seconds where the time was stopped.
+byte tseconds = 0, tminutes = 0, thours = 0;
// after void setup() {
+rtc.begin(); // Initialize the rtc object
// after void loop() { // needs to be replaced!
-timer.reset();
+timerreset ();
-timer.getTime()
+timergetTime()
-timer.getTotalSeconds()
+timer_getTotalSeconds ()
-timer.stop();
+timerstop ();
-timer.stop();
+timerstop ();
-timer.start();
+timerstart();
// add to end of program
//------------------------ DS3231 Timer implementation --------------------------------------
void timerstart() {
if (timerstatus == 2) {
timerstatus = 1;
stopSeconds = rtc.getUnixTime(rtc.getTime()) - timestopped + stopSeconds;
}}
unsigned int timer_getTotalSeconds () {
if (timerstatus == 1) {
elapsedtime = rtc.getUnixTime(rtc.getTime());
timedelta = elapsedtime - timestart - stopSeconds;
return timedelta;
} else {
// parameter already in timedelta
return timedelta;
}
}
String timergetTime() {
unsigned long tmp = timedelta;
tseconds = tmp % 60;
tmp = (tmp - tseconds)/60;
tminutes = tmp % 60;
tmp = (tmp - tminutes)/60;
thours = tmp;
return align (thours) + ":" + align (tminutes) + ":" + align (tseconds);
}
String align (byte number) {
if (number
Great job!!
Thank you for sharing
Thank you.Big thanks and thumbs up.
great! i think it would be useful if we can tweak cut-off voltage every time with the encoder (not only in software). So menu with couple of batteries types and one user-defined profile would be nice :) thanks!
Thanks. I am looking at possibly a menu for different battery types. Hope to cover this in Part 3.
Very nicely done project! Could this be used to test the battery capacity of other batteries such as a 12V lead-acid battery or gel battery of 17Ah capacity?
Thanks. Yes it could we just need to add the addition requirements for different battery type. I am looking at adding some options in Part 3 of this project.
Thumbs up for sure. Lots to think about.
Thanks enjoy.
You are a great teacher! I have built part 1 and will definitely try this. Have you done any work with IGBTs? I saw it being used in a analogue DC load to handle very high currents.
Thanks. I have not done any projects with IGBT's yet but may look at them in the future. My understanding is that they are a cross between a Bipolar Junction Transistor and MOSFET transistors. They have the output switching and conduction characteristics of a bipolar transistor, but are voltage-controlled devices just like a MOSFET. However, the IGBT does not have a body-drain diode which MOSFET's have, also they have a negative temperature coefficient, which could lead to thermal runaway. Having said that it would be interesting to have a further look at IGBT's and see if they are suitable for future projects. Thanks for reminding me of them. Regards, Louis
The only place I've seen them used, is to drive ignition coils. Provides insulation from high voltage. Maybe they wanted to be insulated from the load and required very high operating voltage. All else, inferior to MOSFET.
Excellent project and very good exlanations. I just found a little mistake in the schematics, You mention there that R6 and R16 should be .1% ones. Guess you mean R6 and R14. Right?
Yes your correct.
Excellent project !!! Thanks a Lot
Thanks.
Thank You very Much fpor your advices i'm going to realize this project.
Thanks, enjoy your own build of this project.
This is gold
Why didn't you use a rotary encoder instead of all these P/B switches?
Fantastic project.. Karl
Thanks Karl.
Excellent video! Thanks!
Thanks.
the thermal resistance calculation confuses me. by that it would seem that it would be best to have no heatsink at all eliminating the thermal resistance. doesnt the thermal dissipation of the heatsink have to be factored in as well to understand the temperature?
First remember thermal resistance is the reciprocal of thermal conductance. So therefore thermal resistance is actually a heat property which is a measuring a temperature difference by which an object or material resists heat flow. In my example I show that effect for each element between the mosfet die and ambient temperature and how that effects heat flow. The important thing here is to remember that thermal resistance measured in °C is the amount by which it restricts the heat flow or heat transfer from the mosfet die. So it is important to keep the total thermal resistance as low as possible.
In my example I showed the effect of running the Mosfet at 30 volt at 3 amp which means the total power being dissipated is 90 watt. The thermal resistance illustrated is therefore calculated at 90 watt.
You will see at that power level the heatsinks resistance to heat flow is 54°C, this is calculated from the datasheet of the heatsink which shows its resistance to heat flow at 0.6°C per watt. Therefore 90 watts at 0.6°C per watt = 54°C.
If you did not use a heatsink as you suggest then the thermal resistance of the mosfet junction die to ambient temperature would be 60°C instead of 0.43°C (look at the datasheet for the BUK965R1-100E (table 6, thermal characteristics). So this would have to be factored in to the calculation when you take out the heatsink thermal resistance and would have a drastic effect on the ability of the mosfet die to keep with in its maximum allowed temperature.
Hope this helps to understand. I appreciate this can be a difficult subject to understand and id not easy to explain in just a few words.
Regards,
Louis
A great project, Louis. Do you or any of your viewers know of a similar project that it capable of 130V at 1A (up to 140V preferred)? I assume there would be too much to change in SW and HW to adapt your project?
Hi Bob, As 130V at 1A is 130W it should be possible. You would need to use a higher voltage drain/source rated logic level Mosfet and use say two or more wired in parallel to share the power load. The BUK956R1-100E I am currently using is limited to an absolute maximum drain/source voltage of 100V.
The Vishay IRL640 is a 200V logic level mosfet which may do the job for you but I think you would need to wire 3 in parallel to get a stable 130W (and a good heat sink) - I have not tested this mosfet yet in my circuit but it may be worth a try. You would also need to isolate the mosfet from the heat sink (suggest a mica washer and heat sink compound) for safety, due to the higher voltage. You would also need to adjust the resistor values in my circuit which monitors the load voltage so as to ensure that it drops to a range from 0 to 5V (may need also a tweak in software to allow for that and do the appropriate calculations in software). Regards, Louis
Scullcom Hobby Electronics - thank you, Louis. That all sounds very doable and encouraging . I think I have some IRL610's in my transistor trays, so I will look into it more. Many thanks again.
Its the IRL640 - think you made a miss typo.
First, excellent tutorial and thanks for share your knowledge.
At 13:10 you started talk about the temperature dissipation on heat sink, so in my case, I'm using two mosfets to control and share the power. How should I calculate the power dissipation in this case?
Again, thanks. I appreciate your initiative.
I will be looking at parallel mosfets in Part 4 as well as some alternative Mosfet types. As they will share the load the power dissipation will be approx. half in theory although in practice you would have to add a little leeway.
Ok, understood.
May I suggest IRFP27N60K for analysis?
Thanks
The IRFP27N60K although it will handle the power it is not a Logic Level Mosfet so it would not work in this project. The IRFP27N60K requires higher gate/source voltages of more than 5 volts so if you were to use it in this project you would need to add a gate drive IC to provide the higher gate/source voltage it needs.
Thanks for a nice video. (Again) Keep up up the nice work.
Thanks Christian. I have some further updates planned for Part 3.
thinks you so much
Thank you.
So I've been looking into making an AC voltage reference but cab literally find no information on anyone doing one which confuses me. Maybe this would be a good future project?
Thanks this may be a good idea for a new project.
Scullcom Hobby Electronics I started a discussion about it on eevblog. its getting pretty good.
Scullcom Hobby Electronics Ok I really hope you do because I'm completely baffled about how to do this economically and practically. I really thought there would be an ic where you put in so many volts DC and it spits out so many volts ac rms or P 2 P But apparently it's a lot more complicated than that.
Hi, there great work love these projects.
But can anyone please tell me how to put the firmware on the Arduino?
easy step ect as am new to programming.
Thank you
Hi, Louis! I tried to use a ds1307 and M41T56. I changed all 0x6f address to 0x68 in MCP79410_Timer.cpp and in "timer" example. But it seems to me this is not enough. Can you present a text from port terminal for "timer" example? I get the following data:
Timer Test
==========
-----------
Timer Reset
-----------
-------------
Timer Started
-------------
1
00:00:01
1
00:00:01
1
00:00:01
1
00:00:01
1
00:00:01
-------------
Timer Stopped
-------------
1
00:00:01
2
00:00:02
-------------
Timer Started
-------------
3
00:00:03
3
00:00:03
3
00:00:03
3
00:00:03
3
00:00:03
-----------
Timer Reset
-----------
0
00:00:00
1
00:00:01
-------------
Timer Started
-------------
2
00:00:02
2
00:00:02
2
00:00:02
2
00:00:02
2
00:00:02
==========
Test Ended
==========
Super!
Hmmm... The mosfet part number after a quick search at mouser seems to be obsolete...
Yeah same for Digi-Key and Newark... Perhaps something in the ST line?
Would also be interested in an alternative due to availability
Question, are a MOSFet designed to operate in linear mode? I've always thought for this type of application it's better to use a bipolar transistor? In my DC-load I use a TIP35c. Am I wrong?
Yes some MOSFETS are designed to operate in the Linear Mode.
A MOSFET is a voltage driven device, so the op-amp doesn't have to drive any current. With a Bipolar Power Transistor (BJT) the Op-Amp must supply the base current. Bipolar Power Transistors may require a high drive current. You could use a BJT in these type of circuits but there would be an error due to the base current being added to the load current passing through the sense resistor. At higher voltages and high current levels MOSFET's are more efficient than BJTs.
Viewers may be interested in using the TI CSD18502KCS Power MOSFET. While not specified as a "logic level" device, its Rds(on) @ 4.5 Vgs is 4.5 mOhms and @ 10 Vgs is 2.9 mOhms. However, as a substitute, it should be fine, and might be more easily sourced, especially in North America. It is $2.15 in singles from DigiKey.
Thanks Richard for the info. I had a look at the Mosfet you suggested as a possible alternative. I think it should work but with a few caveats. It should be OK for up to 30V Load, absolute Max Drain/Source voltage is 40V. Looking at the Safe Operating Area in the datasheet, at 30V its Max current at DC level would be 2 Amp. Also the Junction-to-Case Thermal Resistance is higher at 0.6°C/W so you would have to take that in to account. Having said all that it should be OK if you limit your maximum power to say 50 watts and have a good heatsink or better still use 2 in parallel. I will get hold of a few of these Mosfets and test them. I will be testing a range of alternative Mosfets and feed the info back in later parts of the project. Thanks again Richard for your input, I am sure it will be of help to other viewers. Regards, Louis
I'm using the heat sink/fan unit from an old graphics card. This are "throw away's" from old PCI bus XP machines that had upgraded graphics capability.
Yes these old PC heatsink and fan come in handy and for free!
Hi, where you could get 0R1 35 watt resistor? thanks
great video thank you!
Thanks.
Great. Thumbs up.
Thanks.
Yay part 2.....
Thanks.
Thanks. Nice job. Thumbs up....PEACE
Thanks.
I'm curious where or how you found the .6 c/w for the heatsink I can't seem to find the specs for most of the heatsinks fan combos on eBay for example. most say "95watts max" but no thermal resistance data.
it seems some of the more well known cpu heatsink manufactures do post thermal performance graphs which allows us to calculate c/w, thermaltake is one example (www.thermaltakeusa.com/products-model.aspx?id=C_00002014).
I also found the follow calculator that helps estimate thermal resistance based on material type, base thickness, fin thickness and number of fins.
www.myheatsinks.com/calculate/thermal-resistance-plate-fin/#
Hi David,
You can find details for a wide range of heatsinks fan combos on the link below:
www.dansdata.com/coolercomp.htm
Regards,
Louis
Thanks David for the links. However they did not provide specification in °C/W. The link I sent you however does provide that data.
Regards,
Louis
Keep up your good Work please;)
Thanks.
♥️👍👏