I vote yes for a CV mode follow up!👍 I do find it funny how these chinese brands use the silliest names! Atorch, Junctek, Ultrafire etc.. This one is more relevant I guess 😁 That hysteresis on the fans could get annoying, I assume there is no control for speed.
You are missing the voltage sense wires (4 terminals into the board, the right hand two) you'll need those so that it's battery/source voltage measurement is not being affected by the current draw on the main lines. It's in the manual I think...
A quick way to clear mAh/Wh is to hold + and - at the same time for a couple seconds. Big time saver over entering the menu and going all the way down. Any idea what that last setting in the menu is? Something about no load current? Also have you tried calibrating it? Mine came fairly accurate out of the box but could use a minor tweaking. Just don’t want to mess it up although I think you can change it back to the default original calibration.
The DL24 calibration is in a "hidden" menu and is only in Chinese. To access the menu, here are some instructions found online: Hello, friend, when the ATORCH mark appears during startup, double-click the M key to enter the calibration mode. First, use 1V to clear and calibrate, then use 30V voltage to calibrate, and then use 3A current to calibrate" Since it is only in Chinese, you can use a translation "app" with the possibility of graphic input
Resistors probably are for ballancing mosfets becouse they are not ideal and have diffrent Rdson. Also you can mesure votage drop at 35A on input diode and calculate power on it Vf x 35A
My biggest gripe with these units is where the thermistor is located for the board temp. It's next to the shunt resistors onboard and a few mm away from the heatsink. The temperature of the silicon is much higher than reported by the board temp reading. Running the board at its full rated 150W will result in really hot (100c+) mosfets while the controller only reports 60c
You have two ways to remedy this problem. The first is a partial but easy remedy, just put some thermally conductive but electrically insulated material on the thermistor until it makes thermal contact with the heatsink. I did it and the temperature can be 3-5 degrees closer to the real one, but the time delay remains, i.e. the time it takes for the heat to reach the thermistor. The other solution is to use the socket for the external sensor and place it in close contact with one of the mosfets, in this case you will have an almost perfect reading in real time
I would say that is what they are for. They are probably not balancing power through the modules other than those resistors. Maybe that is why they are using that special mosfet with a high on resistance because it is more consistent in it's linear region. I would still always double check the heat on each mosfet to see which one runs the warmest. It will probably be on the mother board because that has the highest voltage and most gate voltage.
It's always in DC load mode. Like it or not, MOSFETs have to work in this mode. PWM mode is undesirable because many power sources do not function correctly in it.
Is it capable of handling 35/40A for a longer period? I am looking for a decent electronic load for testing heavy LFP cells up to 280Ah, so i need to burn a lot of current/power. Did you also test for accuracy? These are the 2 main parameters of dummy loads - i guess 🙂
I really don't think so without making major changes that would significantly increase the total cost. It is not a product for professional use and was not created less for intensive and heavy use
@@rilosvideos877 Looking at the DL24MP under review here, does the connector on the main board through which the 40 Amperes pass, or should pass, seem adequate for that current? Does the (double) schottky diode at the input seem adequate for 40 A? do the same tracks of the main board PCB that carry the 40 A flow seem suitable to you? Does a card with those 4 heatsinks and the 8cm fan seem suitable for dissipating 150W continuously? Look, there are many things that make one think that the data declared is a lot of advertising. In reality the object is nice and at half the declared values it is quite fine, it is not the object that is a fake but the data too pumped up on the power and Ampere
Bought one & answered my own question. "Setup" in the mobile app does nothing. The pc version offers full control with curve graphs & data output retention.
What did you test the constant voltage with ? For constant voltage you will need to use some sort of constant current source like say a solar PV panel else of course it can not do any constant voltage limiting. You can also test constant voltage if you add some resistive load in series with a voltage source like say adding an incandescent lamp in series with your battery or super capacitors. Maybe you already know all this and the DC Load can not maintain a constant voltage for some reason. If the other modes work not sure why constant voltage will not work.
I tried a power supply in CC mode - wildly unstable. Then I tried a battery in series with a bulb - also very unstable. Then the battery in series with a power resistor - more stable, but still not good.
@@JulianIlett Maybe with the control loop of the CC power supply and the DC load interfered with each other. The battery and resistor should work and if that is not working then solar panel will also not work.
The battery and bulb surprised me. I'm guessing it's because of the bulb's non-linear resistance. The bulb was squealing like crazy - worth a video just to hear it :)
@@JulianIlett Yes bulb will be no linear resistor still resistance change will not be that fast. Maybe is the higher inductance of the bulb that created the problem compared to a power resistor.
That reverse polarity input diode will not be carrying anything, until you reverse the polarity, then it protects the board by conducting hopefully enough to blow a fuse or trip.
True. It will be against to the nature of the unit to put it in line (series) because of the voltage drop it will cause. It must be reverse biased parallel, but then why put a heat sink on it?
I think the heatsink is to stop it overheating whilst it trips any input fuse. If it fails then the unit is damaged. This is a firesafe device.@@borayurt66
@@G1ZQCArtwork Depends on what you connect there. This being a DC load / battery capacity meter, batteries with very high short circuit currents will often get connected. With such a battery having hundreds of amps under shortcircuit, I doubt that the heat sink can do much to keep that diode from splitting to its molecules in a blink. 😀
The protection of the diode which blows a fuse because it becomes conductive when you put the wires with reversed polarity is used in the DL24EW model, not in this one. Here the diode conducts when the connections are with the correct polarity and prevents the current from flowing if the connection polarity is wrong. This obviously involves, under normal conditions of use, a power dissipation on the diode which becomes "considerable" when the currents are high; this is why there is a heat sink which is also hit by the air flow of the fan.
Před 3 měsíci
Can you try to bring this capacity tester to the limit? I was buying the small version rated 180 watts. After I connected it to a 84 volts battery, I had to replace the MOSFET 3 times at only 0.8 amps because it blew up 3x.
It may happen that the DL24MP reads small currents even when there is no load connected, this is due to the sensitivity which has been increased compared to previous models and very small shunt sensors. With the last setting you can cancel this bias error and make the measurement more reliable. But remember that everything is not stable over time and with external temperature fields. Too many figures to claim to be precise and reliable on an object of this type. Furthermore, I wouldn't think about using it to the limit, as I have already written elsewhere, too many savings: switching type MOSFETs not suitable for linear use as needed in an electronic load and often also "fake"; "ridiculous" heat sinks for the intended task anyone who knows how to assemble a PC knows what it takes to dissipate 150W away from a CPU. And then everything is up in the air, no safety regulations and even dangerous voltages (200v are dangerous) on the entire surface of the heat sinks. Oh by the way when did you replace the mosfets where did you buy them? do you know that the market is full of fakes of this type that can't last?
Před 2 měsíci
@@francescacosentino6975 when it's rated 150 watts you wouldn't go near to 150? I'm aware of the replicas. I was buying from a reliable supplier. They blew up same fast as the original MOSFET that came with the board. Finally I draw only 50 of 180 watts in 10 days. The temperature was not close to anything dangerous. Maybe the blew-up's caused by the avalanches from the PWM. Did you use a battery with higher voltages than 48 volts?
@ When you write They exploded at the same speed as the original mosfet why do you use the "singular" are you perhaps talking about the previous model which only had one mosfet? I don't come close to the declared power of 150W because it is a Chinese product made in an artisanal way and with many defects. First of all the use of switching MOSFETs where the manufacturer does not guarantee the survival in the conditions of use of this device: in fact in the SOA (Safety operation area) graph of this MOSFET there is neither the curve referring to DC operation nor with the case and the mosfet at 75°: The most demanding condition where the manufacturer takes responsibility for guaranteeing the functioning of the mosfet is therefore with pulses of 10 msec max and with a case temperature of 25°C, both very far from those when using the DL24MP. The second major flaw is the heat dissipation with heat sinks that are too small for the powers involved even with forced ventilation. By installing better heat sinks I managed to have a temperature of around 10-15°C lower with the same dissipated power
Oh boy I’m so excited to see your opinion on this. I got the 4 module 600W/40A version. Can’t get a 600W/40A electronic load for anywhere NEAR the $62 I paid on sale. Like not even in the same galaxy can you get an electronic load that powerful. But….the design sucks. I mean it’s truly a poorly implemented idea that makes sense and doesn’t make sense. Compressing PCB’s together?! With a design that leaves some of them not level with the rest? WTF? I initially couldn’t draw more than 33/34A. Atorch told me to change the jumper configuration for 450W….which was an issue because at 40A I’d be drawing more than 480W… So I went back over all the nuts and bolts and very carefully applied a bit more torque and found a couple that were just slightly less torqued than the rest. Keep in mind you have to do the upside down T configuration that’s 3 wide and 2 tall in the middle. Nowhere in the literature do they say how to assemble the damn thing. I had considered the 2x2 square but that would still leave 2 modules not level with the other two but I figured I could mod the feet with shims. I think right now I have the middle unit on top of the outer modules which works ok. I always have the unit elevated an additional 12 inches off the ground because additional heat dissipation is absolutely necessary. The heat sinks are TERRIBLE and the fans not running until a set temp is a terrible design. I’d much rather have a variable speed fan that goes on over 10% of the units max rating or some 5-10 Celsius over ambient temp. Currently I have it set for I believe the lowest setting, 40 Celsius. Those motherfuckers are loud and for how loud they are they provide minimal cooling. 75% of my cooling comes from either a 90mm PC fan ducted through a Poptart box to go underneath the elevated PCB’s and another less powerful PC fan on top of that to get the power resistors on the board some airflow. Seems to work ok but if my testing is longer than a minute I use a quite powerful 12” desk fan on its highest setting when I’m doing anything over 200W. Over 400W I use two PC fans and the 12” desktop fan to keep things under 60 Celsius but it’s bloody freaking loud. Really tempted to change the stock fans and honestly might look at the heatsinks as well. If I can find salvaged heatsinks that would fit I’d gladly swap them out. Would love to find some copper.
I agree the initial idea may be interesting but the realization makes you cry. First of all the choice of mosfets. The IRFP264 that is used is not suitable for linear uses but is a switching one. Its SOA (safety operation area) does not exist for continuous use, therefore the manufacturer does not guarantee that it will not go up in smoke when used as a linear load. Then I agree that the heat sinks are unnecessarily small and a little was enough to obtain good advantages. For the use of copper I have seen that it has great advantages on the propagation of heat and therefore in transients; but when fully operational it does not have much greater dissipation towards the air than aluminum and in any case is not able to justify the price which exceeds that of the same DL24MP device
The Atorch set system is a bit mickey mouse, takes a while to make it work for you. Alright once set up for a batch of batteries say , 100's on a test bench, but only 450W for all that faff and $$$$. I made up a bucket of water and some nichrome wire to do a 50A test on a 25v lipo bank, much cheaper $ and I wanted to watch temp rise over 15-30mins and check all over the bank for hot spots
WASTE of my money. it doesn't even charge it to full first. time and more money gone trying to simply match some old cells. I can do this with a few light bulbs
I vote yes for a CV mode follow up!👍
I do find it funny how these chinese brands use the silliest names! Atorch, Junctek, Ultrafire etc.. This one is more relevant I guess 😁 That hysteresis on the fans could get annoying, I assume there is no control for speed.
Yes, the fans run at a fixed speed, just on and off.
Yes, Julian… I’m voting for a follow-up video as well. Thank you! This is intriguing!
You are missing the voltage sense wires (4 terminals into the board, the right hand two) you'll need those so that it's battery/source voltage measurement is not being affected by the current draw on the main lines. It's in the manual I think...
Yes, if I was taking accurate measurements, I'd use 4-wires.
@@JulianIlett I was suggesting that might be why you were having problems with the CV mode...?
Yes, I see what you mean. I'll try it again using shorter wires and 4-wire measurement.
A quick way to clear mAh/Wh is to hold + and - at the same time for a couple seconds. Big time saver over entering the menu and going all the way down.
Any idea what that last setting in the menu is? Something about no load current?
Also have you tried calibrating it? Mine came fairly accurate out of the box but could use a minor tweaking. Just don’t want to mess it up although I think you can change it back to the default original calibration.
The DL24 calibration is in a "hidden" menu and is only in Chinese.
To access the menu, here are some instructions found online:
Hello, friend, when the ATORCH mark appears during startup, double-click the M key to enter the calibration mode.
First, use 1V to clear and calibrate, then use 30V voltage to calibrate, and then use 3A current to calibrate"
Since it is only in Chinese, you can use a translation "app" with the possibility of graphic input
Resistors probably are for ballancing mosfets becouse they are not ideal and have diffrent Rdson. Also you can mesure votage drop at 35A on input diode and calculate power on it Vf x 35A
Yes, I'm thinking that's what "In Temperature" is showing. It's the temperature of the input diode calculated from its forward voltage.
My biggest gripe with these units is where the thermistor is located for the board temp. It's next to the shunt resistors onboard and a few mm away from the heatsink. The temperature of the silicon is much higher than reported by the board temp reading. Running the board at its full rated 150W will result in really hot (100c+) mosfets while the controller only reports 60c
You have two ways to remedy this problem.
The first is a partial but easy remedy, just put some thermally conductive but electrically insulated material on the thermistor until it makes thermal contact with the heatsink.
I did it and the temperature can be 3-5 degrees closer to the real one, but the time delay remains, i.e. the time it takes for the heat to reach the thermistor.
The other solution is to use the socket for the external sensor and place it in close contact with one of the mosfets, in this case you will have an almost perfect reading in real time
Those 0.25 ohm resistors... are they for current sharing in the fets?
Yes, I guess so.
I would say that is what they are for. They are probably not balancing power through the modules other than those resistors. Maybe that is why they are using that special mosfet with a high on resistance because it is more consistent in it's linear region.
I would still always double check the heat on each mosfet to see which one runs the warmest. It will probably be on the mother board because that has the highest voltage and most gate voltage.
I'm not sure you'd see much difference. The MOSFETs are on common heatsinks, so any heat generated would quickly spread among all the transistors.
Actually there's a mixture of .25 ohm and .22 ohm, so the resistor value can't be that critical.
@@JulianIlett true, the difference seems to be negligible but at maximum power it could become a factor, at least between each board.
Thanks! Rumors that the Control Panel PCB has an unpopulated socket for rotary encoder. Can you confirm or deny?
is it acting as a DC resistive load or a PWM load? normal mosfets doesn't like the linear region, by what I understand.
It's always in DC load mode. Like it or not, MOSFETs have to work in this mode. PWM mode is undesirable because many power sources do not function correctly in it.
Yes for the problems video. Also, is there no sense (4 wire) input?
There is yes but he had it hooked up in 2 wire mode
Can't believe you didn't pull out your thermal camera on this :)
I'm sure it would just confirm temperatures between 30C and 40C as indicated by the board temperature parameter.
Is it capable of handling 35/40A for a longer period? I am looking for a decent electronic load for testing heavy LFP cells up to 280Ah, so i need to burn a lot of current/power. Did you also test for accuracy? These are the 2 main parameters of dummy loads - i guess 🙂
I really don't think so without making major changes that would significantly increase the total cost.
It is not a product for professional use and was not created less for intensive and heavy use
@@francescacosentino6975but it claims to run 40A cont.? So you guess this is chinese fake?
@@rilosvideos877 Looking at the DL24MP under review here, does the connector on the main board through which the 40 Amperes pass, or should pass, seem adequate for that current? Does the (double) schottky diode at the input seem adequate for 40 A? do the same tracks of the main board PCB that carry the 40 A flow seem suitable to you? Does a card with those 4 heatsinks and the 8cm fan seem suitable for dissipating 150W continuously?
Look, there are many things that make one think that the data declared is a lot of advertising.
In reality the object is nice and at half the declared values it is quite fine, it is not the object that is a fake but the data too pumped up on the power and Ampere
Does this mean you can have one bit in each part of the modular shed, 🙂
I'm sure that could be made to work :)
Can the settings be changed inside the app? What's in the "setup" (bottom left button) part?
I think 'setup' is the M button.
@@JulianIlett In the app. Bottom left.... "Setup".
Where does that lead?
Bought one & answered my own question. "Setup" in the mobile app does nothing. The pc version offers full control with curve graphs & data output retention.
What did you test the constant voltage with ? For constant voltage you will need to use some sort of constant current source like say a solar PV panel else of course it can not do any constant voltage limiting. You can also test constant voltage if you add some resistive load in series with a voltage source like say adding an incandescent lamp in series with your battery or super capacitors. Maybe you already know all this and the DC Load can not maintain a constant voltage for some reason. If the other modes work not sure why constant voltage will not work.
I tried a power supply in CC mode - wildly unstable. Then I tried a battery in series with a bulb - also very unstable. Then the battery in series with a power resistor - more stable, but still not good.
Maybe it would work OK with a solar panel - I didn't try that.
@@JulianIlett Maybe with the control loop of the CC power supply and the DC load interfered with each other. The battery and resistor should work and if that is not working then solar panel will also not work.
The battery and bulb surprised me. I'm guessing it's because of the bulb's non-linear resistance. The bulb was squealing like crazy - worth a video just to hear it :)
@@JulianIlett Yes bulb will be no linear resistor still resistance change will not be that fast. Maybe is the higher inductance of the bulb that created the problem compared to a power resistor.
That reverse polarity input diode will not be carrying anything, until you reverse the polarity, then it protects the board by conducting hopefully enough to blow a fuse or trip.
True. It will be against to the nature of the unit to put it in line (series) because of the voltage drop it will cause. It must be reverse biased parallel, but then why put a heat sink on it?
I think the heatsink is to stop it overheating whilst it trips any input fuse. If it fails then the unit is damaged. This is a firesafe device.@@borayurt66
@@G1ZQCArtwork Depends on what you connect there. This being a DC load / battery capacity meter, batteries with very high short circuit currents will often get connected. With such a battery having hundreds of amps under shortcircuit, I doubt that the heat sink can do much to keep that diode from splitting to its molecules in a blink. 😀
The protection of the diode which blows a fuse because it becomes conductive when you put the wires with reversed polarity is used in the DL24EW model, not in this one.
Here the diode conducts when the connections are with the correct polarity and prevents the current from flowing if the connection polarity is wrong.
This obviously involves, under normal conditions of use, a power dissipation on the diode which becomes "considerable" when the currents are high; this is why there is a heat sink which is also hit by the air flow of the fan.
Can you try to bring this capacity tester to the limit? I was buying the small version rated 180 watts. After I connected it to a 84 volts battery, I had to replace the MOSFET 3 times at only 0.8 amps because it blew up 3x.
It may happen that the DL24MP reads small currents even when there is no load connected, this is due to the sensitivity which has been increased compared to previous models and very small shunt sensors.
With the last setting you can cancel this bias error and make the measurement more reliable.
But remember that everything is not stable over time and with external temperature fields.
Too many figures to claim to be precise and reliable on an object of this type.
Furthermore, I wouldn't think about using it to the limit, as I have already written elsewhere, too many savings: switching type MOSFETs not suitable for linear use as needed in an electronic load and often also "fake"; "ridiculous" heat sinks for the intended task anyone who knows how to assemble a PC knows what it takes to dissipate 150W away from a CPU. And then everything is up in the air, no safety regulations and even dangerous voltages (200v are dangerous) on the entire surface of the heat sinks.
Oh by the way when did you replace the mosfets where did you buy them? do you know that the market is full of fakes of this type that can't last?
@@francescacosentino6975 when it's rated 150 watts you wouldn't go near to 150? I'm aware of the replicas. I was buying from a reliable supplier. They blew up same fast as the original MOSFET that came with the board. Finally I draw only 50 of 180 watts in 10 days. The temperature was not close to anything dangerous. Maybe the blew-up's caused by the avalanches from the PWM. Did you use a battery with higher voltages than 48 volts?
@ When you write
They exploded at the same speed as the original mosfet why do you use the "singular" are you perhaps talking about the previous model which only had one mosfet?
I don't come close to the declared power of 150W because it is a Chinese product made in an artisanal way and with many defects. First of all the use of switching MOSFETs where the manufacturer does not guarantee the survival in the conditions of use of this device: in fact in the SOA (Safety operation area) graph of this MOSFET there is neither the curve referring to DC operation nor with the case and the mosfet at 75°: The most demanding condition where the manufacturer takes responsibility for guaranteeing the functioning of the mosfet is therefore with pulses of 10 msec max and with a case temperature of 25°C, both very far from those when using the DL24MP. The second major flaw is the heat dissipation with heat sinks that are too small for the powers involved even with forced ventilation. By installing better heat sinks I managed to have a temperature of around 10-15°C lower with the same dissipated power
You need to connect the inner connections to the outer ones to get the CV function to work!
7:01 using u shape terminals.
I think that is a DL24MP (The P is for Pink latest upgrade)
Couldn't you use this for a linear power supply
Oh boy I’m so excited to see your opinion on this.
I got the 4 module 600W/40A version. Can’t get a 600W/40A electronic load for anywhere NEAR the $62 I paid on sale. Like not even in the same galaxy can you get an electronic load that powerful.
But….the design sucks. I mean it’s truly a poorly implemented idea that makes sense and doesn’t make sense. Compressing PCB’s together?! With a design that leaves some of them not level with the rest? WTF? I initially couldn’t draw more than 33/34A. Atorch told me to change the jumper configuration for 450W….which was an issue because at 40A I’d be drawing more than 480W…
So I went back over all the nuts and bolts and very carefully applied a bit more torque and found a couple that were just slightly less torqued than the rest. Keep in mind you have to do the upside down T configuration that’s 3 wide and 2 tall in the middle.
Nowhere in the literature do they say how to assemble the damn thing. I had considered the 2x2 square but that would still leave 2 modules not level with the other two but I figured I could mod the feet with shims. I think right now I have the middle unit on top of the outer modules which works ok. I always have the unit elevated an additional 12 inches off the ground because additional heat dissipation is absolutely necessary. The heat sinks are TERRIBLE and the fans not running until a set temp is a terrible design. I’d much rather have a variable speed fan that goes on over 10% of the units max rating or some 5-10 Celsius over ambient temp. Currently I have it set for I believe the lowest setting, 40 Celsius. Those motherfuckers are loud and for how loud they are they provide minimal cooling. 75% of my cooling comes from either a 90mm PC fan ducted through a Poptart box to go underneath the elevated PCB’s and another less powerful PC fan on top of that to get the power resistors on the board some airflow. Seems to work ok but if my testing is longer than a minute I use a quite powerful 12” desk fan on its highest setting when I’m doing anything over 200W. Over 400W I use two PC fans and the 12” desktop fan to keep things under 60 Celsius but it’s bloody freaking loud.
Really tempted to change the stock fans and honestly might look at the heatsinks as well. If I can find salvaged heatsinks that would fit I’d gladly swap them out. Would love to find some copper.
Completely agree. It's good value for money, but far from perfect. I would never run it at its theoretical maximum power.
I agree the initial idea may be interesting but the realization makes you cry.
First of all the choice of mosfets. The IRFP264 that is used is not suitable for linear uses but is a switching one. Its SOA (safety operation area) does not exist for continuous use, therefore the manufacturer does not guarantee that it will not go up in smoke when used as a linear load.
Then I agree that the heat sinks are unnecessarily small and a little was enough to obtain good advantages.
For the use of copper I have seen that it has great advantages on the propagation of heat and therefore in transients; but when fully operational it does not have much greater dissipation towards the air than aluminum and in any case is not able to justify the price which exceeds that of the same DL24MP device
The Atorch set system is a bit mickey mouse, takes a while to make it work for you. Alright once set up for a batch of batteries say , 100's on a test bench, but only 450W for all that faff and $$$$. I made up a bucket of water and some nichrome wire to do a 50A test on a 25v lipo bank, much cheaper $ and I wanted to watch temp rise over 15-30mins and check all over the bank for hot spots
WASTE of my money. it doesn't even charge it to full first. time and more money gone trying to simply match some old cells. I can do this with a few light bulbs
hello Julian Ilett im looking for youre 3d files but i can find them at thingverse. i thouth the files where on there?
Sorry, I've never used thingiverse.
you made a 3d file case around youre pcb's for the rd series powersupply or not ?@@JulianIlett
ooh no sorry it was simple electronics hahah ty for the reply@@JulianIlett