Postbag Update - 25A Buck Converter

[@rilosvideos877]

The voltage difference (input to output) is very low - its actually more the natural drop down so it doesn't really have to take a load. Give it 50 or 60V input and a 12V / 20A output to see if it really can handle the load? I doubt it a bit, to be honest. 50V/20A would mean 1 kW for this little unit - not impossible but ... Lets give it a decent test! 🙂 The 120 A rating of the mosfets in the datasheet doesn't mean anything practically. Its a labority value under ridicoulos cooling conditions - you can never achieve in practice. Is it current limiting? If not, the sense resistors would do anything if shorted - only in cc-mode this would have an effect, i guess.

[@Sylvan_dB]

It's rated for up to 25 amps OR up to 600 watts.

[@grahamhall2662]

Yes my thoughts. Not really a test at all.

[@anwin85alon]

Pm me can you to install do sola paint of charger`

[@earlyadapter643]

The mosfet current rating is for the chip inside but in most cases the to220 transistor is package limited :)

[@alasdair4161]

I have been using a couple of those regulators to drop HV solar panels down to a suitable voltage to feed my cheap fake MPPT solar charge regulators. What I found most interesting is the final system efficiency is even better than a much more expensive MPPT solar regulator. I set the buck regulator output voltage to 16V into the cheap PWM regulator which delivers a full 22Amps into my load, with the MPPT controller I get just under 20A from the same panel, both running at 14.5V final output. Without the buck regulator I get just over 6A with PWM regulator alone, but with this system, as the current drops the solar panel voltage soon exceeds the PWM maximum input voltage. So after discovering this hidden benefit I did a second panel the same way and get consistently better performance than the big dollar mppt controller, the cheap solar charge regulator gives me full data logging and LVC battery protection along with USB outputs and direct in/out ports. The total expenditure is so low that I can afford to do multiples to gain large energy levels. I also purchased several of the more expensive 800 watt buck versions with CC/CV to use with my 430Watt panels and I get almost the same gains. I also find the buck efficiency gets more charge energy on cloudy days and early/late sun than the MPPT chargers manage to do, the high panel voltage coupled with high buck efficiency to transform the voltage down just works so well at very low light levels. I suspect my overall gains are probably much better, mainly as it is reaping benefits by efficiently charging low voltage (12V) batteries from HV panels. Time will tell as I log more data.

[@JulianIlett]

Have you looked at the buck/boost units with MPPT pot?

[@alasdair4161]

@@JulianIlett Not as yet, if they can squeeze even more efficiency out I'll have to give them a try... do you have any video reviews on them?

[@tomgrant6563]

I was thinking of doing similar- but for the purpose of keeping my pv wire voltage high with 4 or 5 panels in series to reduce losses over my 80m long wires. I would need a buck converter that stepped down the roughly 250vdc to about 100v which I would then feed into my mppt chargers. I can't find any converters that run at that high a voltage though.

[@alasdair4161]

@@tomgrant6563 Good idea, I did a similar trick to operate our gates about 140m from the shed, I run 24VDC into a boost converter, send it to the gate box at 55VDC (maximum the cheap 10A boost converter would do), then I buck regulate it back down again to 24V at the gate box. On load when the gates are moving the peak voltage drop over the 140m is about 20V, but the buck keeps the output at exactly 24V up to about 7A load. It was much cheaper than trenching in mains or running super heavy gauge wire. You may be able to run HVDC into a regular switchmode power supply for broad voltage input conversion down to whatever voltage you like. Most will work happily on DC input up to 340V (250VAC), and down to around 150DC(100VAC) as they internally rectify the AC input anyway and self regulate for 100-250VAC input.. Good luck.

[@tomgrant6563]

@@alasdair4161 haha that's a bizarre- I figured that out today before I saw this comment! after a fair few days of researching and scratching my head! Exactly as you described. Switch mode power supply off the dc.

[@GannDolph]

It's not really bucking in this test - just a fancy diode drop. Try feeding 48V maybe?

[@anwin85alon]

Pm me can you to install do sola paint of charger`

[@GannDolph]

@@anwin85alon man, wut you smok'n ?!!??

[@i-TEQ_power]

Buck is tl494 and driver ir2104, my findings based on the circuit schematics

[@anwin85alon]

Pm me can you to install do sola paint of charger`

[@earlyadapter643]

Thanks a lot. Do you know what is the small buck voltage regulator that makes the 12 V for the tl494 and the fan ?

[@ahaveland]

ah, banana plugs and spade terminals. Reminds me of my electronics class 50 years ago!

[@JulianIlett]

Those and XT60s are my favourites :)

[@ahaveland]

@@JulianIlett Agree with the XT60s - such a useful open-source and versatile connector, it's like Lego, so easy to make distribution blocks and adapters with them, though it's a struggle with 10AWG cable. I also have to use XT90s since I can use 100s of amps for large ebike and inverters.

[@JulianIlett]

Yes, I recently used 12AWG silicone wire with XT60s - it wasn't easy. The male connector has deep recesses behind the solder cups, but the female doesn't. I'm tempted to make a video about that.

[@ahaveland]

@@JulianIlett tip, when soldering XT connectors, plug it into its opposite connector to hold the pins in place in case the plastic starts to soften or melt. Holding vertical may be best. The void could be filled with a piece of solid core copper from cooker cable. Saves solder and improves conductivity. Also good for building distribution blocks splitters, series, parallel and gender changers. Encase in hot glue and shrink-wrap.

[@anwin85alon]

Pm me can you to install do sola paint of charger`

[@SuperBrainAK]

I think those shunt resistors are irrelevant except for a hard set over current protection, most likely set at 25A. They do nothing when running in a CV mode which is where it was running the entire time you were testing. I suggest soldering right over them as your overcurrent will be from upstream devices anyway. This will keep any ground loops as minimal as possible so you don't damage any small interconnects between the cards.

[@anwin85alon]

Pm me can you to install do sola paint of charger`

[@Sylvan_dB]

A patch of black vinyl electrical tape or even black marker can help reading temperature off the heat sink.

[@anwin85alon]

Pm me can you to install do sola paint of charger`

[@james10739]

I definitely need a 25a buck converter laying around especially at that price

[@alextrezvy6889]

Thanx for the review. This is exactly what I was looking

[@ericklein5097]

They do make a version of this with current control. The pot would be on the left side near that mosfet you looked up. Haven’t looked what else they change on the PCB. I’m only interested in these buck converters if they have current control. Obviously the other issue is the lack of displays. I have seen similar buck converters that also use small 7 segment displays. For under $20 (the variable voltage/current model is $12’ish, the one with displays is a bit more) this isn’t bad for a 15A stable, perhaps 20A with upgraded heat dissipation.

[@JulianIlett]

Yes, I got that version. It looks like a bit of a fudge. The current pot is wedged in, touching the MOSFET. I can't see myself using it.

[@anwin85alon]

Pm me can you to install do sola paint of charger`

[@jayc2469]

I bought something extremely similar to this in around 2019 but mine was sold as a Yeeco DC Buck Converter DC-DC Step Down 10A Voltage Regulator but mine has an LCD display showing the Input and Output Voltage and Current. I could see exactly what was happening when you plugged the lights in - it immediately started to Clamp the output. Adjusting the Current pot would have remedied it but on my board it has 3 Indicator lights denoting when Current Clamping occurs, allowing it to be adjusted up or down. My board came with a clear acrylic 'shell' but I recently had to cut myself a new one using my solid state laser

[@anwin85alon]

Pm me can you to install do sola paint of charger`

[@____________________________.x]

If you had a scope you could measure the noise, which is quite significant on these modules

[@james10739]

I feel like he does have an oscilloscope

[@____________________________.x]

@@james10739 Never seen one on this channel. His bench is oddly unequipped for the hobby, not even a bench power supply which is generally the first thing everyone builds?

[@james10739]

I think I have but I'm not that sure

[@james10739]

And he has 12v and plenty of boost and buck converters so a power supply might be a little more convenient but not really necessary

[@____________________________.x]

@@james10739 it’s pretty central to have a box with V & A meters and current control, he wastes hours fiddling with DC-DC modules instead of just hooking one up

[@earlyadapter643]

Very informative video, thanks for it, i recently bought two such modules and there the transistors are HYG043N10. I'm playing around with swapping transistors, for example the 400 W boost converter that has very cheap transistors can be upgraded to go to higher voltages, and also the commutation losses depend not only on R_DS,on but transconductance and input capacity so the best is just to try which one gets less hot. Nice to learn from the comments that the parts are tl494 and driver ir2104, so i guess the other ic is a little buck converter only for the fan ? The part numbers are erased on my modules.

[@johnnodge4327]

The more current you pull through those measuring resistors, the more effect bridging them out will have.

[@JulianIlett]

I'm not so sure now. There's another variant of this converter that has a current limit pot. Maybe this unit uses those current sensing resistors to perform a simple shutdown at 25A. Not sure. I've bought the other variant.

[@rich1051414]

If those use a closed loop voltage follower to regulate it's output, I would fear them fighting each other if you deisolate them by allowing the grounds to reference together. No two voltage references are exactly the same, and it would send each other into current regulation allowing all the voltage followers to free for all deathmatch regarding what is actually '12v'. A diode would be enough to prevent this from being an issue, but you would have to compensate for that voltage drop and it would be a possible point of efficiency loss. I would use schottky and not silicon diodes.

[@SuperBrainAK]

I'm not sure diodes would actually solve that problem. Besides you should use ideal diode modules if you are working with high currents. So yes with closed loop voltage regulation the problem you have is that one module will always be set at a higher voltage than the others, so because it is the highest voltage the majority of the current will flow from it as it tries to keep the higher voltage than the rest, the others see 12.01v when set to 12v so they shut off completely. It is only until the resistance in the connections or really precise trim pot adjustment will that balance out. What you want is actually a resistor on the positive side so that a small change in voltage doesn't have a large change in current. This has been done many times before with audio amplifiers with multiple transistors in parallel. The reson you want it on the positive side is due to ground loop reasons, if you were to inadvertently ground the chassis of the AntMiner with a stray wire or other metal enclosure your negative side resistance will change, defeating it's purpose. Putting it on the positive is ideal because it is generated at the buck converter and consumed by the miner cards, so nowhere else to potentially connect. All current shunts in my opinion should be on the positive side to avoid creating negative and ground potential differences. I personally use the WD4050 module it is much more expensive but it has all the right features, positive shunt, sunchronous, display, and CV+CV adjustment.

[@SuperBrainAK]

I also believe running one module per card would also completely eliminate the "fighting each other" issue, if the 12v output is not in parallel on the positive side then one card can run just fine at 12.1v where another card can run from another buck module outputting 11.9v. Both modules and cards will output and draw the same wattage. The 11.9v fed card will just draw slightly more current and would run a little less efficient than the rest but there would be no cross fighting since they won't have a way to influence each other. Very good topics to discuss for sure.

[@rich1051414]

@@SuperBrainAK I assume there are positive rail interconnects between the cards. Perhaps the real danger is damaging those interconnects, or anything else that is between them and the differing rail voltages. A diode would limit that effect since the power supply won't try to sink any current itself. But yeah, that would be a 0.5v drop you would have to deal with by increasing the power supply voltage and dealing with the heat generation. A resistor to limit current on each would generate even more heat.

[@anwin85alon]

Pm me can you to install do sola paint of charger`

[@earlyadapter643]

I have one more question if the discussion here is still active : does anybody know if the choke is made with aluminum wire and the efficiency can be increased by re-winding it with copper ? There is a discussion going on about the very cheap 400 W stepup module that can be much improved by changing the choke wiring.

[@johnconrad5487]

at 8:40. two .01 resistors in parallel and not .02. the parallel resistance is 0.005 Ohms.

[@anwin85alon]

Pm me can you to install do sola paint of charger`

[@robert574]

Thanks for a great video, I really like your lamp set up and the connectors with the fork terminals. All of your connections really. I have something similar with a lot of socket adapters that will accept US 120V bulbs (it's always been a work in progress). While I was watching this, I would have liked to see a test of the output voltage, set it down to just below the minimum you had with all the lamps on, say 10.000V and then plug and unplug all the lamps to see if the output voltage remained the same. I did just order one of these. It says they operate in the Continuous Current Mode CCM not sure how that's different than Constant Current CC. My reading of CCM finds that the inductor always has current flowing and never zero. hmm... Nothing to do with Constant Current which I care about. It does say it has Constant Current (and short circuit protection). It also needs a flyback diode if powering any inductors. It has no reverse polarity protection on the input (easy to add). There are some really cool small fans that are squirrel cage and don't whine all the time, I need money, where are the car keys... wait, that's something else. I like this little buck converter.👍👍 ⭐

[@anwin85alon]

Pm me can you to install do sola paint of charger`

[@leybraith3561]

Re 12v double filament Headlight Bulbs (@16mins), unless both filaments are same wattage they will have different voltages across them when filaments are wired in series. The lower wattage filament has the higher voltage and will be stressed / may blow especially if you apply the voltage of two fully charged car batteries across them (2x14.4 = 28.8volts). Also having both filaments lit results in an awful lot of heat being produced and is probably not recommended. Years ago I was using 4 paralleled bulbs in this fashion while testing a beefy 28v power supply, It was entertainingly very bright and hot for a couple of minutes before they all blew the low wattage filaments.

[@anwin85alon]

Pm me can you to install do sola paint of charger`

[@Hopatzful]

this dc converter can handle 40v input, output 24v 20a max?

[@fahimahmadosmani3618]

can i use this with solar panel to charge lead acid battery?

[@무-o2v]

Can i replace mdp1991 with other mosfet? Etc ifrp250n?

[@paulfrindle7144]

You haven't got enough voltage difference between in and out to get it to switch and stabilize. It's just passing straight through!

[@mohammadasif8501]

What happend, if we directly connect it to the solar panel and run some DC devices?

[@danyalibnekhalil]

Wah wah Asif saab sadqy tmary sawalon pe 😅

[@panospapadimitriou3498]

watching it this morning the 12a ones a watched for boost my delta2 were sold out ~~~

[@anwin85alon]

Pm me can you to install do sola paint of charger`

[@TheEmbeddedHobbyist]

i expect that that package rating is with the device connected to incredibly big heatsink. it's almost impossible to get the currents they claim on the data sheets in real life. I had a 1KW rated resistor which with 800W blow in about 3 seconds, it was mounted to a heatsink with two very powerful fans but it still died. Later found that we had to machine the heatsink as plain aluminium finish was not flat enough to allow the heat to flow even with a pad. this was in some very very small print. 🙂 we spent as much time designing the dummy load as we did the circuit we were testing.

[@Sylvan_dB]

In software development with thorough test code it is common to spend 2x to 3x the time writing test code as the production code (what we'll deliver). You can get about 50% test coverage if you spend about the same time on test as production code.

[@TheEmbeddedHobbyist]

@@Sylvan_dB Don't get me started on software testing. having done projects to DO-254 (HW) and DO-178C (SW) i hate the saying "can we just make a minor change" 6 months in to testing.

[@anwin85alon]

Pm me can you to install do sola paint of charger`

[@111genti]

Hello Julian, I see almost the same at AE but with a current regulator/pot at about 12euros. Would it fit your purpose?

[@JulianIlett]

I found it (and bought it)

[@anwin85alon]

Pm me can you to install do sola paint of charger`

[@EsotericArctos]

Not sure the unit needs derating due to the connector. 300V at 20 Amp is 6 000 Watts. That is very different to 20 Amps at 12 Volts which is only 240Watts. So the connectors will handle a lot more than 20 Amps at 12 Volts. Derating it for other reasons may be worth while, but not for the connectors.

[@JulianIlett]

Yeah, there are lots of reasons to derate these things. Interestingly, they seem to have fitted a higher rated output connector now: www.aliexpress.com/item/1005005639051641.html

[@neur303]

Hmmm, The connectors don't depend on the wattage of the load but of the wattage of the connection due to the resistance of the junction and materials used, which would be I times the voltage drop along the connector or simpified I^2 * R. I.e. it should only depend on current and connection. Where am I wrong?

[@RexxSchneider]

@@neur303 You're not wrong. 20A means 20A maximum current for the connector, regardless of whatever potential it is at, because of the I2R losses causing overheating. The reason the connectors also carry a voltage rating is because of the possibility of insulator breakdown. Two separate failure mechanisms for overcurrent and overvoltage means you don't just multiply them together as if it were a maximum wattage!

[@legominimovieproductions]

I recently ordered a buck converter on amazon, it had like 15 grams of solder spread as one giant blob over every mosfet terminal and half the IC's XD

[@anwin85alon]

Pm me can you to install do sola paint of charger`

[@legominimovieproductions]

@@anwin85alon Hi, what do you mean with "do sola paint of charger"?

[@fredflintstone1]

Hmm what current can those R010 resistors take would the act as a fuse???

[@rilosvideos877]

I think they are for current sensing in parallel.

[@fredflintstone1]

@@rilosvideos877 yes but they also carry current hence the heat

[@rilosvideos877]

@@fredflintstone1 Yes, its up to 1 W if each one carries 10 Amps, might be o.k. when soldered to big pads on the board. But its probably close to the limit.

[@ahaveland]

Perhaps it's a crude short circuit protection mechanism so if the voltage drop increases then that tells the buck chip to stop.

[@JulianIlett]

I think it might be that. @ahaveland

[@anwin85alon]

Pm me can you to install do sola paint of charger`

[@AZ-yj2eh]

❤❤

[@jlucasound]

Hi Julian!! 🤩⚡🎛🎚📻🔦🎬👍

[@anwin85alon]

Pm me can you to install do sola paint of charger`

[@kkathulgangadharan2262]

Intresting.need one video of pwm5

[@JulianIlett]

One last video?

[@publicmail2]

Over rated crap 10-15 amps maybe depending on voltages...

[@JulianIlett]

At least this one has the fan included.