🔥 Unleash LIMITLESS Power with Parallel DC-DC Boost Converter!⚡
ฝัง
- เผยแพร่เมื่อ 15 ส.ค. 2023
- Welcome to our channel, where we explore the fascinating world of electronics and DIY projects! In this video, we're diving into the realm of power electronics as we embark on an exciting experiment: combining two 1800W boost converters in parallel to create a single, high-power 3600W boost converter. If you're curious about increasing power output efficiently, you're in for a treat.
Boost converters, also known as step-up converters, are devices that allow us to amplify the voltage of a power source while trading off for current. They're utilized in a wide range of applications, from battery charging to renewable energy systems. Today, we're pushing the boundaries by harnessing the power of parallel connections to achieve greater power output. Before we dive into the experiment, let's lay the foundation. We'll provide a brief overview of boost converters, discussing their basic operation, components, and efficiency factors. Understanding these concepts is essential for grasping how parallel connections can be leveraged to achieve our goal.
Testing:
In this section, we'll explore the advantages and possible disadvantages of parallel connection. By linking two 1800W boost converters in parallel, we can double the current-handling capacity while maintaining a steady voltage. This idea is to know if this concept allows us to handle larger loads without compromising efficiency.
RELEVANT COMPONENTS: Affiliate links
1500W DC to DC Boost Converter: amzn.to/4a0GASD
Power Inductor: amzn.to/44rjPpC
22 Gauge Copper Inductor Wire: amzn.to/3UIrwVx
12V 750W Server Power Supplies: amzn.to/3Wn5KaO
Kaiweets Smart Digital Clamp Meter: amzn.to/3UF4gro
0 to 300V, 0 to 100A, DC Digital Power Monitor: amzn.to/4djjZnd
48V 1000W Water Heater-Test Load: amzn.to/3UKHBtS
Rotating Display: amzn.to/3UGcC1I
6 Gauge High Current Silicone Wires: amzn.to/3UGoRLJ
Big Aluminum Heat Sink: amzn.to/44mTl8K
100A Circuit Breaker: amzn.to/3y95xxS
60A Circuit Breaker: amzn.to/4doyBS6 - วิทยาศาสตร์และเทคโนโลยี
Wonderful work on the project!
Great video! ❤
I am really glad you made the video, lots of lessons learned from!
Glad you enjoyed it!
Thanks for sharing, very systematic and professional video👍
Thanks a lot for the great feedback! We appreciate it.
👍😎
The Innovati0n Lab
Great setup, and thanks for the test. These boost converters have a maximum input current of 40A and output current of 22A. So when driving them with 24V nominal, they are in fact, at best 960W boost converters. When boosting to 32V, the most you will ever get is 704W. You were hoping for ~3000W combined output. Perhaps re-evaluate what success/failure is. Best analogy is this: if you strangle a top fuel dragster with 87 octane unleaded, you cannot call it a failure if the engine explodes, or they cannot achieve a 4 second quarter mile.
You are right, and we have covered the power limits at varying input voltages - in multiple of our videos.
The summary statement I made in the end was not with respect to the 24V that was used to conduct this test. I was pointing out the fact that each converter is rated for 1800W and if the paralleled topology cannot reliably provide at least 3000W (blanket statement), then in the end, it will be a pointless effort....yes, you will expect the full power at a 60V input, we know that for a fact.
These videos build on each other, and I recommend you watch previous ones. Maybe I should make a playlist for all the videos we have made on these converters.
I explained the effect of applied input voltages
In these videos.
👇👇👇
Product Review video:
th-cam.com/video/2KyBwwldqkk/w-d-xo.htmlsi=pvxi_N1KQ1GejY_d
th-cam.com/video/WDWJ77WWkXc/w-d-xo.htmlsi=v-pqGX8uAawTwNP_
th-cam.com/video/ZWcnCeuy_kA/w-d-xo.htmlsi=KR2Cc0IuJVl5PKD2
Thanks again for your feedback. In the next video, I will be sure to make my summary notes clearer to avoid confusion.
I appreciate your wonderful comment.
👍😎
The Innovati0n Lab
😲👍👍👍 maybe make a xmas tree to ramp up the volts 10 at a time and decrease the amps 10 at each boost step..up..
So use 5x24vdc boosteretsrss in and 2 boosters on the 5the step up at 50..60..70..vdc.out..
12..then 24......32...42...52...62...output low amps but the hi 60..70..volts ..out..
Adddd a flow restrictoR in the line in the inputter pwer feed line to divid up the flow of hi amps .. to each boooster...only need 1 or .5 oHms..i hears. Itd probly yy the hi v outputter booster blue first .. cos it got hot first .. cos it was putting out more pwr first
I wants to make 20vdc from 12v to pwr my 18v tools out hear in ta bUshs... oz..🙋♀️🙅♀️😉👍👍👍 ..
From my many testing experiments 30A on the input is the hard limit on those modules, theoretically their wattage rating is only with 60v on their input, so at only 24v in you will be severely limited. The thermal pad and mosfet driving circuitry are their weakest points. I successfully made my own "3000W" boost converter by using one switching circuit from one module and with galvanically isolated mosfet drivers and putting the mosfets directly on a copper CPU cooler it works rather well.
You made a good attempt at it, the output diodes are not necessary but balancing resistors are a must. That is what you see on large audio amplifiers with many switching elements in parallel. So that way tiny differences do not result in drastic current imbalances because the resistance is larger than the very low value of the resistors.
Thanks for the video!
Wonderful!
Thanks a lot for the great feedback. It seems as if we are two like minds, doing similar things in a parallel universe, hahaha.
I have made multiple videos on these converters. In some of my videos, I have talked about the relationship between the current limit, input voltage, efficiency, and the achievable output power.
However, if you take a closer look at the TL494 PWM circuit, you will see that the current limit can only be implemented on the output section of the converter. You are right about the some input current limitation....Again, when you take a closer look at the power inductor design with respect to the saturation current, skin effect calculations on the copper windings used, the converter switching frequency.....you will find the answer and it will all make sense. I am currently working on my own boost converter design that will at least expand that current limit by a lot.
Check out the follow-on video below.
👇👇👇
th-cam.com/video/oPhLpekQ4q4/w-d-xo.htmlsi=uLDv6F36HVjhNR3b
Success wishes on your projects, my friend.
👍😎
-The Innovati0n Lab.
@@theinnovati0nlab782 Yes we did some similar experiments and came to similar conclusions!
Which of your videos did you explain the current limit design? Also by "implemented on the output section of the converter" do you mean of the TL494 chip? For example the chip creates the PWM signal that would normally go to the mosfet, but the current limit is implemented between the TL494 and mosfet. So it blocks the PWM signal from the TL494?
This immediately sounds like a bad way to do it and might explain why the module in undervoltage protection or constant current mode gets very warm when little power is actually allowed to flow.
I saw your video of your, lets say, "expanded" converter. 😁 It was a great start and shows a lot of promise! I can't wait for the new design! But don't rush it, take time to do it right. It is always worth it in the end.
My own version is in the video below: 👇👇Time skipped to the relevant part, as I load tested my upgraded single module converter first.
th-cam.com/video/UbMUy68kWZc/w-d-xo.htmlsi=l8-anxqdidRyFTsG&t=489
I thank you for the wishes, I am always glad to share my experiences and insights! Many hopes to you to have successful projects or great insight in the inevitable failures! 😂
@SuperBrainAK
It's a bit hard to explain over text without looking at the TL494 schematic.
The TL494 has two separate Op Amps that it uses for error correction on the feedback control loop. One monitors the output voltage via a resistor divider - essentially the trim potentiometer. The second monitors the output current by comparing the voltage set across the output shut (V = I*R) with the set reference voltage. When there is a positive or negative difference between the reference voltage and the voltage set by the output current shunt, the TL494 chip will modify the duty cycle of the PWM sent to the power MOSFETs that drives the ppwer inductor. This is to allow the chip to adapt to the change in output load or power demand.
Thus:
Error correction 1 - allows you to adjust the output voltage and, afterwards, tells the chip to maintain the voltage level you have set via a Burst-mode control function....
Error correction 2 - allows you to adjust the output current and also allows the chip to maintain a constant current drive to your output load.
You can see what I mean that it's a bit challenging to explain over text.
I hope this helps.
👍😎
The Innovati0n Lab
I was also thinking about using a capacitor as an alternative is that not a good idea?
nice project man, I hope new projects like this will come
Absolutely!
Thanks a lot for the wonderful feedback.
Please, stay tuned.
👍👍👍
The Innovati0n Lab
I’m glad that you are covering these boost converters right now since I recently received one that I purchased for
Thanks a lot for the feedback.
Make sure you save all your failed converters because I will be making a repair video soon. Also, we are working on a new approach that could use the parts from those converters. Having those could come in handy for you.
Thanks again.
👍😎
The Innovati0n Lab
Ha i don't see the mosfet repair vid ... 😡🤔🤔🤔👍
Thanks that was very helpful. Can you do a video by connecting them in series making the voltage double?
Thanks for the wonderful feedback.
👍👍👍
A lot of people have been asking about this and it has been on my to-do list.
Stay Tuned, my friend. 😊
👍😎
The Innovati0n Lab.
Awesome. Great, friend. Congratulations for sharing the video.
Thanks a lot, friend!
I am glad you found it useful.
👍😎
The Innovati0n Lab.
@@theinnovati0nlab782 thank you again.
I am literally DAYS away from my planned attempt to do the exact same thing! Wow! I didn't expect that result.. I have the exact same model converters. Thank you for this video! I hope you can come up with a way to make it work!! =)
Thanks a lot for the feedback. I didn't mean to ruin the experience for you. But I have been essentially going at this for some time now, and I have grave concerns about the safety of this approach. I know that with careful tuning, maybe at lower output voltages, you could probably find a way to make it kind of work in the interim. But my biggest concern is regarding the safety and reliability of this topology - given how much power we will be expecting the converters to handle at high loads.
I will be working on a synchronous distribution approach next.
But if you try the paralleling approach and it works reliably for you, please be sure to let us know how you did it. This way, we can learn from you as well.
Stay tuned.
Thanks again.
👍😎
The Innovati0n Lab
What happens if you use a Individual powersupply for each converter at the input, does it work, i guess only with a diode that seperates each Power supply from another right?
It only works well if the individual power supplies are ground isolated and the output of your converters are connected in series and not parallel.
I have made a whole series of videos on this comcept.
See below.
👇👇👇
th-cam.com/video/vcY87smODeY/w-d-xo.htmlsi=_CBtlTsdvrMmSoFE
th-cam.com/video/roI7LExA9_Q/w-d-xo.htmlsi=NDYDRfkmSEmSV_HS
th-cam.com/video/MPxp9OyEbHI/w-d-xo.htmlsi=XtkITNFZcWzQOvwy
I hope this helps.
👍😎
The Innovati0n Lab💥
I really enjoy the long form videos with more information and clear description of the idea, problems and solutions related to the project. Keep up the good work and it will be well received! and keep us posted with what more you find!
Thank you
Thanks a lot!
I appreciate the great feedback, and I am glad you found the video useful.
👍👍
The Innovati0n Lab
Nice video!
And two qestions: will this work with two different batterys at 12V? and if so , can you also film or report of Ah eficciecy?
As i would like to use two or more 15-21V Tool battery to power a 48V Inverter
And alternative powered from four 12V battery, charged from different sources via inverters like this.
Thanks a lot for the feedback. We appreciate it. 👍👍 The test was conducted with the idea of using a single input source that is common to both converters.
Honestly, I love your idea of using two different input sources. Actually, with this method, you can possibly connect the individual outputs of the boost converters in series - at least in theory. However, it still needs to be fully tested to see how the control loops of the converters will hold up.
I am sure I will try this concept in the future. However, for now, we need our videos to do well. This way, we will be encouraged to spend more time and resources in making future videos. Please do your part by helping us to share our videos.
👍😎
The Innovati0n Lab.
Great video mate. I like those stands you 3d printed👍
Thanks a lot!
👍👍
We appreciate the amazing feedback.
Glad you liked the video.
👍😎
The Innovati0n Lab.
Thanks for doing this experiment so we don't waste our money on this kind of experiment:)
Thanks for the wonderful feedback.
I am glad.
Excellent educational video. Congratulations. Would it be possible to power the step ups in parallel and use the outputs in series? By making them have 127vdc on the output, do we set a limit for the output current, thus protecting the converters, or does the operating circuit prevent the serial connection on the output? Another question: did the converter burn out due to exceeding the power or for some other reason such as heating or lack of ventilation?
If you looked at the output meter he was outputting 600w at 61v which is 10amps, his input voltage is 12v so 600w/12v = 50amps. the man state not to exceed 40amps input so the experiment didnt fail, he just push too much power through the input. I use those boust converters so have some experience with them, Its a bit misleading about it producing 1800w since to get that power you need at least 45v on the input
Thanks for the feedback and the questions.
Per your questions,
The outputs of these converters can not be connected in series because the ground (-ve) is shared between the input and the output. To accomplish a serialized output, you will need two 'Input to Output' isolated converters. The isolated converter topology essentially uses transformers with isolated primary and secondary windings.....and not inductors as in the non-isolated flyback topology.....sorry about the too much techo babble..Hahaha.
Per your second question:
The converter burned out as a result of an instantaneous overload on one converter. This is the typical reason why I am discouraging this approach because standalone SMPSs are notorious for doing when forced to work in parallel without any synchronous control circuit. The converter feedback loops respond better at lower loads, but at heavier loads, one converter attempts to drive all the load as the other converter's control loop is still trying to respond.......when this happens, there will be a failure.
Thanks again for the thoughtful question.
👍😎
The Innovati0n Lab
Awesome observation on the output parameters. However, your input current calculation is off by 2x factor because we used a 24V LiFePO4 pack to conduct this test...and you can see the input voltage reading on the DMM on the test setup.
The truth is that you are correct about an instantaneous transient overload, and this was what caused the observed failure. This happened at the 62V output test. You are right about the failed converter pulling close to 600W at the point of failure, while the other converter was merely supplying about 56W.
Yeah, too much to troubleshoot, too much to figure out... and in the end, I still won't trust this configuration without some extensive testing and characterisation.
Regarding the failure, see my loop stability explanation to @tenhoduvidas974 above.
Thanks again for the observation
And if we use separate sources, where none of the + or - inputs will be shared between the converters, I believe that this way we can make a series circuit at the output. Even with the input and output being shared in the converters. Is this idea valid?And if we use separate sources, where none of the + or - inputs will be shared between the converters, I believe that this way we can make a series circuit at the output. Even with the input and output being shared in the converters. Is this idea valid?
that could work if Vmax were higher than Vout, but they are nearly equal at 100v
Very nice video like it ❤
Thanks a lot for the great feedback. I am glad you liked the video.
👍👍
The Innovati0n Lab.
??? It it reading possible frequency in the kHz range? I really wana try to apply something like this to my upcoming hho+ design
Yes, the switching frequency of the converters should be around 90 to 100KHz
Whilst not particularly efficient; if you utilised three in parallel, a single device failure will not cause complete power failure. Possible application in unmanned remote location solar power system. Nice video, keep up the good work.
You are absolutely right!
I like the way you think.
👍👍
Actually, what you mentioned is the bread and butter of our 'Limitless Power' concept.
I will be showing this concept in future videos.
Stay tuned, friend!
👍😎
The Innovati0n Lab
Thanks for a great video with loads of learnings... Thanks to your findings, I now question the setup I'm currently putting together; 24VDC LiFePo4 battery system charged from Solar as well as one of these DC-DC boost converters feeding from a 12VDC battery system. I'm not pushing the power you do in this test but still the Solar regulator can in theory push about 580W, but irl no more than 350W, into the battery system and the DC-DC boost converter can in my setup push 480W into the battery system as well. These two feeders to the battery system will never be in sync. Naturally I have schottky diodes protecting any back feed to the DC-DC boost converter. I expect the Victron Smart Solar to be able to deal with this usecase as Victron offer this usecase with their own products, Orion DC-DC chargers.
Am I'm missing anything here or wouldn't my setup be pretty much the same as your test setup? giving that, although I do run much lower voltages and less steep voltage boost (12VDC-24VDC), wouldn't it facing the same risk of failure? Or will the two different power source make the situation different, with perhaps less issue of failure?
Perhaps I have nothing to worry about if I would to replace all critical components with known good parts as suggested by other comments under this video?
Thanks a lot for the wonderful feedback.
👍👍
I didn't fully understand your setup. As you know, looking at a schematic or block diagram makes it easier to understand designs.. but unfortunately, we don't have that luxury on this platform.
However, from your description, it doesn't seem to me that you have two of these converters specifically connected in parallel like we showed in the video. Also, since you are operating at a lower output voltage (24VDC) and less power, I will be less concerned.
As you saw in the video, the system failed when we pushed it a little - at high voltages and above 1000 watts of power.
However, make sure you thoroughly test and keep a close eye on your setup until you gain full confidence in your design.
I hope this helps.
👍😎
The Innovati0n Lab
Is it possible to increase its input voltage to 150v dc ..??
did this blow up from overheat, faulty feedbackloop, or other?
inhave driven these single from 40v supply up to 96vdc, no problems. i want to knownif parallel converters are stable
I have provided some explanation earlier on in the comments threads. When you asynchronously connect two SMPS in parallel, the individual control loops fight for control when a heavy load is applied at the combined output. The converter that wins the control will naturally want to drive all the load and if the loads happened to be greater than the rated output capacity of the converter - then bad things happen......blown Fuses, blown MOSFETs, rectifier diodes..etc.
I hope this helps.
The Innovati0n Lab
I have two of the same models. I can’t get them to load share. They seem to draw amps from one or the other. My thought to get around this is to drive each output into an mppt charge controller and then connect each charge controller to the battery. I believe this will work.
Thanks for the wonderful feedback. I like the way you think.
That will be a good experiment to conduct, and I will never discourage that.
However, the first thought that comes to mind is regarding the feasibility and cost effectiveness of this approach. As you know, reliable MPPT charge controllers are not cheap, plus the fact that we don't know how MPPT controllers will behave when connected in parallel at the common load (your battery). So, in the end, you may just be shifting the same exact problem from the converter to the MPPT controllers.
I hope this helps.
👍😎
The Innovati0n Lab
The simple fact of the matter is that it is very difficult to get good sharing between constant voltage power supplies. The circuits are often made for very good voltage regulation at DC. That means it if one supply in a parallel connected group is delivering just a few millivolts more than the others it will take most of the current.
One way to get moderately good sharing is by using "ballast resistors" between the output of each supply and the common connection. In the video the connecting wires and the measuring boxes would be acting as ballasting to some extent. The drawback to ballasting is that it degrades voltage regulation with load current changes and wastes some power.
I've designed a few MPP controllers. I can't imagine how one would solve your problem. They basically regulate INPUT voltage if the input source can't deliver sufficient power to meet the load requirement.
Even if an MPP circuit did work, it is likely you'd get better overall efficiency with ballast resistors.
@d614gakadoug9
Spoken like someone with a wealth of experience...hahaha.
Thanks again for sharing your wonderful experience with us here!
Greatly appreciated.
This is the definition of constructive feedback to me.
👍😎
The Innovati0n Lab
Good stuff. I was thinking putting a volt meter across the + rails after the diodes to null out any offset caused by monitoring meter inaccuracies and component variations might work.
Hi friend,
Thanks a lot for the feedback. I see what you are saying. However. the issue with paralleling switch-mode power supplies goes beyond managing the offsets - if I understood you correctly.
I appreciate your constructive feedback.
@@theinnovati0nlab782 Thanks for the warm response, BTW, did you try them in parallel, as series connected could cause instability due to them becoming voltage dividers.
Nice test.
Thanks a lot!
👍
The Innovati0n Lab
Where can I find the 3D print file(s) for this stand ?
Our website will be up soon. My goal is to be able to share some of these designs with our audience.
Stay tuned, my friend.
👍😎
The Innovati0n Lab 💥
Thank you for sharing this video, it damage your boost converter but it will save others that will do the same.
Thanks a lot for the feedback. I am glad you found the video useful.
👍😎
The Innovati0n Lab
Sir, can we use this converter for a 36v inverter/UPS from a 12v battery ??? UPS model: Eaton-9SX-1000
Please guide!
Please see my earlier feedback and watch the review video that I recommend to you.
@@theinnovati0nlab782 thank you, Sir
Hello my friend...
Would you feel that the 1800W boost converter could safely output 30A at 50v with a high current 24v input (two of the 750W server supplies in series) for a couple of hours with extra fan cooling?
THANKS MUCH
--dalE
Thanks for another thoughtful question, friend! 30A at 50V will be 1500W at the output and if you account for efficiency losses, that could be >1600W at the input. At an input voltage of 24V, that will be >70A.
That will be too much for these converters.....just my humble opinion.
But you are welcome to give a test and let us know the outcome.
👍😎
The Innovati0n Lab
YES, perhaps >=36v. I just found another way that works, but If I do try this way I'll let you know!
Thanks for your expert help!! @@theinnovati0nlab782
Se puede anular la oscilacion en uno de los modulos y alimetar el gate del mosfet desde el otro modulo. ¿Crees que funcionaria? Para mi que sí!! Saludos desde Santa Fe - Argentina
Awesome suggestion! 👍👍
I thought about doing something like that at some point on these converters instead I tried am interleaved concept...almost the same as what you were recommending.
See the video below
👇👇👇
th-cam.com/video/oPhLpekQ4q4/w-d-xo.html
I am pretty much done playing with these converters and I am currently working g on my own high power design.
However, please feel free to experiment on your ideas and share your results with us here - if you don't mind.
Update video coming soon.
Stay tuned.
👍😎
The Innovati0n Lab💥
I watched this video as I was looking for info before attempting a similar experience...
Please correct me if I'm wrong but I am under the impression that you have connected both batteries in parallel with both converters and then connected both converters in parallel to load... i t did not even occur to me to attempt a setup like this. I plan to connect 1 battery to 1 converter and 1 battery to another converter then connect both output in parallel to load, I'll let you know what happened.
thank you for sharing
cheers
Wonderful!
Please take the time to watch the video before attempting this.
The batteries are connected in series 12V+ 12V = 24V. I believe I covered the test setup in this in the video - as I usually do in all my videos.
Watch the video to the end - for my summary and lessons learned notes. If I am being honest, my opinion is that this is not a great idea for high power applications, and I made this video to discourage the audience from trying a similar concept. But then, I could be wrong, and you could make a major breakthrough....if this is the case, please share your findings with our community here.🙏
The Innovati0n Lab💥
@@theinnovati0nlab782 so if your batteries are in series it means that both converters are connected to the same power source. mY experiment consists to connect each converter to its own(different) battery. as converter#1 connected to battery#1 ans converter#2 connected to battery #2 and then connect the load in parallel . note that only the load here would be connected in parallel. the converter's input wouldn't have any common connections only the output. I'll be taking security measures to ensure safe experiment.
I'll let you know how that went
cheers
@hansplourde
Wonderful!
Hmmmm, I think you may be on to something.👍👍
You are right. I had the same converters pulling power from the same source as that was the premise of the experiment. Keep in my that the converters are not input to output isolated because they share a common ground.
Well, keep me posted on your progress. It would be great if you were able to double your output power using this method.
Success wishes to you, my friend!
The Innovati0n Lab💥
hello. I would like to know if it is possible to use this buck converter to replace a 24v battery to connect to a 24v 220v converter to power electrical devices. Thanks for your help.
Thanks a lot for asking such a thoughtful question. 👍👍👍
Unfortunately, I wouldn't recommend that you attempt using these boost converters as battery replacements. This is mostly because these converters have an output current limit as they can only deliver about 30 to 40A to your load.
I have this helps.
👍😎
The Innovati0n Lab.
@@theinnovati0nlab782 thank you for your reply. Is it possible to use this buck converter to replace a solar panel to power an mppt solar regulator ?
@@user-nu8cb7tv3g
Wonderful question.👍👍👍
We have made a few videos on this subject.
Please check out the videos and let me know if you have any questions.
👇👇👇
th-cam.com/video/6J20RW8_pRM/w-d-xo.html
th-cam.com/video/4RfOrXcBvOY/w-d-xo.html
I hope you find these videos helpful.
👍😎
The Innovati0n Lab💥
I wonder what the outputs limits would be using a 48V 20Ah thru a boost converter? 🤩
Well, the specs of the converter says that it should be able to handle 960W ===> 48 × 20.
However, you might need at a 36V input power to accomplish that.
I hope this helps
👍😎
The Innovati0n Lab.
@@theinnovati0nlab782 Thank you... ;)
I'm going to need 12v to 48v at 15amps output. Less than what this guy asked. How hot do they get? @@theinnovati0nlab782
i paralelled 4 12000 watt bost converters and actually pulled it off using blocking diodes at the outputs of the hot wires of the boost converters. i thought i had it nailed down because i was going to build a turbo booster for my ebike. but itried to repeat the conversion it faied and 3 of the 4 convverters died on me.
Hi friend.
Thanks a lot for sharing your experience. We appreciate it. This is exactly why I mentioned that this approach is not reliable and will not be something I would recommend.
The paralleling approach seems to work at low power demand - less than 500W. But then that becomes pointless, especially if your goal is to try to increase the power output beyond the rated power of a single converter.
At high surges, the converters tend to fight each other for control, and this is when the failures happen.
I have heard people say that they were able to get it to work using low-resistance droop resistors. But then it seems like it only works at low power demands.
👍😎
The Innovati0n Lab.
What are those transistor things on the heatsink for? I missed that part. I wish he explained more instead of the music.
I usually do my best to get to every question.
Lucky for me, this time, this one was directed to the audience.
Hi, good info on all of your videos can you print more of these booster holders and sell them to us?
Thanks for the wonderful feedback!
Absolutely! We are working on our website and this will be made possible soon enough.
Please stay tuned.
We appreciate your support.
👍😎
The Innovati0n Lab 💥
Curious about the other way, buxk converters 48V -> 12V
Thanks for the feedback.
You are right though, I have been giving too much attention to the boost converter side of things.
I will do more content in the buck side of things.
- The Innovati0n Lab💥
Have you shared your 3d print files for your converter stand and meter box? If not will you? Thanks for your videos.
I would love to.
I just have to figure out a way to do it. The youtube platform only supports video uploads....I will be looking into a safe way to do that.
Stay tuned.
👍😎
The Innovati0n Lab.
@@theinnovati0nlab782 you link a share link file as from dropdox
Or gitgrub
@@theinnovati0nlab782 you can put links in description, thingiverse or google drive. Is this meter (in printed box) accurate?
Hello from Canada.thank you for all your videos.I have a two 24 volt 25 ah reongy lifepo4 battery.which can only be run in parallel not series or I’d have my 48 volts.
I am looking for a dc to dc booster to turn my 24 v to 48 v.
from there it feeds into a 48 volt brushless controller.model# 4t48zwsrm-gh04q.
rated current 9a,maximum current 18+1a,speed limit 1-4.2v,brake input:low level,low voltage protection 40v+0.5v.
From there feeds into bafang 48 v 500watt hub motor.
Can you or your viewers please recommend a boost converter to Change battery from 24v to 48 v and that will work safely with the controller and motor.
I am doing my best to figure it out myself but my brain and math/numbers don’t work so well together.
My fear is damaging my controller/motor using the wrong converter or it not working properly.
I have a bpm imports fat tire trike 500 watt 48 volt.
You can not find these trikes in 500 watt because bpm Canada never got off the ground due to Covid.
The USA versions come in 750watt and up.
Thank you again for all your great videos.
Happy Monday to ya.
Hi friend,
Thanks a lot for watching our videos and for leaving great feedback.
Regarding converter recommendations, there are lots of dc to dc boost converters out there that could meet your needs. It really depends on your budget.
The converters shown in this video in theory should be able to individually give you the power you need at a 48V output with an input voltage of 24V. The rated current limit is about 30A, so at an input power of 24V, you should be expecting about 720W to be processed at the input of each converter.
However, the output power has to really account for the efficiency of the converter system at 24V input.
Assuming a worse-case efficiency of 80% (it should really be better than this),
The available delivered power at the output = 720 X 0.8 = 576W.
Hence, if my theoretical calculations are correct, and your motor doesn't care too much about ripple noise levels,.....then one of these $30 converters should give you what you are looking for.
VERY IMPORTANT NOTES:
1. Make sure that the UVP is set to below 18V.
2. Ensure that the constant current potentiometer is set to the Max output current position.
I will recommend you take some time to watch all the videos below and I am confident that they will answer any further questions that you may have regarding these converters. Just an FYI, I am not affiliated with either the vendors or the manufacturer of these converters. My reviews are based on my personal ho est opinion.
th-cam.com/video/2KyBwwldqkk/w-d-xo.htmlsi=lckQUYefk8TwmDYh
th-cam.com/video/t8Aaz4iQeXY/w-d-xo.htmlsi=B3kS3kXSpnG3AoB0
th-cam.com/video/WDWJ77WWkXc/w-d-xo.htmlsi=JLzxKk_AIR57YMFN
th-cam.com/video/0Ct_sgbAjU4/w-d-xo.htmlsi=pXKiSgV4C5Xa8BU7
I hope this helps.
And if it does, please support our channel by sharing our videos.😂
👍😎
The Innovati0n Lab
Thank you very much friend.
I appreciate you taking the time to reply.
I will watch the recommended videos you sent.
I just seen the new booster on your shorts.
I am going to look for led screen setting buttons booster since I do not have electrical devices that you have to set or read the booster.
I am going to try to find a really good booster with screen and buttons,the one on your shorts gets very mixed reviews on Amazon.
If I can’t find a better option I will order the one from your shorts.
Again thank you very much for your reply and your videos.
I will like button them all.
I would share them but I don’t have social media.
Happy Monday to you good sir.
How did you make out with your experiment. Im looking at same idea with my trike.
Why didn’t you use a pair of steering diodes at the outputs of the boost converters to keep the voltage from one converter feeding back into the other converter? Back when I had a single trace scope, and I wanted to cheat and make it a dual trace. I used a pair of steering diode, and I brought the two diets together deposit side and the ran the positive positive input of each source into each diode. this is also also worked for me in isolating, power supplies, and preventing them from seeing each other and causing such a problem
Thanks for the comment.
However, you just described the exact setup that was shown in the video. I found that sometimes people comment on a video by just looking at the thumbnail image, and I honestly don't encourage this. Were you able to actually watch this video?
I always encourage my audience to leave us feedback so we can learn from the audience as well, but the only thing I ask is that we watch the video before commenting. This is a reasonable ask, yes?
This saves me a lot of time on having to re-explain things that I already showed in the video.
Respectfully,
The Innovati0n Lab💥
www.theinnovati0nlab.com
@@theinnovati0nlab782 sorry I did not catch the steering diodes I for my project as well as I need to decrease the charge time without each DC to DC converter affecting each one you had so much going on in your video that I must’ve missed them. Please give me a part number on the Steering diodes you used this will save me a lot of time as I do not wish to burn up components and then repair them
@@theinnovati0nlab782 and it was because of your video you tipped me off about the grounds and I checked my DC to DC up converters to see if the ground was common between input and output and found my up converters to have the same situation. Thank you. Your video was very informative as far as that went. You have saved me from burning out expensive parts.
@@TheCommonDanger
It makes me very happy to hear that. I am glad you found some good information in the video.
👍👍👍
The Innovati0n Lab💥
So what did you use as far as the semi conductors isolate the outputs and bring them together that was one more question I had that was not answered for a brief moment after re-watching your video I spotted what appeared to be too show me conductors bolted to a heat sink after the far right your video did you failed to mention or point to them in your video? Or maybe you just didn’t bring enough attention to them “going from the output for safety purposes you go to a circuit breaker for safety purposes“ I’ve now watched it three times in this section at eight minutes and 13 seconds and you made no mention of the part number or type of conductor used to isolate the two outputs and bring them together
Good video, I think the problem is the output current is not balanced, in your video one shows more amps than the other, if one has more current than the other most of the current will surge to the higher current output/path of lest resistance.
I really want this to work so that I can connect two boost converters to a 12v server supply and charge lifepo4 at 70-80amps 14.6v
Thanks a lot for the wonderful feedback.
You are right about the current balancing observation. The first test was such a low power output that I didn't put in much effort in tuning the outputs. The subsequent tests were a bit more balanced after some tuning. Maybe I will give it another try in the future.
But no promises for now. This type of test video takes lots of time and effort to make.😊
Stay tuned.
Thanks again.
👍😎
The Innovati0n Lab
If is nearly impossible to get good constant-voltage power supplies to share current well. "Ballasting" by adding resistance between each supply and the common connection point helps, but it degrades voltage regulation and efficiency.
Power supplies operated in constant current mode can be used in parallel, each contributing current according to their setpoint.
If you need to limit charging current to your battery anyway, this can work quite well. The supplies can run in constant current mode until the batteries are nearly charged, then cross over to constant voltage mode. One supply may deliver most of the current in CV mode, but that may be OK if the current isn't too high. If one supply is always the one that takes most of the load it will "wear out" faster (capacitor ripple current handling is usually the limiting factor).
Hello my wonderful friend. It was not clear what had failed and why it had failed. Individually the buck converter should have outputted more than when in parallel so why did it fail with two. Is one feeding power back to the other. I noticed you stepped the voltage up on both of them then ran the load, but what if you had stepped each one individually while concurrently powering the load so eg one would be at 32V while the other was increased to 42V. Would that scenario have failed? I assume a diode would fix the problem. I think of it as the same as if I used the buck converter to charge a battery. The battery (equivalent to a 2nd converter) could have a higher voltage than the first buck converter (and that would work OK without a diode), and if the output of the buck was higher than the first buck converter (like charging a battery), it should work. You might want to talk about blocking vs bypass diodes in the same video.
Liked: My wonderful friend!😊👍👍
I hear what you are saying, and seeing all these responses now makes me realize that I could have probably taken a couple more minutes to explain a bit more regarding my observations.
It's just that this very video was a bit tougher to make, plus I try to keep the videos right around 15 minutes long.....people these days have super short attention span!
The behavior that led to these failures is typical of SMPSs - when asynchronously connected in parallel. At heavier loads, the converter with a faster loop response attempts to deliver all the output power demanded by the load. This is a transient behavior that usually ends up with a mildly catastrophic outcome.
I am going to make an unedited explainer video this weekend.
Stay tuned.
Thanks again, my wonderful friend!
👍😎
The Innovati0n Lab
It only works if the boost converter are in synch , or only one TL494 drives the tow gates of the Mosfest on the tow bord, make the connection just right and it will work
Absolutely,
This is the other idea I talked about in the video.
👍😎
Stay tuned.
The Innovati0n Lab
can i get stl files for those stands for 1500w converters?
Yes!
I know I have been saying this, but TH-cam does not support file sharing, and I am working on a way to make this possible very soon. I won't be able to share some things that I want to keep as proprietary, but I can share these.
👍😎
The Innovati0n Lab 💥
How can I order, buy the converter. And what is the maximum wattage I can get I can get?
@richardleslie4306
I may be selling some of these on my website soon.
But for now, you can get them from Amazon and ebay.
the question however remains ..... would the converter have failed anyway if used stand alone for the same power output.....
i believe that IF the converters don't have a diode in the output, put a nice high amp schottky version on both units and something like a small resistor to even out differences in output like 0.01 ohm or so .... won't screw up the output voltage too much but evens out about 0.2V difference ;)
Great suggestion.👍👍👍
The straight and simple answer is that SMPS don't like to be connected in parallel unless they are designed for synchronous sharing.
The control loops fight each other once an instantaneous high load is applied. Your setup could work reasonably below a couple hundred watts.....the true test is when you slam the system with a 1500W load step.....that is when the majestic blue smoke comes out to grace the occasion....to say thank you to playing...hahaha.
Anyway, I am done with that project. Please feel free to order a few of these converters and try out your idea.
Let me know how it goes.
My prediction is that 0.01 ohms would not have your anticipated effect.....that's basically wire resistance for you.....but becareful to resist the temptation of increasing the droop resistance values because the amount of dissipation you will see on those will shock you.....please do the math. One single decimal increase to 0.1 ohms will cost you a whooping 180W - if your goal it to drive the system to the Max rated of 30A.
I hope this helps
👍😎
The Innovati0n Lab💥
www.theinnovati0nlab.com
U need dioode each output for counter reverse feedback
Thar is correct.
You need to adjust the same amount of current going thru each converter or one will be pulling all the weight of the load and over heat. Need to have them balanced to do the appropriate test here. You notice the amperage pull on the one device was greater then the recommended current draw because they were not balanced between the two. The fault was not with the parallel converters it was the operator's fault doing the test circuit. You didn't stay in parameters of each devices capabilities of it maximum output power and current ratings. That is why each has its own adjusting potentiometer to regulate the current so the load can be evenly matched between the two converters. Re-think your theory and concept here. You were the one who failed during this test my brother. I like the scope on the circuit too, seeing that ripple current or noisy DC the converter is making I might need to build my own filtering circuits to bring that ripple under control for my RF device or it might lead to a humming noise being picking up on the receivers hearing my transmittions.
Thanks for watching our videos and leaving feedback. We appreciate it. 👍👍👍
However, your comments were a bit presumptive. The current limits of both converters we carefully dialed in before the test loads were applied.
The test was not perfect, and I have identified some areas of improvement - not exactly the things you mentioned because they were already implemented in the tests shown in the video. The load sharing you saw on the tests before the 62V (failed) test didn't not happen by accident......it came from a meticulous effort of adjusting the output current of both boost converters.
The failure came of a much deeper level (converter loop response) that have talked about on the subsequent video - which you have also commented on. This is a known behavior of asynchronously paralleled switch-mode power supplies.
There is a great deal of effort (research, calculations, testing, etc) that goes into conducting these tests and now capturing them on video.....So we shouldn't make blanket or degrading statements for things that we do not understand. This is what questions are meant for.
All the same, thanks again, and we appreciate your honest opinion.
The innovati0n lab.
are you driving a LDMOS device? do you use a PI filter circuit for the ripple?
I have designed a lot of switchmode power supplies for industrial applications. I take one look at photos of those low-cost supplies sold to the hobby/DIY market and know beyond doubt that their claims for power handling are, in general, grossly exaggerated.
The modules in the video do look better than many. The core material is possibly "sendust" (I'm basing that on the fact they are painted black; Magnetics Inc's "Kool Mu" cores are a high-quality example) which is considerably better than powdered iron. There are multiple input and output capacitors which improves ripple current handling (boost converters cause high ripple current in the output caps but aren't so bad for the input caps). Provided those are very high quality capacitors they'll likely hold up OK, but I suspect they are still being overworked and ripple current is too high for good long-term reliability.
Connecting constant-voltage power supplies in parallel is usually problematic. If the gain of the error amplifiers is high (so output regulation is good) it can be very difficult to get good sharing. One supply will tend to take most of the load current. Diodes at the outputs don't help in any direct way. One thing that may be helping in the test setup is the "ballasting" created by the connecting wires and the measuring boxes. Adding some resistance between the output of each supply and the common connection helps with sharing but at the cost of lower efficiency.
If I were testing these modules I'd collect data on output voltage versus output current to get an idea of how well they regulate with varying load. The poorer the regulation, the better they are likely to share.
Wow!!
What a wonderful feedback!
Thank you so much.
👍👍👍
It seems like you have a wealth of experience on the subject, and I appreciate your input/ suggestions.
As a matter of fact, I was not surprised with the outcome of the test shown in the video. I had to do it because I felt it would be a good lesson learned for my audience - who kept on asking about SMPS paralleling as a means to meet very power demands. Well, this was because of their high power and charging current needs. However, it seems to me that what you were describing is the droop sharing method with some low value resistors connected to the output of each boost converter. The issue I had with that concept in the past was that it works well at low power demands - usually below 500W shared between two or more individual converters. My audience here is looking for converter systems capable of delivering >100A at an output voltages >100V..hahaha.
I don't think we can efficiently manage/balance this type of power demand via droop sharing - without some more advanced distributive synchronous sharing design approach.
What do you think?
This was essentially why I am working on what we call the "Limitless Power" design concept. Essentially, multiple converter branches are driven by the same control/feedback loop. The 'Limitless' in the name basically points to the fact that the design concept is a system that can grow limitlessly to an nth number of converters - with respect to the designer's power need. We fully recognize that limitless power is a pipe dream - at least on a single board.
I have shown this concept in a later video.
👇👇👇
th-cam.com/video/oPhLpekQ4q4/w-d-xo.htmlsi=uf4WftIkxJyYq0Tr
Let me know what you think, and thanks again for such constructive and inspiring feedback!
👍😎
The Innovati0n Lab
@@theinnovati0nlab782
I'm pleased to try to help. You did a very nice, professional job of your video!
I've never heard it called "droop sharing" before, but that's a pretty good term.
I agree. At high current or more accurately, over a wide current range, it is not very practical. You _might_ take the position that you want good sharing _only_ at high current and not worry about good sharing at low current. In that case you could use the lowest value resistors that would give you the sharing accuracy you want at high current, but that wouldn't help much, if at all, at low current.
At one time Unitrode, now part of Texas Instruments, made an IC specifically to manage sharing of power supplies. I've never used it and don't remember any details. I suspect it probably isn't even available anymore.
I can think of a few ways to actively adjust each supply but it would require modifying the supplies and adding extra circuitry. One of the problems with active circuitry or a microcontroller to do something like making a small change in the reference voltage or DC feedback network on each board is the common ground. You don't want power currents finding their way into signal paths. Then you have the problem of different supplies being made differently, so a desirable "universal" solution would probably be difficult.
Your viewers are lucky to have someone with the knowledge, ability and willingness to make videos to answer their questions. Best wishes!
@d614gakadoug9
You are absolutely right.
Power sharing for converters is a tricky business that requires complex and carefully dialed-in control circuitry. Some of the more high-end manufacturers use master-slave synchronous buses. I believe it's called a democratic bus or sharing method. This is a control protocol where one of the converters in the paralleled system takes over at power-ON, and essentially shares its PWM control signal and switching frequency (in a broad spectrum design) to drive the applied load.
Thanks again!
It's always great to talk to like-minded people in here.
👍😎❤️
-The Innovati0n Lab
This isn't so much a reply as it is an invitation to critique my thoughts and let me know if I understand the concepts...
Yeah you can see at the moment of failure, one module is supplying 598W and the other one 56.6W. At 23V input (I saw the input voltage at the 52V test) the input current is going to be around 30A for the high module, maybe more.
It's also important to match the parameters of the experiment with the equipment... The max input current for those modules is stated as 40A. If you want to see if the total power output can get to 3600W, you have to supply each module with 1800W... at 40A that means you need at least 45V input, probably more like 52V to account for conversion losses.
You also need an adequate load... there was no problem at 32V despite one module supplying almost twice the power of the other (2/3) of the total, because the resistance was so high that neither module, even on its own, could push enough current through the circuit to get the input current to a dangerous level.
At 42V you had near perfect matching of voltage and nearly perfect load sharing... again, 42V isn't enough to push enough current through the resistance to kill either module even if it were alone.
At 52V, again, nearly perfect voltage match and very good load sharing. With a total current of about 18.5A, one module on its own would have drawn at least 18.5x(52/24) A, which would likely have killed it, but the voltage match meant amazingly even load sharing. It is, however, into the danger zone where a lot of heat is produced and heating induced changes in the performance of either module could cause a change in voltage in one unit... if that change is upwards, then that module assumes more load, producing more heat, causing further voltage increase and so on in a rapidly escalating positive feedback loop.
At 62V, the instant the breaker was closed, there was a 0.5V difference between them, resulting in the higher V module assuming 90% of the load. This occurred, I would say, because boosting from 24V (probably less given that 52V drooped the supply to 23V) to 62V is right at the limit of the modules. They were both 62V in open circuit, and the slightly better module maintained that voltage, supplying 600W, from 23V, giving an input current that, while within the specs, is probably somewhat wishful thinking.
Would have been interested to see if you had flicked the breaker, readjusted the voltage on the lower volt module to, say, 62.1V, tried it again, could it have been made to work? I'm certain that a higher supply voltage would have saved the system too. I think it's possible to parallel two modules and get greater output than one module alone, but trying to get double the output? You're pushing components to the limit AND basically using each one to try and overpower the other... think of a tug of war, quite safe on the ground... then do it on a tightrope 200m up.
how long do you test
The test shown in this video was a short duration test of about 5 to 10 minutes.
The vastly different output current levels are an indication of a serious problem.
I see your point, and that was why I changed setup midway through the testing - this was just a measurement error on the power monitors because they were not designed be connected it series.
But then, I made this video to show the audience that this approach does not work as a strategy for boosting converter output power.
Thanks a lot!
👍😎
The Innovati0n Lab💥
www.theinnovati0nlab.com
I used those modules, to make them reliable replace all big capacitors and the power fets and diodes by parts you obtain from a reliable distributor like Mouser. The original components are underrated and will fail under load.
Thanks a lot for such a wonderful suggestion. I will definitely this approach.
This is really a great idea.
👍😎
- The Innovati0n Lab.
@@theinnovati0nlab782 I have been using the 757-TK100A10N1S4X available @Mouser it's an N-Fet 100V 100A with a RdsOn of 3.1 mOhms. The low rdsOn helps efficiency of these modules, less heat also.
I did a 120 ac 4 amp 13.8v dc and a 120 ac 10 amp 15v dc with 2 - 20 amp diodes together to charge a battery! It kind of worked! The solder started to melt in the diodes even throw they were twice a big!
Wonderful!
Thanks for the feedback.
I am glad it works for you. However, my concern is not that it won't work at all, I am only concerned that it is not safe, especially at high loads and higher voltages.
@@theinnovati0nlab782 I disconnected the 15v about 10 minutes after it was to hot.
Im not sure i get the concept of this! surely if you step up the voltage you will drain the batteries even faster? why would you not just use an inverter to power heavier loads briefly ?
Thanks a lot for the wonderful feedback.
👍👍👍
The concept has a single very focused goal which is to provide a variable high voltage DC source.
The video was made as a response to the high voltage needs of our audience. The isolated battery method was shown as one of the possible ways of accomplishing the function...but not the most feasible.
A power inverter will only provide a fixed AC output (120/220/240Vac...etc) and not the needed variable high voltage DC. This system offers a wide output voltage range from 24V to about 190V DC.
As always, we make sure to remind the audience that high voltage DC is lethal and much care and experience is required for handling such a system.
👍😎
- The Innovati0n Lab 💥
Knowing the current in each converter is measured by a shunt in the negative line, if you connect the two negative input and the two negative output lines, the two shunts are interconnected in parallel... each current measure is influenced by the other converter current and the converter "sense" a very wrong current. As the current limit, avoiding to burn the mosfet, is wrong, there is no limit anymore and the system drives the modfet to over current and ... boom..
In all tests at different voltages, the bottom converter gives more current than the upper one. (@32V we had 4.31A and 7.48A respectively. almost double current for the lower converter... this is due to the voltmeter precision of the measuring tools. Maybe better to adjust with an external Vm (the fluke?) on each converter output. But this will probably not balance the current due to the differences of resistances in the different circuits. Better to adjust one converter voltage and than adjust the second one to balance the currents... just my two sents....
Thanks a lot for such constructive feedback!👍👍👍 I greatly appreciate it.
I think we can both agree that this is not the best approach. Honestly, I have always known that this was not the best idea, but I made this video to deter our audience from trying this approach....it is simply wasteful.
SMPSs, in general, do not operate well in a paralleled configuration - unless they are designed for synchronous sharing....that's it.
You are right about the disparities in the current sense shunt I*R voltage translations to the internal comparators/Error amps of the TL494 PWM controllers. However, this essentially boils down to a tug of war between the two independent control loops of each of the converters.
Paralleling works better with batteries and capacitors because they are simply just obeying the fundermental laws of physics....Ohms law, Kirchhoff's laws, superposition....etc.
Switch-mode power supplies, on the other hand, are designed to circumvent those laws - depending on how you look at it.
Btw, please check out our latest video as it sheds more light on this very subject. We found a hack that you could use to double the power output of these converters...at even better efficiencies.
👇👇👇
th-cam.com/video/_cG3smKwTzE/w-d-xo.htmlsi=E2xBW6ZOPdLO10Dm
I hope this helps.
The Innovati0n Lab💥
Add a 1ohm resistor to each input lines..
Leave diodes out .
Rf man 12 24 in 50vdc out ..2000..1500 watts.. out of 4x 1800 w booostesresrrs
1500w watt i think is max he uses ..4radio eqip.
Thanks a lot for the suggestion. I know of this approach, and it is called the "Droop" approach. However, I have only used this for low power applications because at high loads, the droop resitors will dissipate a lot of power - per ohms law.
Is this something you have tried for high power applications, 2000W and above?
Example:
Consider a situation at high power demand where each of the 1-Ohm resistors is sourcing >30A.
Per Ohms law;
Total Power Dissipation = (30 X 30) + (30 X 30) = 1800W dissipated by the Droop resistors alone.
On the contrary, the recommendation is usually to use lower resistance values for droop applications - simply because of the math above, and again, i would only use it for a low poerr application (
Isn't there any buck inverter to get 2000 watts?
I see your point.
Of course there is. But they will be super expensive. My channel is about helping people find innovative, low-cost alternatives.
I am sure that if some alhas $300 to $2000 to spend, they will find some top-notch converters with all the bells and whistles. These one cost only $30 with 'promises' of 1800W.
Thanks again for the thoughtful question.
👍😎
The Innovati0n Lab.
@@theinnovati0nlab782 thank you very so much for replying.
If you increase the number of MOSFETs, can you increase the Ampere capacity?
Thanks a lot for the great feedback! I like the way you think. 👍👍
Your thought on adding more MOSFETs is on the right track. However, it will only help if you are using low power MOSFETs. Also, the inductor saturation current will be another bottleneck to consider. If the inductor saturation is low, there will be no need to add more MOSFETs....Basically.
For example, if the inductor saturation current is 30A, then paralleling MOSFETs that can drive 60A wouldn't really help your power output. It might be good for redundancy/reliability, but that's it.
I hope this helps.
👍😎
The Innovati0n Lab
@@theinnovati0nlab782 Thank you very much for sharing your knowledge 👍👍👍👍👍🙏❤️
what about the input voltage? definitely needs a big current.
I have paralleled 2 power supplies with the same capacity, the goal is only to change the single voltage to be symmetrical, namely + ct _, but the supply current from the source becomes very large, so the breaker trips
@@theinnovati0nlab782
sorry just a discussion@@theinnovati0nlab782
you just need to put high current diode at the output of each boost converter and then connect them together.
Thanks for the feedback.
However, your suggestion was already implemented on the test setup shown in the video, and I always take the time to explain my test setup in great detail. Its ok olif you skipped through some sections. 👍👍
I hope that you will be able to watch the entire video.
Let me know if you have any questions after watching it.
👍😎
- The Innovati0n Lab.
@@theinnovati0nlab782 could you give me the pointer to time stamp where you put a diode at each output of boost converter and connected the the other side of the diodes together?
--thanks
@@aduedc It's shown in the circuit diagram near the beginning, then not explicitly shown in the physical setup. There is a heat sink with some large diodes attached. Unclear if these are required or why one failed even with the diodes present.
It's hard to replace the factory smoke, you're not supposed to let it out.
Hahaha!
That's a good one.
I like it.
If there was a watt meter after the diodes, we could see the effect of paralleling on efficiency. the interesting thing is that the one with the higher output voltage is broken, I would expect the reverse to happen. it seems that there was no deterioration due to reverse feeding. I think that even if the deteriorated one was alone, it would not be able to bear such a large resistive load and would deteriorate. Great work thanks!
Thanks for the wonderful feedback! I appreciate your comments. I will probably make an unedited explainer to go over this failure this weekend.
👍😎
The Innovati0n Lab
The higher voltage module assumes most of the load, which is why it failed. The input voltage was about 22V (at the 52V stage, the voltage from the batteries drooped to 23V, so I think it's safe to assume it droop further at 62V).
The higher voltage module was supplying 9.66A, which would mean the input current is 9.66x (62/24) = 25A x (1/.8) = 31A. The last factor is 80% efficiency which is about right for that extreme level of boost.
The specs are 40A input max, but usually that's with increased heatsink capacity.
Не верится что на хлипком радиаторе и на одном транзисторе можно вытянуть 1800 ватт. Это можно если ключь будет полностью открываться и закрываться. Плюс сопротивление канала должно быть минимальным
Please watch the video below.
Summary: I have been able to show that this is possible and it has been tested to 1500W - using a 36V input.
In my next video - I will be pushing our modified converters to a goal of 3000W with an input voltage of 60V.
Is this possible?
Can we pull this off?
Stay tuned.
BOOSTED CONVERTER:
👇👇👇
th-cam.com/video/_cG3smKwTzE/w-d-xo.htmlsi=-Ig506ThdRR2p2nt
👍😎
The Innovati0n Lab💥
❤❤❤
Thanks a lot!
We appreciate all your inspiring feedback!
👍❤️❤️
The Innovati0n Lab
Always here to give the first comments ❤❤
Hi friend!
Thank you so much for all your wonderful support. We greatly appreciate it.
👍👍👍
This video took quite some time and lots of effort to make. I am glad you liked it.
👍😎
Thanks again!
-The Innovati0n Lab.
u need add Diode to mix 2 DC output
Thanks a lot for the feedback. However, please watch the video before leaving a comment.
What you are suggesting was already implemented in the test setup that was shown in this video.
Again, watch the video and let me know if you have any questions.
- The Innovati0n Lab.
How many consant current give in outpit
Each converter is rated for a maximum current of 40A. The steady state limit should really be about 30A.
@@theinnovati0nlab782 how much output as parallel
2 Diod Conecter
Yes.
I'm awiating my 1800w boost from ali at the moment, I like the holder that you printed, one inprovement I would suggest which I will do with mine is to move the cooling fan to blow down the heatsink instead of the way that they have mounted it. I will be replacing the fan for a pc psu fan which is alot quieter . the inductor gets very hot if you push more than 15 amp through it. by the way when the converter failed you were pulling 50 amps from the battery which exceed the 40 amps fuse so I woudlnt agree with you that its a fail
Thanks a lot for the great feedback. Wonderful observations, I appreciate it. The idea was for the overall current draw to be shared between the converters. However, the issue is that sometimes on converter always wants to take all the loaf before you have had the chance to make the needed adjustments. This behavior is typical of SMPSs, and that is why the paralleled configuration is always tough for SMPSs.
Thanks again for the wonderful feedback.
👍😎
The Innovati0n Lab
The Person Who Recorded the Video Already Changed the Module Fuses to 60 Amperes
@@koksaltasci2607 60 amps blade fuse are normally light-blue colour, the fuse in the booster was yellow, unless the person who recorded the video made his own using the yellow housing, blade fuses are colour coded as a standard
@@danny323dee This is the thing I can't go on saying. This Module Never Provides 60 Amperes From Its Output Channel. Therefore, 60 Ampere Fuse is Unnecessary and Very Ridiculous.
@@danny323dee Also, Blue Color Fuses Must Be 20 Amperes
hiya with this type of boost converter for what you are doing you would need to be able to adjust the input voltage and input current ,or you will quickly over run the max input current and bang ,you killed them,these units must have input and output monitoring
Great idea.
Thanks.
Make both negative out put common and take both the postive out to two ends of heater one end of heater connect to one end of heater and other end of out put to other end of heater out put will be double
Common of both DC DC booster is not connected to heater only parallel booster of two booster out put connected to heater
Two phase DC out put
Thanks for the suggestion.
👍👍
Take output has two phase voltage will be double
Thanks for the feedback.
Please can you elaborate?
Testing starts at 9:30 .......
Thanks a lot for the feedback. But I am not clear on what you are trying to point out.
Let us know if you have any recommendations on how we can make the videos better.
👍😎
The Innovati0n Lab💥
Sir Peralal connection use mosfet number
These Boost converters use HY3810 MOSFETs. The earlier versions use HY3912. You find these on Ebay.
I hope this helps.
Success wishes on your projects
👍😎
The Innovati0n Lab
Connect both DC DC booster out put in series
Connected in series voltage output will be double
@HarvinderSingh-vs2yl it will not work because the converter input and output ground are not isolated.
your volt in parallel each test 32+32=64volt
No, it is more like 32V in parallel with 32V = 32V. When you connect two dc sources with equal voltages in parallel, the resultant voltage stays the same.
Yea man when it takes 150 amp from the battery blows it so next time when you do that use a fuse from the battery at least 40 amp fuse
@maxbridget3614
There is an entire segment of this video titled "Safety Review" in anticipation of concerns/comments of this nature.
I am hoping that you didn't skip that segment of the video?
Thanks regardless.
-The Innovati0n Lab
I am sorry but you dont understand you went way over the max wattage ,you cant output 1800 watt converting 12 volt to 42 or 52 the converter can only do 300 watts or so at this convertion ratio
Thanks a lot for the wonderful engagement.
You shouldn't be sorry for providing good feedback and constructive feedback.
However, I would like to point out the following;
- Nominal Test Input voltage used for all the tests was 24 - 27V. We had 2X 12V lithium phosphate battery packs connected in series.
As for the overall output load. The concept was to use a 3.3 Ohms fixed resistance load. Hence, if you do the match (V^2/R) the overall power output expected from the system was around 1200W....at an output of 62V.
Which should translate to about 1400W at the input - accounting for efficiency losses. The goal of the test was never really to drive the system to full power. The goal was to test the power distribution/sharing between the two boost converters.
You can double-check my math and let me know if you think I am wrong.
Again, you should never apologize for pointing out where you think I made an error - no one is above mistakes.
I hope this helps clarify things.
Thanks again!
👍😎
The Innovati0n Lab
remember amps x volts equals the real watts ,those metters arnt giving you a real reading ,with 12 to 42 at 350 watts will be 25-30 amps on the input
You are making a logical point. However, how do you know that those meters are inaccurate?
its not that the meters are inacurate ,they are most likely reading the right figures ,but its the way the converters work ,if you boost from 12 to 52 the convertercan onlty outputv 3 hundred of so watts before the input current reaches 25-30 amps @@theinnovati0nlab782
52 volts x 25 amps equals 1300 watts,but your output would only say 3 hundred watts or so,its the way the maths is worked out
so having meters on the input will allways let you know the true ststus of the unit
and current limiting and voltage adjustment on the input will let you set ,your desired output without damaging anything
exactly what i said would happen 9 amps on the output would have been pulling 30 on input
if you arnt measuring your input with there converters you will allways have trouble ,you allways have to input the highest voltage possible that you can ,so the converter is not stressed
and you can never get equal load sharing if you dont know the input
See my comment above.
I hope it makes sense to you.
Isn't there a converter that has a higher output power than this type on the market
??
I am sure there are, of course, if you have hundreds of dollars budgeted for your hobby/DIY projects.
Our vision is to find low-cost innovative alternatives to super expensive devices that most people around the world can never afford.
A professionally built and tested 3000W variable output voltage, constant current DC-DC boost converter that offers great reliability will probably run you upwards of $500 or more - if you can actually find it. These converters only go for about $30 each....I think that they are great for the price.
Thanks again.
I hope this helps.
👍😎
The Innovati0n Lab
I have recently seen a similar spec’d but much more robust boost converter with a higher power rating on Ali and eBay. The price is just over $100 if I remember correctly but I’ll be watching in hopes that the price will go down in the future because they look nice!
@@maukaman any links to the site
@@maukaman
Wonderful,
Please share the link of you don't mind.
Thanks again.
Good demonstration but video too long
That video is barely 15 minutes, and that is too long? Well, you should check out our new format.
Some of my videos are meant to focus on capturing all the relevant details. This way, we can use them as educational references in the future. You don't have to watch all the sections in one go.
👍😎
The Innovati0n Lab💥
stop saying lets get started. just start!
Sounds good, thanks.
I Dont understand your conclusion. There are thousands of models out there. How do you wipe them all away and give that conclusion based on no examination and only one test with one model?
Imagine i bought a lightbulb 💡 and after 5min it went blink and stopped working... Then i concluded that lightbulbs are not a great idea.
You didn't test how they affect eachother either.. like.. if you set one to 32v 10a.. and the other.. test them individually.. then what do they do when they are turned on simultaneously.. do they then give 20a combined? Does it drop to 14a? Is it split even?
Think the video was a bit.. short.. and needs a bit more.. enthusiasm and curiosity.
Thank you for the effort and time.
Thanks a lot for such wonderful feedback.
👍👍
I see all your points.
However, my conclusion was not that this concept is an absolute impossibility. I focused on the safety of the topology.
Secondly, the typical behavior of standalone SMPSs is that they do not like to work well together when connected asynchronously. I made this video to push the boundaries a bit, and it really took a long time to make. All those tests you highlighted will take a team of high payed engineers to complete, and we do not have the resources to do that. Maybe someday we will. But for now, I mostly work alone.
Also, using light bulbs in your analogy to describe a test that involved non-isolated asynchronous flyback converters is simply not fair.
If you put diesel in a gas engine muilple times - even in different makes/models.... the result/outcome will end up being the same - bad.
There are external circuits you could add to make the converters work better together. However, you will be dealing with tradeoffs like cost, efficiency, and lots and lots of development time and testing.
For what it's worth, I didn't give up....I am stubborn that way. 😊
I have another approach that I am convincing will be more effective.
Stay tuned.
👍😎
The Innovati0n Lab
@@theinnovati0nlab782
It took me 30sec and im not a team of highly skilled engineers :). That said i do use step down not step up
Looking forward to your next video. If nothing else they look fun :)
* This PCB no good , if you need to reduce the current it will be drop very fast , actually the current will increase or decrease slowly , but this one can not , no good ,
It seems to me that you have a faulty converter with a bad potentiometer.
I have never experienced this with any of my converters.
However, thanks for sharing your experience.
👍😎
The Innovati0n Lab💥
www.theinnovati0nlab.com