From the title, I thought this was going to be a dull video - I was wrong. Great job - I did not know of the alternatives and just assumed that if you needed dc, you needed a diode to make it. (my 55 year old Navy training is letting me down!). Speaking of my Navy time, have you ever played around with 3 phase power. I was trained on a system that used 400hz, 3 phase power. To get dc, all the system did was use a 3 phase full wave rectifier and a simple LC filter. No active components, no regulation, no feedback. And it made an extremely clean dc (for it's time). Good enough to run the transistor based computers we were using.
This was a great video. This will come in handy for me an my future endeavors. Thank you so much for taking the time to create the content and putting it out there.
This is great! I never thought about controlling transistors to replace diodes in a bridge rectifier. Graet videos man, you always renew my passion for electronics!
Active Rectifier's are great when you have a DC supply that need to be fed from a few different supplies and a backup battery. Only a little loss as the body diode conducts before it bypassed by Ron. AC backup is no problem as the UPS's do it all for you, but high current DC can leave you with a bit of head scratching.
Great video topic! I’d taken the good old bridge rectifier for granted, and fully overlooked that there might be some smarter modern alternative solutions. It’s always good to come across a video that re-educates you. :)
I'm impressed with your really well produced and informative videos, I always learn a lot from them. When I saw the title, I wondered if you'd pay homage to a certain Full Bridge Rectifier person so was pleased to see an early acknowledgement!
If after watching this video I felt that something was really badly wrong, technically or otherwise, I would reach out to SDG via the comments to try and start a discussion, get things sorted. Thankfully this video is top quality, free and educational. As are all of Steve's work. I have nothing but respect for SDG and cannot imagine what the circumstances are that make someone give this video a thumbs down. It's about them, isn't it?
Edit: Nevermind, you already had a followup video with the capacitors. I would liked to see LT chip waveforms and efficiency with large capacitor installed. It showed taking operating voltage from DC side of rectifier, so getting stable voltage over 9V could increase effciency as it would be in operating state all time. Also if capactior keep voltage drop small enough, mosfets not turned on part of curve won't be used and efficiency should go up a lot.
Did you ever fit the output smoothing caps and test? Would be interesting to see if they all acted in a similar manner or whether you have to tweak the values around somewhat. Anyway, great video. You've found a new subscriber! :)
I expect the Mosfet Fullbridge to perfrom much better than the others when using smoothing, because then all the current conducting is going to be done above the 9V where controller chip starts operating
yeah these mosfets are way to 'small' to produce such high switching losses and we saw on scope that Vpk was almost identical on the output, so there should be a huge difference.
Great video!! Really enjoyed! Can you do a video of converting the main supply to 3.3V in the smallest full print possible? For example to supply an ESP.
I've just been looking online for this NMLU1210 - it seems they are hard to get hold of? Might be worth me muckin' about with a similar circuit using separate components one day - something to do this winter during the long nights. I guess I won't get the same efficiency as this carefully crafted design though.
I like the quad mosfet bridge the best :) It might probably even be the most cost efficient and environmentally friendly solution even the initial part cost is higher. what you pay extra for the ICs you can save in heat sinking and power waste .. over a long time period they might be even cheaper, depending on the overall power throughput of course. I thought I have seen a quad mosfet bridge with no active components ... maybe mixed that up with an H-Bridge. Interesting stuff, thanks Steve !
@@sdgelectronics That one kept me busy quite a while, but I (now) think it is not possible, you'd need a square wave to switch the gates at least ... there is an article on the eevblog (bridge-rectifier-made-with-mosfets) that came closest to an answer .. sorry for the mixup. Still I think quad mosfet is the way to go even if you need some ic with a little bit of brain. ;)
As far as I'm aware the LT4320 is essentially useless for mains rectification due to its internal charge pumps being unable to generate sufficient gate drive voltage. (the datasheet is rather unclear about voltage ratings but there is no indication that it can be used for mains voltages). It's also useless for secondary side SMPS rectification due to the max 600 HZ switching frequency - not nearly fast enough. For rectifying small signals you'd probably opt for a "precision rectifier" type circuit - there are plenty of cheaper and better options for low current. That pretty much relegates this chip to a few niche low voltage, low frequency, high current applications such as "hot swap" load switching and other non-rectification ideal diode uses. An interesting sounding chip without many real applications sadly.
Indeed a very, very interesting topic. Recently I've repaired some 1541 drives. Their bridge recitifiers get really hot and I had some of them fail on me already. I was wondering if I could simply replace them with Schottky bridges for these low voltages. Then I noticed that Schottky diode bridge rectifiers are not commonly available at all. But why? Anyway, those active recitifiers seem to be a great solution. Maybe I can find some comparable products that are actually available and affordable.
Interesting! I didn't know there was more sophisticated alternatives to the good old full bridge rectifier. BTW; what type/brand is that 12ohm load resistor?
Steve, ive take inspiration from this video to use of LT4320 in my next project (didnt know about this solution before) as a 15A full bridge recifier for low noise power supply. I have to say, speaking about losses on mosfets, everything works fine, but i discovered some spurious in output spectrum every 10ms (100Hz) which datasheet is not mentioning. This looks like there is some "ringing" during mosfet switching :( I was not able to find email contact to you to send you screenshots.
If you go to the "ABOUT" page on my channel, my e-mail address is there. The ringing can be due to several things so I'll be interested to see what's going on.
Is there a better ideal IC that runs at a lower voltage than 9V, ie, switch on sooner, ie, have a better trace. I tried to buy some caps (3 different values) from RS here down-under, and they don't like selling stuff unless you buy multiplies for about $10 per cap value, ie, $30 just to replace a few caps on a TV tuner card. Plus with the virus you can't pick them up, so it's another $12.50 postage, ie $40 total. I went Ali for $12 total - lol -.
Shame the NMLU1210 is being discontinued from Onsemi, but I guess one can make it from discrete components. You should have also considered the LM74670 from TI, though not cheap and you need 4 and some capacitors.
Ah I hadn't noticed that. I think in the PACE soldering station they had a discrete transistor solution for the bridge rectifier. I'll have to trace that out to see how they've done it.
@@sdgelectronics There are mosfet rectifiers by using two dual (n and p in the same package) that I have seen though you cannot use smoothing caps afterwards, so you still possibly need a diode, though that diode is conducting at all time, the nmlu1210 uses two that conduct alternatively so that should help with generated heat. I'll consider making it with discrete components and test it, great tip from your video! Though curious how the rectifier in the soldering station is laid out.
I'm disappointed that you didn't solder the capacitor to the LT4320. As you said yourself, it needs 9V to turn on, so it needs a capacitor at its output that keeps the output voltage above 9V. That would probably increase the efficiency substantially.
Ok, I am really confused right now, from videos that I searched, the theoretical maximum efficiency for rectifiers are 82%, how did you get 96%? Could you make a video explaining that?
I'd be interested to see the solutions placed into a more favorable competition. Change out the 4007 for something that can take 3A and then re-run the comparison, as that would help justify the use of these solutions to users.
Off topic here, but how about reviewing a 'chip audio amp' like the TDA1517 but built onto an aluminium PCB, I say this because another favourite you tuber of mine "John Audio Tech" just added another video to his series called the 'chip amps done right' these are great videos and I wonder how some of these 1 package amps perform as heatsinking is a huge factor in the designs, I believe it would be interesting as this is new technology for the hobbyist (at least for me !) just thought I would give you my thoughts....cheers.
Steve im looking at trying 4g Bb hear as im out in the sticks, and best upload speed i can get is 8meg. Not good to build youtube channel with What conclusion did you come to on service and hardware. Steve
I have pretty consistent results on the Mikrotik once the TTL setting is changed to 64. 24 Mbps most of the time. I would consider the one with the parabolic reflector though
I was hoping you'd try the active solutions at 2 and 3 Arms and also with come capacitance on the output as that may effect the LTC4320 efficiency, a little disappointing because i think that's where you'd see bigger improvements. Ahhwell...
Remind you viewers to NEVER do what you said: ground one side and probe the other side of an AC input, at least not with an earth-grounded scope. I know you meant a generalized AC circuit, but the closest one is from the wall!
Well, not 'never', it would be fine on isolated AC from transformer even if scope is grounded. And his schematic was correct, as this is the symbol for common potential and not protective earth / chassis ground. It's just the annotation wording that was sloppy.
![Lp. Сердце Вселенной #30 ДВА В ОДНОМ [Один человек]• Майнкрафт](http://i.ytimg.com/vi/nrXy15orKhY/mqdefault.jpg)
Watching PhotonInduction play with 1000's of amps and Kilovolts in his living room makes my tummy all wobbly.
Imagine a guy like PI living next door t your place:))
From the title, I thought this was going to be a dull video - I was wrong. Great job - I did not know of the alternatives and just assumed that if you needed dc, you needed a diode to make it. (my 55 year old Navy training is letting me down!). Speaking of my Navy time, have you ever played around with 3 phase power. I was trained on a system that used 400hz, 3 phase power. To get dc, all the system did was use a 3 phase full wave rectifier and a simple LC filter. No active components, no regulation, no feedback. And it made an extremely clean dc (for it's time). Good enough to run the transistor based computers we were using.
Exceptional video. Thanks.
Please do this for mains voltage and larger currents.
Nice video , you explain very good the teory .
Great video, and discussion! Thanks!
Interesting stuff, that matte green did look nice....cheers.
"Quite chunky SOICs" had me smiling! Everything is relative. 😄
Fascinating experiment, Steve. I wasn't aware of other options to Full Bridge Rectification. Thank you.
I wonder if it's even possible these days for a tech youtuber to say FULL BRIDGE RECTIFIER without the camera shaking?
No, not possible
@@sdgelectronics And where's the eyebrow dance? I was really looking forward to it :-D
Clear and concise. Nice job. And here I was thinking I knew all I needed to know about FWBR!
Really interesting video. I really like your content. Keep going!!!!
This was a great video. This will come in handy for me an my future endeavors. Thank you so much for taking the time to create the content and putting it out there.
This is great! I never thought about controlling transistors to replace diodes in a bridge rectifier. Graet videos man, you always renew my passion for electronics!
The PCB looks very nice.
Very good idea for a video and interesting results. I'm aware of the integrated FBR's, but didn't know about these active solutions, so good to know.
Great video thank you I have never come across those active bridge rectifiers before so really good info thanks
Great topic from 87 to 96 %. Reliability and sensitivity to dirty input maybe something to consider. Great idea and nice video. Thank you very much.
Active Rectifier's are great when you have a DC supply that need to be fed from a few different supplies and a backup battery. Only a little loss as the body diode conducts before it bypassed by Ron.
AC backup is no problem as the UPS's do it all for you, but high current DC can leave you with a bit of head scratching.
Great video topic! I’d taken the good old bridge rectifier for granted, and fully overlooked that there might be some smarter modern alternative solutions. It’s always good to come across a video that re-educates you. :)
Very informative. Thanks
I'm impressed with your really well produced and informative videos, I always learn a lot from them. When I saw the title, I wondered if you'd pay homage to a certain Full Bridge Rectifier person so was pleased to see an early acknowledgement!
If after watching this video I felt that something was really badly wrong, technically or otherwise, I would reach out to SDG via the comments to try and start a discussion, get things sorted. Thankfully this video is top quality, free and educational. As are all of Steve's work.
I have nothing but respect for SDG and cannot imagine what the circumstances are that make someone give this video a thumbs down. It's about them, isn't it?
It's usually people not liking sponsored videos, it's counted positively by UA-cam though as your video forced someone to interact.
It's a very helpful video.
You've been thinking about bridge rectifiers since 1970? Think there is a clear winner for cost and efficiency. Good video
Edit: Nevermind, you already had a followup video with the capacitors.
I would liked to see LT chip waveforms and efficiency with large capacitor installed. It showed taking operating voltage from DC side of rectifier, so getting stable voltage over 9V could increase effciency as it would be in operating state all time. Also if capactior keep voltage drop small enough, mosfets not turned on part of curve won't be used and efficiency should go up a lot.
Did you ever fit the output smoothing caps and test? Would be interesting to see if they all acted in a similar manner or whether you have to tweak the values around somewhat.
Anyway, great video. You've found a new subscriber! :)
No I ran out of time, but I will do a short follow-up
I expect the Mosfet Fullbridge to perfrom much better than the others when using smoothing, because then all the current conducting is going to be done above the 9V where controller chip starts operating
@@elektro-peter1954 thanks, exactly my thought. I was really confused why he didn't after he specifically pointed out that it had to be above 9V
yeah these mosfets are way to 'small' to produce such high switching losses and we saw on scope that Vpk was almost identical on the output, so there should be a huge difference.
That green PCB looks nice. I avoid matte black PCBs like the plague, although they look rather nice. +1 for the FULL BRIDGE RECTIFIER bit.
I see a monobrow, I click!
Great video!! Really enjoyed!
Can you do a video of converting the main supply to 3.3V in the smallest full print possible? For example to supply an ESP.
I've just been looking online for this NMLU1210 - it seems they are hard to get hold of? Might be worth me muckin' about with a similar circuit using separate components one day - something to do this winter during the long nights. I guess I won't get the same efficiency as this carefully crafted design though.
Yeah, sadly I hadn't noticed - RS had lots in stock so I didn't think twice about investigating it. I'll look for a other options.
very interesting!
"FULL BRIDGE RECTIFIER"reminds me of electroboom NOSTALGIA,ELECTROBooM 🔥🔥
That was purposeful.
Nice unibrow thumbnail ! :)
Could you try adding MLCC capacitors to the Standard Bridge Rectifier ? 😀
I like the quad mosfet bridge the best :)
It might probably even be the most cost efficient and environmentally friendly solution even the initial part cost is higher. what you pay extra for the ICs you can save in heat sinking and power waste .. over a long time period they might be even cheaper, depending on the overall power throughput of course. I thought I have seen a quad mosfet bridge with no active components ... maybe mixed that up with an H-Bridge.
Interesting stuff, thanks Steve !
I think there's one in the PACE soldering station, I need to trace it out as I can't quite think how it would work.
@@sdgelectronics That one kept me busy quite a while, but I (now) think it is not possible, you'd need a square wave to switch the gates at least ... there is an article on the eevblog (bridge-rectifier-made-with-mosfets) that came closest to an answer .. sorry for the mixup. Still I think quad mosfet is the way to go even if you need some ic with a little bit of brain. ;)
I thought it was called a "full 'wave' bridge rectifier"
As far as I'm aware the LT4320 is essentially useless for mains rectification due to its internal charge pumps being unable to generate sufficient gate drive voltage. (the datasheet is rather unclear about voltage ratings but there is no indication that it can be used for mains voltages). It's also useless for secondary side SMPS rectification due to the max 600 HZ switching frequency - not nearly fast enough. For rectifying small signals you'd probably opt for a "precision rectifier" type circuit - there are plenty of cheaper and better options for low current. That pretty much relegates this chip to a few niche low voltage, low frequency, high current applications such as "hot swap" load switching and other non-rectification ideal diode uses. An interesting sounding chip without many real applications sadly.
Indeed a very, very interesting topic. Recently I've repaired some 1541 drives. Their bridge recitifiers get really hot and I had some of them fail on me already. I was wondering if I could simply replace them with Schottky bridges for these low voltages. Then I noticed that Schottky diode bridge rectifiers are not commonly available at all. But why? Anyway, those active recitifiers seem to be a great solution. Maybe I can find some comparable products that are actually available and affordable.
Actually I'm more reminded of ElectroBOOM!
Were your efficiency calculations including the losses from the resistors you put inline in each circuit?
I ended up bypassing them in the testing as they weren't needed.
great video :)
As most power supplies use standard diodes, they really gotta push for that 80% efficiency rating
Interesting! I didn't know there was more sophisticated alternatives to the good old full bridge rectifier. BTW; what type/brand is that 12ohm load resistor?
It was from RS, their own brand RSPro uk.rs-online.com/web/p/panel-mount-fixed-resistors/1852424
Thanks@@sdgelectronics ! I've been soldering together 5W power resistors too many times :)
Steve, ive take inspiration from this video to use of LT4320 in my next project (didnt know about this solution before) as a 15A full bridge recifier for low noise power supply. I have to say, speaking about losses on mosfets, everything works fine, but i discovered some spurious in output spectrum every 10ms (100Hz) which datasheet is not mentioning. This looks like there is some "ringing" during mosfet switching :( I was not able to find email contact to you to send you screenshots.
If you go to the "ABOUT" page on my channel, my e-mail address is there. The ringing can be due to several things so I'll be interested to see what's going on.
Is there a better ideal IC that runs at a lower voltage than 9V, ie, switch on sooner, ie, have a better trace.
I tried to buy some caps (3 different values) from RS here down-under, and they don't like selling stuff unless you buy multiplies for about $10 per cap value, ie, $30 just to replace a few caps on a TV tuner card. Plus with the virus you can't pick them up, so it's another $12.50 postage, ie $40 total. I went Ali for $12 total - lol -.
I didn't find one specifically. I think you can do it without any active parts if you don't mind using 2x N channel MOSFETs and 2x P channel MOSFETs
Where can I get a link to the tool you used to pick-n-place the components?
ua-cam.com/video/1FnGqH_WkL4/v-deo.html
Shame the NMLU1210 is being discontinued from Onsemi, but I guess one can make it from discrete components. You should have also considered the LM74670 from TI, though not cheap and you need 4 and some capacitors.
Ah I hadn't noticed that. I think in the PACE soldering station they had a discrete transistor solution for the bridge rectifier. I'll have to trace that out to see how they've done it.
@@sdgelectronics There are mosfet rectifiers by using two dual (n and p in the same package) that I have seen though you cannot use smoothing caps afterwards, so you still possibly need a diode, though that diode is conducting at all time, the nmlu1210 uses two that conduct alternatively so that should help with generated heat. I'll consider making it with discrete components and test it, great tip from your video! Though curious how the rectifier in the soldering station is laid out.
Can this be done with op amps?
I'm disappointed that you didn't solder the capacitor to the LT4320. As you said yourself, it needs 9V to turn on, so it needs a capacitor at its output that keeps the output voltage above 9V. That would probably increase the efficiency substantially.
There's a follow-up coming on Thursday.
Ok, I am really confused right now, from videos that I searched, the theoretical maximum efficiency for rectifiers are 82%, how did you get 96%?
Could you make a video explaining that?
Because the low efficiency is due to the voltage drop across the diode, hence the losses. In this case, there is no diode loss.
That is a cool little test and learning board. Are you selling them or is it a shared design on pcbway?
It's on PCBWay: www.pcbway.com/project/shareproject/High_Efficiency_Full_Bridge_Rectifier_Demo_Boards.html
Will you do a follow up with capacitors installed?
yes, should be available to watch on Thursday
@@sdgelectronics look forward to it!
I'd be interested to see the solutions placed into a more favorable competition. Change out the 4007 for something that can take 3A and then re-run the comparison, as that would help justify the use of these solutions to users.
It's true, with the original schottky diodes I had in the BOM, you'd get closer to 92%
Off topic here, but how about reviewing a 'chip audio amp' like the TDA1517 but built onto an aluminium PCB, I say this because another favourite you tuber of mine "John Audio Tech" just added another video to his series called the 'chip amps done right' these are great videos and I wonder how some of these 1 package amps perform as heatsinking is a huge factor in the designs, I believe it would be interesting as this is new technology for the hobbyist (at least for me !) just thought I would give you my thoughts....cheers.
Nice thought, I also watch his content.
Steve im looking at trying 4g Bb hear as im out in the sticks, and best upload speed i can get is 8meg. Not good to build youtube channel with
What conclusion did you come to on service and hardware. Steve
I have pretty consistent results on the Mikrotik once the TTL setting is changed to 64. 24 Mbps most of the time. I would consider the one with the parabolic reflector though
@@sdgelectronics ok steve ill look into it
I was hoping you'd try the active solutions at 2 and 3 Arms and also with come capacitance on the output as that may effect the LTC4320 efficiency, a little disappointing because i think that's where you'd see bigger improvements. Ahhwell...
OK I'll do a 2 minute follow-up
@@sdgelectronics Cheers mate!
Interesting components. LT/AnalogDevices chips are always pretty darned expensive, it's a shame there's some distortion in the output of the LT part.
Congrats on not drawing the rectifier circuit as the classic diamond layout 😁
Can PCB Way make flat flexible cables too ?
You can make them from a flexi-pcb, but it may not be that cost effective.
Unfortunately no subtitles are available. Thanks anyway!
It's a fullWAVE bridge rectifier.
Remind you viewers to NEVER do what you said: ground one side and probe the other side of an AC input, at least not with an earth-grounded scope.
I know you meant a generalized AC circuit, but the closest one is from the wall!
If I had an Oscilloscope I would totally RTFM, it is probably the first sentence in Chapter one in bold Text. ;)
Sure, that's true but not everyone reads it.
Well, not 'never', it would be fine on isolated AC from transformer even if scope is grounded.
And his schematic was correct, as this is the symbol for common potential and not protective earth / chassis ground.
It's just the annotation wording that was sloppy.
Where there is advertising, THE TRUTH DOES NOT SHINE
Hey @electroboom you could learn something here 🙃😂🙃😂
Is that you snoopy. 🤣
There is no such thing as a full bridge rectifier.
Don't let people know this though.