EEVblog #238 - Power Supply Design Part 7.

[EEVblog]

Will there be a revision C schematic? Is the design frozen at this stage? If so, I may volunteer to create a multi page schematic (Altium designer based) if that would be useful? It may take a few weeks though as I'm still learning how to use AD10. It will also give me a chance to get some practice. Please let me know if this would be a useful contribution.

Thanks, but I'm not after a multi-page schematic. I prefer the single page version.
There's always room for a Rev C!
But the PCB has been ordered for Rev B

Dave.

[thilo]

Will there be a revision C schematic? Is the design frozen at this stage? If so, I may volunteer to create a multi page schematic (Altium designer based) if that would be useful? It may take a few weeks though as I'm still learning how to use AD10. It will also give me a chance to get some practice. Please let me know if this would be a useful contribution.

For an open source project that's hardly useful, because the schematics should not only be available as a PDF, but also as project files for a CAD program that many people have access to

[djsb]

OK then, If I did a version using Kicad would you provide that for download on your site Dave?

David.

P.S I'm not going to put in the effort if there is no demand for KiCAD project files and unless they can be shared with the community.

[RJSC]

I don't want to bug you with the "hi, please implement my feature" talk, I can mod to my own needs, but I want to ask you:

Is there anything against using I2C DAC and ADC?
Why have yet another bus and more used pins?
Since the feedback loop is in the LT3080 itself, the is no need to have super fast DAC and ADC communication with the minimal possible latency.
Having only one bus would simplify PCB routing (especially to people who are doing the PCB them selfs at home and you most definitely will like to do it single sided, not so much for Dave who will have them made at a fabhouse and probably double sided anyway).

I have seen power supplies where the feedback loop is implemented in the microcontroller driving a transistor directly, but since it's not the case, do you see any problems with using an I2C ADC and DAC?

[EEVblog]

Will there be a revision C schematic? Is the design frozen at this stage? If so, I may volunteer to create a multi page schematic (Altium designer based) if that would be useful? It may take a few weeks though as I'm still learning how to use AD10. It will also give me a chance to get some practice. Please let me know if this would be a useful contribution.

For an open source project that's hardly useful, because the schematics should not only be available as a PDF, but also as project files for a CAD program that many people have access to

I used Altium for this project, and that's what format the files are in.

Dave.

[EEVblog]

I don't want to bug you with the "hi, please implement my feature" talk, I can mod to my own needs, but I want to ask you:

Is there anything against using I2C DAC and ADC?
Why have yet another bus and more used pins?
Since the feedback loop is in the LT3080 itself, the is no need to have super fast DAC and ADC communication with the minimal possible latency.
Having only one bus would simplify PCB routing (especially to people who are doing the PCB them selfs at home and you most definitely will like to do it single sided, not so much for Dave who will have them made at a fabhouse and probably double sided anyway).

I have seen power supplies where the feedback loop is implemented in the microcontroller driving a transistor directly, but since it's not the case, do you see any problems with using an I2C ADC and DAC?

The SPI ones were cheaper.

Dave.

[EEVblog]

OK then, If I did a version using Kicad would you provide that for download on your site Dave?

Sure, no problem.

Dave.

[metalphreak]

There's a difference between "Open Source" and doing all the work for you

Recreating the PDF schematic in whatever CAD software you use isn't that difficult, and it encourages you to fully understand the circuit you are building, rather than just copy/paste and lay out a PCB. Open Source is about encouraging learning, innovation, and sharing, not tailoring free designs to each individuals needs and wants. Dave has done a wonderful job explaining how each part of his design works, which provides a fantastic base for you to build upon adding any extra features or specs you require.

If you do end up recreating it in KiCAD or Eagle for your own purposes then it will be greatly appreciated if you do share it

[Janaha]

Hy Folks, hy Dave,

i just want to say that i realy realy like the Power Supply series of videos. For me it's the first series of videos where not only an already made laboratory power supply is presenet to the public. It's the first show, where someone shows off all the details of his design desicions. And not only this, he makes it also so that it's understandable by someone like me, who only does some elektronic designs with microcontrollers as a hobbyist.

Power supply design just seemed to be the holy grail of electronics desings to me. Here in germany we have a german forum called www.microcontroller.net. I have seen so many posts there where some guy just kindly presented a power supply design as open source and got almost instantly flamed with a ton of negative posts, for what he did wrong in his design. But there never was a guy who showed how to make it right. And no one has the power supply that pleases all their desires.

So i want to say, keep up the good work. And many thanks for all the information and the very good presentation.
Janaha

[Zad]

Thanks, but I'm not after a multi-page schematic. I prefer the single page version.

As an aside, this reminds me of one of my pet annoyances. Designers (or, more likely, design department rules and guidelines) who put one IC or a handful of passive components on one sheet, and then group all the signals together into buses. CAD tools and abstraction concepts are great, but it can be taken too far. An engineer should be able to see the interconnections at a glance.

If it gets too complicated for  1 sheet, I tend to split it into analogue + digital. That doesn't disturb the visual flow too badly.

[PeteInTexas]

Did that reset pin really need to drive the LED? Is that LED really needed? Isn't there an LCD already?  Was finding screen real estate more costly than adding two more chips?

It's want I WANTED. I know that might be hard for some people to understand, but try, please.
If you want something different, go design your own.

Yes, of course.  So sorry.  Being from Texas I shoot from the hip a lot in speaking my mind.  But you're right.

And yes, I will be forking your design for myself.  I'll take everything after the DAC and before the ADC in rev A and hook it up to one of those duino thingys.  Knowing me, this will get done by 2025.

Quite frankly I'm disappointed.  I know that means nothing to you and thats fine.

Correct. Because if I cared what every single person thought about every aspect of any one of my videos, I'd end "doing a photonic induction". Seriously, I would have to quit the whole damn thing.
So why don't you just quit it with the unproductive comments? I don't need it.

No, no, no, nooooo! you're right, you're right!  Just don't pull a photonic induction on us. 

[enz]

There's always room for a Rev C!
But the PCB has been ordered for Rev B

Dave.

Hi Dave,

i don't know if someone mentioned this already, but i think you will really need a Rev C.
I just had a closer look to the Rev B schematic and i think you accidentally swapped the inputs of the current-control Opamp (U3B), in Rev A this seems to be correct.

Martin

Edit: typo removed

[senso]

If you dont mind I will start to copy this schematic to Eagle but use an Atmega644p, to have more voltage swing, maybe in the 15/18v range all I need to do is have some more gain in the voltage op-amp?

[MickM]

Hi Dave;
   I want to tell you how cool it is to watch a pro design something.
The decisions,and reasons why are very interesting.
I have learned a lot from watching your videos.

Have you decided how to sell them?
Bare PCB, PCB and a Bag-O-Parts, PCB with SMD installed, fully built and tested PCB, fully built and tested unit.
I would go for PCB + Bag-O-Parts.

Just a warning - you showed  your Philips frequency counter.
I have the PM6671.
So I pulled it out and it was a bit dusty (no front cover).

I wiped it down with regular rubbing alcohol (91% Isopropyl) - and the lettering started to smear.

Arrgh!! alcohol soluble paint - who knew!.
So now you are warned.

MickM (Seattle)

[Shale]

Hi dave. First post here. I have watched several of your videos and I am learning a lot. I like how you are doing this set of videos. I am learning about how and why you are making design decisions. I am very new to electronics and I do not know what everything is by any stretch but thats what google is for. Im sure it can be a pain in the tail with some people complaining about things, but I would like it if you did more builds like you are doing this one.

[amspire]

If you dont mind I will start to copy this schematic to Eagle but use an Atmega644p, to have more voltage swing, maybe in the 15/18v range all I need to do is have some more gain in the voltage op-amp?

You can increase the output voltage if you lower the current. You do not want the regulator IC going into thermal current limiting as you will start to get different output current limits to the one you have set.

At a current of 1A maximum, the maximum practical output is something like 10 -12 Volts. Reduce the current to 0.5A and you could increase the output to 20V with something like a 24V source for the supply.

Richard.

[BravoV]

Codename Platypus ??

[mobbarley]

Just saw that on twitter - nice sneak peak

[Nick Gammon]

Dave,

You really inspired me (in earlier videos) to get back into electronics after a lengthy break. Thank you very much!

Your videos about the power supply show clearly the sort of things that happen to a designer during the design process. Sometimes you can get half or three-quarters of the way through and realize you need to make major changes. That is normal, and you show how it happens.

Personally I don't care if the final version has Ethernet, Bluetooth, infrared remote or motion-sensing capabilities. The important thing is the knowledge you are imparting about design decisions. Knowledge we can all take with us to improve our own processes.

Remember: "A camel is a horse designed by a committee".

[ronwoch]

Just want to add to the chorus of voices saying thanks for letting us inside your head as you build this project! I am looking forward to the 'big reveal' as to what you have in mind, specifically, for this design.
Hoping the kits will be available around the time that my tax return is! 

Again, thanks for all you do Dave, it inspires everyone here. (Or else they shouldn't be here )

[sonicj]

scaled & tiled for US letter x 4

[krivx]

Is it too early to ask about pricing for kits?

[FreeThinker]

I Think Dave was aiming at around $100 or so .... But that was Yesterday

https://www.eevblog.com/forum/index.php?topic=6497.msg85141#msg85141

[krivx]

That seems very reasonable, I'm looking forward to it now.

[senso]

Thanks for the information Amspire, I will settle with the standart 10v version as it is enough to do electronics works, motors and powerfull things dont need a super precise power supply, but I have one question is there any substitute for the MAX4239ASA+, Mouser doesn't carry it and Maxim wants me to sign some strange contract to send me two samples of that chip, and I don't like to sign things that are under the USA laws..