How to Design a Low-Noise Dual Rail Voltage Supply

[@napent]

It would be great to see the spectrum analyzer output of the signals

[@kfcmcdonalds]

Switchmode and Low Noise are kinda mutually exclusive. Ultra gets even harder. The simplest design with two +/- 9v batteries will be the quietest you can get, across the full noise bandwidth.

[@M1America]

Not necessarily. you can always do both

[@EE-Mike]

Got a bit lost on the noise measurement, but other than that really nice design. I have seen a lot of this ground stitching but never knew why until now!

[@maks886]

That is a fantastic video, hats off!

[@AltiumAcademy]

Thank you so much 😀

[@oliverer3]

Figured I'd watch this before I start laying this out and funnily enough I'd selected almost the exact same components. I'll take that to mean I must have been doing something right :D

[@fernandoi8958]

Your setup is amazing, would love to know more about which equipment you are using. What are you using to hold the boards in place by the way?

[@Niels_Dn]

Why not use a switched mode regulator on the positive rail as well? By doing it with an LDO directly you’re dissipating over 300mW at full load there, which is a lot in a battery powered system.

[@alperkaplan1839]

thanks great video.

[@AltiumAcademy]

You are welcome!

[@archsys5643]

very nice!!!

[@AltiumAcademy]

Thanks for the visit

[@BMM66666]

Why turns on the bandwidth limit on the oscilloscope?

[@marsrocket]

I didn’t understand why you put the 0 ohm resistor in the design. Are you running a simulation before building it?

[@gpcm]

you only had 69 likes when i watched this video. I will come back later to give it a like because... you know

[@OriginalCircuit]

Nice!

[@felipehmattos]

awesome video! thank you

[@AltiumAcademy]

Glad you liked it!

[@TheOleHermit]

Darn! Hoped this would solve my supply noise problem, but I need to go from a 5VDC USB supply to +/-9VDC rails for my op amps. 😢 Nevertheless, thanks for the tips that I was able to understand.

[@OriginalCircuit]

You should be able to implement an almost identical topology - just add a boost regulator and LC filter before the positive side linear regulator. So your 5V comes in, goes to two switching regulators - a +9.5v and -9.5V, which then both go to LC filters, and then feed a +9V and -9V linear regulator respectively.

[@TheOleHermit]

@@OriginalCircuit Thanks so much for that info. I'm currently using a IH0509S unregulated 5V to +/-9V DC-DC converter, w/10uF caps on the rails for smoothing on the prototype. Assumed that I needed to add +9V & -9V linear regulators, so I'll look for a single split one instead. Any suggestions for the LC filters? Full disclosure, I'm a real dip-shit regarding circuit design, but not too bad with a soldering iron. I really appreciate your help.🙏

[@drhubs7049]

You're a badass!

[@krakamak]

Can someone tell me where can I find a similar silicon piece he used for the reflowing?

[@ShainAndrews]

No time stamp... no answer.

[@krakamak]

@@ShainAndrews 6.50

[@ShainAndrews]

@@krakamak You need to start using yer brain mate. He literally said what it was.

[@edgarwideman737]

@@ShainAndrews Sorry YOU need to start using your brain mate. @Nagy Bálint is asking (WHERE) to get the silicon mat. Not asking WHAT it is.

[@ShainAndrews]

@@edgarwideman737 Says the tard rocking emojis. Once a clown knows what something is... they damn well better be able to find where...

[@tristankordek]

👍

[@tfinmoraes]

Hi Mark, great content! Amazing how one can get an off the charts low noise PS that "easily" these days! Quick question: at kzbin.info/www/bejne/rn3NeIdpgZdgr8U are you using an AZV321 op amp in open loop as a comparator? If yes, a possible suggestion is to use a proper comparator and avoid the unlikely but possible false triggering due to the op amp saturation. Since I don't have an altium licence, I couldn't check the schematics directly.

[@OriginalCircuit]

Yup, I'm using the opamp as a comparator to keep the BOM smaller, its very much a non-critical area of the circuit!

[@DeVibe.]

Being satisfied by not measuring the output noise and assuming it's just negligible is a manifestation of delusion and a demonstration of the worst practices. Building a tiny amplifier to measure that noise would be a trivial task. Not doing it, especially when the PS is part of a very low noise system, is a big failure. (I won't laugh at assuming the noise level was "negligible" since it was lower than my inadequate oscilloscope's own noise or than that of a random resistor 🤭) I thought Altium (as a company) was much more serious than this. 🤦‍♂️ Now I know better. 🤡 Very important note: I still have respect for some key people like the wonderful engineers who made AD, and others like Zachariah Peterson.

[@DeVibe.]

BTW, what makes you still use opamps as comparators? And most importantly, why does the first one lack hysteresis? Imagine what happens when the battery's voltage gets lower much slower than you expected, which, at some point, will cause the "comparator" to act as a crazy noisy amplifier 😂. There is lots of true comparators to choose from, for example, RS8901XF. IMHO, you just made the worst choice.