On 2020-02-19, John Larkin <jlarkin@highland_atwork_technology.com> wrote:> On Wed, 19 Feb 2020 10:42:01 -0800 (PST), Lasse Langwadt Christensen ><langwadt@fonz.dk> wrote: > >>onsdag den 19. februar 2020 kl. 19.35.39 UTC+1 skrev John Larkin: >>> https://www.dropbox.com/sh/6scihn40t60qm7k/AADq46F-s7aBowSgcaVrIr_ua?dl=0 >>> >>> I'm still waiting for the Spice model for the 8078. I think I can make >>> it do positive to dual negative conversion. >>> >> >>https://www.analog.com/media/en/reference-design-documentation/design-notes/dn1021fa.pdf > > We often do that here. But I want the cute little dual LTM to make two > different negative rails. I'd like to Spice that, because the BGA is > hard to breadboard.That's probably why they have that breakout board. https://www.analog.com/en/products/ltm8078.html#product-evaluationkit It looks like it would work, but your middle voltage would be loading the bottom voltage... When I tried that same trick with a LM2596HV trying to make -36V from +12V it made smoke instead. rather than risk more parts trying to figure out why I found a different solution.> I'd love to x-ray the LTM8078. How can they do that?Do what? the datasheet implies that there's a 0.32mm thick PCB with the semiconductors and the inductors mounted on it. it looks like they used green solder mask.> Does anyone have a high-res x-ray? Ours is a big parts counter and > doesn't have useful resolution for ICs.how much do these things cost that you're not prepared to cut one open? -- Jasen.
neat power parts
Started by ●February 19, 2020
Reply by ●February 19, 20202020-02-19
Reply by ●February 19, 20202020-02-19
On Thu, 20 Feb 2020 00:40:55 -0000 (UTC), Jasen Betts <jasen@xnet.co.nz> wrote:>On 2020-02-19, John Larkin <jlarkin@highland_atwork_technology.com> wrote: >> On Wed, 19 Feb 2020 10:42:01 -0800 (PST), Lasse Langwadt Christensen >><langwadt@fonz.dk> wrote: >> >>>onsdag den 19. februar 2020 kl. 19.35.39 UTC+1 skrev John Larkin: >>>> https://www.dropbox.com/sh/6scihn40t60qm7k/AADq46F-s7aBowSgcaVrIr_ua?dl=0 >>>> >>>> I'm still waiting for the Spice model for the 8078. I think I can make >>>> it do positive to dual negative conversion. >>>> >>> >>>https://www.analog.com/media/en/reference-design-documentation/design-notes/dn1021fa.pdf >> >> We often do that here. But I want the cute little dual LTM to make two >> different negative rails. I'd like to Spice that, because the BGA is >> hard to breadboard. > >That's probably why they have that breakout board. >https://www.analog.com/en/products/ltm8078.html#product-evaluationkit > >It looks like it would work, but your middle voltage would be loading >the bottom voltage...We have the eval board on order, but everything is grounded, so it can't be used as a pos-to-neg converter, much less a dual.> >When I tried that same trick with a LM2596HV trying to make -36V from >+12V it made smoke instead. rather than risk more parts trying to >figure out why I found a different solution.We do this a lot https://www.dropbox.com/s/yz6s8k0ng3jd314/LTM8023_Inverter.jpg?raw=1 and it always works. The extra ferrite beads are to keep switcher noise down; there is delicate stuff nearby. -- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
Reply by ●February 20, 20202020-02-20
whit3rd <whit3rd@gmail.com> wrote in news:31b0afad-5e87-4690-a6f0-954923841ded@googlegroups.com:> On Wednesday, February 19, 2020 at 11:04:50 AM UTC-8, John Larkin > wrote: > >> Does anyone have a high-res x-ray? Ours is a big parts counter >> and doesn't have useful resolution for ICs. > > But, an X-ray will only show leads and little squares where the > silicon is. It's not likely to distinguish wiring or doping, > unless you first strip off the epoxy and do fluorescence > microprobing. > > For the kind of resolution that would find component parts of an > IC, you'd have to hang out at a synchrotron next to the X-ray > laser. > > Better to use an electron microscope (you still have to strip the > epoxy). Either way, synchrotron or microscope, you're doing it > with vacuum tubes. >Nobody reverse engineers chips any more. At least not anything with any appreciable density. Too many layers. Too much cost and by the time you finish, the chip you are trying to hack out is obsolete.
Reply by ●February 24, 20202020-02-24
On Wednesday, February 19, 2020 at 11:04:50 AM UTC-8, John Larkin wrote:> I'd love to x-ray the LTM8078. How can they do that? > > Does anyone have a high-res x-ray? Ours is a big parts counter and > doesn't have useful resolution for ICs.No LTM8078s around here, but here is an LTM8049 from my parts drawer: http://www.ke5fx.com/LTM8049_MX20_58256.146068_35kVp_15s.png These are handy because they can be used as dual SEPIC or inverting regulators, or any combination thereof. Especially good for those cases where you want a given output rail that lies within the expected input range. There is no drama or transitional behavior as the input voltage crosses the output voltage, it just works. -- john, KE5FX
Reply by ●February 24, 20202020-02-24
On Mon, 24 Feb 2020 03:01:49 -0800 (PST), "John Miles, KE5FX" <jmiles@gmail.com> wrote:>On Wednesday, February 19, 2020 at 11:04:50 AM UTC-8, John Larkin wrote: >> I'd love to x-ray the LTM8078. How can they do that? >> >> Does anyone have a high-res x-ray? Ours is a big parts counter and >> doesn't have useful resolution for ICs. > >No LTM8078s around here, but here is an LTM8049 from my parts drawer: > >http://www.ke5fx.com/LTM8049_MX20_58256.146068_35kVp_15s.png > >These are handy because they can be used as dual SEPIC or inverting >regulators, or any combination thereof. Especially good for those cases >where you want a given output rail that lies within the expected input >range. There is no drama or transitional behavior as the input voltage >crosses the output voltage, it just works. > >-- john, KE5FXThe older LTM bricks were LGAs, which didn't block the x-ray view like the solder balls do. But then, they made shorts on the PC board. Here's a SEPIC that's adjustable from 0 to +70 volts out, set with a local pot or a customer input. What's useful is that it runs with essentially a constant-power limit. https://www.dropbox.com/s/gc6g6g1cqbwnoh4/T770_Sepic_2.jpg?raw=1 -- John Larkin Highland Technology, Inc The cork popped merrily, and Lord Peter rose to his feet. "Bunter", he said, "I give you a toast. The triumph of Instinct over Reason"
