On 2023-02-05 17:27, Lasse Langwadt Christensen wrote:> søndag den 5. februar 2023 kl. 22.42.57 UTC+1 skrev Phil Hobbs: >> On 2023-02-05 16:37, Lasse Langwadt Christensen wrote: >>> søndag den 5. februar 2023 kl. 22.26.06 UTC+1 skrev Phil Hobbs: >>>> On 2023-02-05 15:46, Lasse Langwadt Christensen wrote: >>>>> søndag den 5. februar 2023 kl. 21.35.16 UTC+1 skrev Phil Hobbs: >>>>>> On 2023-02-03 16:16, piglet wrote: >>>>>>> On 03/02/2023 7:55 pm, Phil Hobbs wrote: >>>>>>>> So we keep being asked for simple TIAs as incidental parts of our >>>>>>>> consulting work. They're useful to have around anyway. >>>>>>>> >>>>>>>> It's a pain getting them to work right if we put SMPSes inside, so >>>>>>>> we're kicking around the idea of using a rail splitter on our usual >>>>>>>> 24V wall warts to make +15/-9 or something like that. We'd probably >>>>>>>> use a TCA0372 and a couple of big caps for the job. >>>>>>>> >>>>>>>> Two-wire warts are isolated, of course, so as long as no bright spark >>>>>>>> decides to try running our box on the same wart as somebody else's, we >>>>>>>> should be fine, I think, but we haven't done it before. >>>>>>>> >>>>>>>> Suggestions? >>>>>> >>>>>>>> >>>>>>> Wall warts are only isolated at DC! Some have horrible HF noise. >>>>>>> >>>>>>> Only you know the kind of users but a lot of folk have trouble getting >>>>>>> their mind around the idea of the non-positive supply rail not being >>>>>>> ground. The near standard barrel DC jack with center pin positive >>>>>>> encourages that notion so you may want to consider using a non-standard >>>>>>> power connector? >>>>>>> >>>>>>> piglet >>>>>>> >>>>>> We have that problem already, even in products with inverting switchers >>>>>> making the negative supply. >>>>>> >>>>>> Simon has been hacking KiCad to make it easy to plunk in our standard IP >>>>>> blocks, such as cap multipliers for various current ranges, fast quiet >>>>>> PGAs, MCUs with Modbus over RS-485, and so on. >>>>>> >>>>>> We have these nice machined brass boxes with feedthrough caps that we're >>>>>> using for lots of stuff these days. >>>>> >>>>> you machine the boxes? then machine a separate cavity for the inverter? >>>>> >>>>> >>>> Yeah, it's pretty cheap to do machined lead brass, because it cuts so >>>> easily. (You lot can't use it on account of the regime, but oh well.) >>>> >>>> We could do separate cavities and all that, but due to the >>>> getting-off-the-couch setup charge, it's better to use one box design >>>> and hang 3D printed stuff off to the side. (It all looks like DKNY >>>> black/gold bling anyway.) >>> >>> SMD RF shield cans are pretty cheap too >>> >> Yup. We use them a fair amount to control local interference, but just >> letting a noisy wire into the box creates problems from unknown outside >> stuff. > > would you have the same issue with a wall wart?Absolutely, if we let the noisy wires inside the machined box with no filtering. Cheers Phil Hobbs -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC / Hobbs ElectroOptics Optics, Electro-optics, Photonics, Analog Electronics Briarcliff Manor NY 10510 http://electrooptical.net http://hobbs-eo.com
Rail-splitting a wall wart
Started by ●February 3, 2023
Reply by ●February 5, 20232023-02-05
Reply by ●February 6, 20232023-02-06
Reply by ●February 6, 20232023-02-06
John Larkin wrote: ------------------------------> https://www.dropbox.com/s/e6okuzog0nulvjm/Rail_Splitter_B.jpg?raw=1** Using a AC ( transformer) wall wart has a few notable advantages it you are going to use linear regs. One with a single 12 volt secondary will supply +/- 16VDC or so using a ( 2 cap, 2 diode) voltage doubler. Highly reliable, no SMPS noise, very safe mains isolation. .... Phil
Reply by ●February 6, 20232023-02-06
On 2/6/2023 12:11 AM, Phil Allison wrote:> John Larkin wrote: > > ------------------------------ >> https://www.dropbox.com/s/e6okuzog0nulvjm/Rail_Splitter_B.jpg?raw=1 > > > ** Using a AC ( transformer) wall wart has a few notable advantages it you are going to use linear regs. > > One with a single 12 volt secondary will supply +/- 16VDC or so using a ( 2 cap, 2 diode) voltage doubler. > Highly reliable, no SMPS noise, very safe mains isolation. > > > .... PhilAnd a user who owns both an AC wall wart and a DC wall wart is bound to stick the AC one into your DC-input device eventually, anyway.
Reply by ●February 6, 20232023-02-06
On a sunny day (Sun, 5 Feb 2023 16:42:49 -0500) it happened Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote in <c5e52388-6aff-79e8-76d8-6e91b5f1eb29@electrooptical.net>:>>> and hang 3D printed stuff off to the side. (It all looks like DKNY >>> black/gold bling anyway.) >> >> SMD RF shield cans are pretty cheap too >> > >Yup. We use them a fair amount to control local interference, but just >letting a noisy wire into the box creates problems from unknown outside >stuff.In the old days when I did some audio design I used potcores with several windings making a sine oscillator to make + an - voltages from a simple 12 V or so DC input. I remember EPROM programmer needed a higher voltage too: http://panteltje.com/pub/8052AH_BASIC_computer/8052AH_BASIC_computer_inside_img_1727.jpg http://panteltje.com/pub/8052AH_BASIC_computer/8052AH_BASIC_computer_8049_programmer_board_img_1731.jpg http://panteltje.com/pub/8052AH_BASIC_computer/dc_dc_converter_detail/5V_to_30V_converter.jpg And all that is still working to this day (tested last year)! Eurocard, 100x160 mm, standard housing http://panteltje.com/pub/8052AH_BASIC_computer/8052AH_BASIC_computer_img_1725.jpg potcores do not radiate much.
Reply by ●February 6, 20232023-02-06
On Fri, 3 Feb 2023 14:55:32 -0500, Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:>So we keep being asked for simple TIAs as incidental parts of our >consulting work. They're useful to have around anyway. > >It's a pain getting them to work right if we put SMPSes inside, so we're >kicking around the idea of using a rail splitter on our usual 24V wall >warts to make +15/-9 or something like that. We'd probably use a >TCA0372 and a couple of big caps for the job.At least in the old days it was common to use an AC wall wart (about 20 Vac) and in the load device use two half wave rectifiers to make +15 Vdc and -15 Vdc for op-amps. What is wrong with this approach ? Is there also a requirement of running the device from a 24 V battery ?>Two-wire warts are isolated, of course, so as long as no bright spark >decides to try running our box on the same wart as somebody else's, we >should be fine, I think, but we haven't done it before.Using an AC wall wart 50/60 Hz iron core transformer with special connector will discourage many experimenters and their devices will not work with AC, possibly releasing some smoke :-)
Reply by ●February 6, 20232023-02-06
On Monday, February 6, 2023 at 12:57:49 AM UTC-8, upsid...@downunder.com wrote:> Using an AC wall wart 50/60 Hz iron core transformer with special > connector will discourage many experimenters and their devices will > not work with AC, possibly releasing some smoke :-)An AC connector can be unpolarized, but no DC connector commonly is. So, you can fit an AC wart with reversible plug, and no DC wallwart is likely to be compatible. I've got a subwoofer that takes AC/DC input, it has a DC-like connector and full wave bridge... can't get the polarity wrong either.
Reply by ●February 6, 20232023-02-06
whit3rd wrote:> upsid...@downunder.com wrote: > >> Using an AC wall wart 50/60 Hz iron core transformer with special >> connector will discourage many experimenters and their devices will >> not work with AC, possibly releasing some smoke :-) > > An AC connector can be unpolarized, but no DC connector commonly is. > So, you can fit an AC wart with reversible plug, and no DC wallwart is > likely to be compatible. > I've got a subwoofer that takes AC/DC input, it has a DC-like connector > and full wave bridge... can't get the polarity wrong either.Even when the voltage and power matches, it can be difficult to find a compatible connector. This thirty-four assortment of different sizes and shapes to connect a nominal 19.5VDC adapter to a laptop fails to cover all possible variances: <https://www.ebay.com/itm/314035139187> Danke, -- Don, KB7RPU, https://www.qsl.net/kb7rpu There was a young lady named Bright Whose speed was far faster than light; She set out one day In a relative way And returned on the previous night.
Reply by ●February 6, 20232023-02-06
On 2023-02-06 01:47, Jan Panteltje wrote:> On a sunny day (Sun, 5 Feb 2023 16:42:49 -0500) it happened Phil Hobbs > <pcdhSpamMeSenseless@electrooptical.net> wrote in > <c5e52388-6aff-79e8-76d8-6e91b5f1eb29@electrooptical.net>: > >>>> and hang 3D printed stuff off to the side. (It all looks like DKNY >>>> black/gold bling anyway.) >>> >>> SMD RF shield cans are pretty cheap too >>> >> >> Yup. We use them a fair amount to control local interference, but just >> letting a noisy wire into the box creates problems from unknown outside >> stuff. > > In the old days when I did some audio design I used potcores with several windings > making a sine oscillator to make + an - voltages from a simple 12 V or so DC input. > I remember EPROM programmer needed a higher voltage too: > http://panteltje.com/pub/8052AH_BASIC_computer/8052AH_BASIC_computer_inside_img_1727.jpg > http://panteltje.com/pub/8052AH_BASIC_computer/8052AH_BASIC_computer_8049_programmer_board_img_1731.jpg > http://panteltje.com/pub/8052AH_BASIC_computer/dc_dc_converter_detail/5V_to_30V_converter.jpg > > And all that is still working to this day (tested last year)! > > Eurocard, 100x160 mm, standard housing > http://panteltje.com/pub/8052AH_BASIC_computer/8052AH_BASIC_computer_img_1725.jpg > > potcores do not radiate much. >We build some things with ~ 2 MHz switchers inside, to keep the spurs out of band. They're a pain, though, because they can do unexpected things such as exciting unexpected PCB resonances in the hundreds of megahertz. Hobbyist approaches such as hand-winding potcores are fun the first couple of times, but get seriously old when you have lots of products to support with very limited manpower. They're also physically large, and have to be soldered manually. The idea here is to make it easy for a CM or a licensee to build and test stuff, so that we get paid without having to do all the work ourselves. Simon has done a lot of work on making bed-of-nails fixtures automatically, using the Kicad board files to get the pin positions, 3D printing to make the actual fixtures, and automatically-generated PCBs to hold the DUT and pins and replace the rat's nest of hand wiring that bed-of-nails testers usually have. It's really slick--he uses standard-ish jigs from AliExpress to hold the jig, DUT, and pressure plate, pressure pins properly distributed to keep the board straight, the whole proverbial nine yards (*) done by a couple of Python scripts. (He blogs about this sort of stuff periodically at <https://maskset.net>.) AC wall warts aren't a solution either. Getting even vaguely enough filtering at 120 Hz takes huge capacitors, because the inductors required to get even two poles would be as big as the entire box. Cheers Phil Hobbs (*) It was once thought that this originally referred to the capacity of a cement truck in cubic yards, but it doesn't. -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC / Hobbs ElectroOptics Optics, Electro-optics, Photonics, Analog Electronics Briarcliff Manor NY 10510 http://electrooptical.net http://hobbs-eo.com
Reply by ●February 6, 20232023-02-06
On 2023-02-05 23:43, John Larkin wrote:> > https://www.dropbox.com/s/e6okuzog0nulvjm/Rail_Splitter_B.jpg?raw=1 >That works great in two quadrants, and has the nice feature of dumping most of the waste heat into the zeners, which are cheap and can work at high temperature. From a belt-and-suspenders POV, it's comforting if it works in all four quadrants. (These boxes don't have motors in them, but it's been known to happen even so.) An extra couple of zeners would fix that up reasonably OK. The op amp gizmo doesn't protect against both outputs being pulled outwards simultaneously, but is good in four quadrants otherwise, and a single unipolar TVS diode on the power input fixes that. (That would be a very uncommon occurrence anyway.) The main high-dissipation scenario for our stuff is if somebody shorts the output when it's trying to produce a large signal. We generally solve that by using 22- or 33-ohm/10 uf RCs in the power leads of the output amp and a 2520-size 50-ohm series termination on the output. With a properly terminated load, 10V output costs 100 mA, which drops the positive supply by a couple of volts. We specify +- 10V outputs, and generally use +-14Vish supplies, so nothing weird happens at that level. The maximum dissipation in the output amp occurs at half-supply, where it dissipates about half a watt--not awful for a beefy part such as a THS3091. (We used to use more of those, but they're getting expensive these days--AD8045s and THS4631s are good. Even LM6171s are pretty good, if noisy, and they have thermal protection too.) Cheers Phil Hobbs -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC / Hobbs ElectroOptics Optics, Electro-optics, Photonics, Analog Electronics Briarcliff Manor NY 10510 http://electrooptical.net http://hobbs-eo.com
