Jihad needs scientists
in [Design]
|
From: nonsense on 10 Mar 2007 18:51 Ken Smith wrote: > In article <72ef1$45f31da7$4fe7333$21289(a)DIALUPUSA.NET>, > nonsense(a)unsettled.com <nonsense(a)unsettled.com> wrote: > >>Ken Smith wrote: >> >>>In article <b6a74$45f2f36e$4fe72af$20401(a)DIALUPUSA.NET>, >>>nonsense(a)unsettled.com <nonsense(a)unsettled.com> wrote: >>> >>> >>>>Ken Smith wrote: >>>> >>>> >>>>>In article <esu5o4$8ss_003(a)s861.apx1.sbo.ma.dialup.rcn.com>, >>>>><jmfbahciv(a)aol.com> wrote: >>>>> >>>>> >>>>> >>>>>>In article <esrr8b$n5i$2(a)blue.rahul.net>, >>>>>>kensmith(a)green.rahul.net (Ken Smith) wrote: >>>>> >>>>>[....] >>>>> >>>>> >>>>> >>>>>>>>These used to be called private packs. The concept has existed since >>>>>>>>the 60s. >>>>>>> >>>>>[....] >>>>> >>>>> >>>>> >>>>>>>The operator allocated them >>>>>>>after you paid large amounts of money. >>>>>> >>>>>>That depended on the site. You seem to be talking from an IBM >>>>>>operational POV. >>>>> >>>>> >>>>>Yes, an IBM environment >>>>> >>>>> >>>>> >>>>> >>>>>>Ours was designed differently. It was easy >>>>>>to redirect any spooling to a pack reserved for that purpose. >>>>>>Video downloads, etc. could be in a similar category. >>>>> >>>>> >>>>>Who did the "reserved for that purpose"? That would be the point where >>>>>money would be needed. >>>>> >>>>>[....] >>>>> >>>>> >>>>> >>>>>>>Where you evolve to depends a lot on where you start. In this case, there >>>>>>>is a large factor from the seemingly unimportant choices made in the early >>>>>>>days. >>>>>> >>>>>>Those weren't unimportant choices. They were deliberately made with >>>>>>certain goals and non-goals in mind. No development was an >>>>>>accident. >>>>> >>>>> >>>>>Sure it was. In both hardware and software design there are often choices >>>>>that look identical today but won't in the future. For a long time logic >>>>>has run on 5V. The selection of 5V can be traced in part to the heater >>>>>voltage on tubes. >>>> >>>>That's not true. Most of the vacuum tubes use(d) 6 volts and upwards. >>> >>> >>>Yes, it is true. The heater voltage on many tubes was 6.3VAC. I'll leave >>>the *sqrt(2) and the headroom numbers for you to go find. When you get >>>done, you will see why 5V was the nearest round number. >> >> >>6.3 * .707 = 4.45 making 4 volts the nearest round number. > > > sqrt(2) != 0.707 Thought you didn't know what you were talking about. It seems I was mistaken about *how* you were wrong. > I'll do it for you: > > 6.3 * sqrt(2) = 6.3*1.414 = 8.9V > > Low line condition = -10% ie: 0.9 times the above > > 8.9 * 0.9 = 8.0 > > The rectifier is a silicon diode: > > 8.0 - 0.7 = 7.3V > > The regulator was from before LDOs were invented. So lets use the LM78XX > as an example design: > > http://www.fairchildsemi.com/pf/LM/LM7805.html says drop out voltage = > 2.0 > > 7.3 - 2.0 = 5.3 > > Round off > > 5V LM78XX series provides: Output Current up to 1A Output Voltages of 5, 6, 8, 9, 10, 12, 15, 18, 24 Thermal Overload Protection Short Circuit Protection Output Transistor Safe Operating Area Protection That there used to be plenty of transformers incorporating a 6.3 volt winding among the multiple outputs common to transformers during vacuum tube days has nothing to do with the reason 5 volts was selected for logic circuits. There's no good reason to have selected 5 volts as opposed to 6 volts, or for that matter, 8 volts. There's nothing sacred about manufacturing 6.3 volt transformers as opposed to any other particular low voltage you might select. An entire electronics industry didn't grow up around some suspect warehouse full of obsolete/surplus 6.3 volt transformers someplace. Looking at the availability of the voltages in the LM series (I had an S-100 buss computer once that had cards loaded down with the things) any voltage could have been just as easily selected.
From: Ken Smith on 10 Mar 2007 19:49 In article <3dbb5$45f34488$4fe71c8$23256(a)DIALUPUSA.NET>, nonsense(a)unsettled.com <nonsense(a)unsettled.com> wrote: [....] >That there used to be plenty of transformers incorporating >a 6.3 volt winding among the multiple outputs There were also huge numbers of "filament transformers" which had only the windings for the heaters. There were vast production lines that made them by the zillions at low costs. [....] > common to >transformers during vacuum tube days has nothing to do >with the reason 5 volts was selected for logic circuits. > >There's no good reason to have selected 5 volts as opposed >to 6 volts, or for that matter, 8 volts. There's nothing >sacred about manufacturing 6.3 volt transformers as opposed >to any other particular low voltage you might select. You could buy a filament transformer for way less than one at any other voltage you may select. > An >entire electronics industry didn't grow up around some suspect >warehouse full of obsolete/surplus 6.3 volt transformers >someplace. They weren't obsolete on the day the choice was made. They were still used in massive numbers. >Looking at the availability of the voltages in the LM series >(I had an S-100 buss computer once that had cards loaded >down with the things) any voltage could have been just as >easily selected. -- -- kensmith(a)rahul.net forging knowledge
From: Eeyore on 10 Mar 2007 23:26 MassiveProng wrote: > Eeyore <rabbitsfriendsandrelations(a)hotmail.com> Gave us: > >"nonsense(a)unsettled.com" wrote: > >> Ken Smith wrote: > >> > > >> > Yes, it is true. The heater voltage on many tubes was 6.3VAC. I'll leave > >> > the *sqrt(2) and the headroom numbers for you to go find. When you get > >> > done, you will see why 5V was the nearest round number. > >> > >> 6.3 * .707 = 4.45 making 4 volts the nearest round number. > > > >How about the forward voltage drop for the rectifier ? > > Semiconductor diodes, yes. > > Do tube rectifiers have the same "forward drop"? It's different. > I doubt it as a hot tube is very happy to pass electrons. Certainly not the amps required. A curious idea to use a vacuum tube rectifier to provide power for ICs btw. I'm sure it would go down well with the audiophools. Graham
From: Eeyore on 11 Mar 2007 00:01 "nonsense(a)unsettled.com" wrote: > There's no good reason to have selected 5 volts as opposed > to 6 volts, or for that matter, 8 volts. Utter nonsense. The IC process used for TTL has a breakdown voltage of around 7 volts. The supply voltage has to be less than that. Graham
From: jmfbahciv on 11 Mar 2007 07:07
In article <esuq2s$ds3$4(a)blue.rahul.net>, kensmith(a)green.rahul.net (Ken Smith) wrote: >In article <esu5o4$8ss_003(a)s861.apx1.sbo.ma.dialup.rcn.com>, > <jmfbahciv(a)aol.com> wrote: >>In article <esrr8b$n5i$2(a)blue.rahul.net>, >> kensmith(a)green.rahul.net (Ken Smith) wrote: >[....] >>>>These used to be called private packs. The concept has existed since >>>>the 60s. >>> >[....] >>> The operator allocated them >>>after you paid large amounts of money. >> >>That depended on the site. You seem to be talking from an IBM >>operational POV. > >Yes, an IBM environment Our products weren't IBM. So we had different constraints and tradeoffs. > >> Ours was designed differently. It was easy >>to redirect any spooling to a pack reserved for that purpose. >>Video downloads, etc. could be in a similar category. > >Who did the "reserved for that purpose"? That depended on what was getting done and who needed it. > That would be the point where >money would be needed. > >[....] >>>Where you evolve to depends a lot on where you start. In this case, there >>>is a large factor from the seemingly unimportant choices made in the early >>>days. >> >>Those weren't unimportant choices. They were deliberately made with >>certain goals and non-goals in mind. No development was an >>accident. > >Sure it was. In both hardware and software design there are often choices >that look identical today but won't in the future. For a long time logic >has run on 5V. The selection of 5V can be traced in part to the heater >voltage on tubes. Now think about that over time. The constraint was 5V so choices made from one development project to the next had to include the old choices. Most projects had to think about backwards compatibility and the parts and power that was available at the time of planning. Then implementation was also determined by the parts and resources available at the time of the implementation. You can't make something that depends on a foobar existing at the time of production. There is also the requirement that there will be enough of the foobars so manufacturing the item won't be held up. /BAH |