23rd Nov 2020, 6:37 pm | #41 |
Nonode
Join Date: Jun 2016
Location: Bognor Regis, West Sussex, UK.
Posts: 2,300
|
Re: Would we have got to the moon with no semiconductors?
I seem to remember an idea to make valve "integrated circuits" using ceramic. The idea if I remember was to machine / mould tiny cavities in a slice of ceramic, print the electrodes and wiring, bond another layer on top and evacuate.
I remember this as a topic of conversation but have not seen any concrete proposals. Peter |
23rd Nov 2020, 6:47 pm | #42 |
Dekatron
Join Date: Aug 2016
Location: Colchester, Essex, UK.
Posts: 4,108
|
Re: Would we have got to the moon with no semiconductors?
I was groping for an example of development potential largely unforeseen, and LED technology seems like a good case.
I bet Fran Blanche would have a well reasoned opinion on the subject of this thread. Dave |
23rd Nov 2020, 7:38 pm | #43 |
Dekatron
Join Date: Jun 2009
Location: Staffordshire Moorlands, UK.
Posts: 5,274
|
Re: Would we have got to the moon with no semiconductors?
The russian space programme also used digital computers. Look on youtube 'Curious Marc' and others, he has bought some of it.
The semiconductors flown on apollo had to be liquid cooled, heat does not convect away easily without gravity, fan assisted or not. It's one of the many reasons that Apollo 13 had to shut right down, the damaged fuel cells being one of the main sources of cooling water. Imagine trying to run valves of any kind in that sort of environment. Even ICs would've overheated if they weren't allowed to shed their heat. The astronauts did not need blankets on their beds because if they laid still, their body heat simply clung to them. The film Apollo 13 portrays the crew's efforts trying to acheive a 'manual burn' of the LM descent engine in order to correct their trajectory on their earthbound journey. Done without the guidance computer, it is portrayed as a 3-man job, frantic and full of swear words a they battle to steer the craft towards earth. That was only a burn of a few seconds, and, if Jim Lovell's autobiography is anything to go by, was quite realistic. If they didn't have the MIT's chip-laden marvels, I wonder how the rest of the manouvres would've gone?
__________________
Kevin |
23rd Nov 2020, 8:49 pm | #44 |
Octode
Join Date: May 2011
Location: Lancashire, UK.
Posts: 1,351
|
Re: Would we have got to the moon with no semiconductors?
There is a brilliant (imo) podcast from the BBC World Service called 'Thirteen Minutes to the Moon' and it is fascinating listen from start to finish. Well worth downloading the episodes and they are great to listen to in the car.
Episode 5, 'The Fourth Astronaut' is all about the computer onboard Apollo. https://www.bbc.co.uk/programmes/p083t547 |
23rd Nov 2020, 10:36 pm | #45 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,902
|
Re: Would we have got to the moon with no semiconductors?
It isn't just the spacecraft and their guidance computer. Having valve-based radios for spacesuits and the larger batteries needed could be a problem.
There is a general issue that for small things batteries seem more efficient in packaging if they're low voltage/few cells. Valves wouldn't enable the high efficiency low voltage SMPS things we enjoy today. I think the Apollo fuel cell was a fairly low voltage thing and relied on switching converters. David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
23rd Nov 2020, 11:19 pm | #46 |
Octode
Join Date: Aug 2006
Location: Stoke-on-Trent, Staffordshire, UK.
Posts: 1,082
|
Re: Would we have got to the moon with no semiconductors?
Really the question should be would we have got back ?
The Russians had a completely manual space craft designed to do the job ! But as we all know the moon was a proxy war between these great super powers Trev |
24th Nov 2020, 12:05 am | #47 | |
Heptode
Join Date: Oct 2011
Location: Culcheth, Cheshire, UK.
Posts: 654
|
Re: Would we have got to the moon with no semiconductors?
Quote:
If Shockley et. al., or some Soviet equivalent, had never invented the transistor in the late 40's, then the last 70 years would have seen valve tech advance at a rate comparable to that seen in the semiconductor industry. Progress would have been driven by the same forces. Bletchley Park had already shown how valves could be used to compute, ( albeit in secret ), and other commercial outfits were soon to follow. The modern day technologies used to manufacture electronics would have enabled valve tech to advance far beyond glass bubbles with wires and plugs. So, in answer to the OP's question, Yes, but only if semiconductors had never been discovered in the first place. Cheers, Buzby |
|
24th Nov 2020, 1:34 am | #48 |
Dekatron
Join Date: Aug 2016
Location: Colchester, Essex, UK.
Posts: 4,108
|
Re: Would we have got to the moon with no semiconductors?
Buzby- you may have something there. We can never underestimate the pace of progress of an existing technology should the supercession technology never have happened..for whatever reason
Dave |
24th Nov 2020, 9:14 am | #49 |
Dekatron
Join Date: Nov 2010
Location: Oxford, UK.
Posts: 4,993
|
Re: Would we have got to the moon with no semiconductors?
|
24th Nov 2020, 10:11 am | #50 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,902
|
Re: Would we have got to the moon with no semiconductors?
As said earlier up the thread (MM) there is no 'PNP' valve, maybe P channel would be a better way of phrasing it. So pullups would have to be resistors (or very complex to do level shifting and inversion) This leaves RC timeconstants all over the place. Also the voltage swings would have to be greater than semiconductor technology because of poorer intrinsic conductivity. As a result, attempts at dense, fast, logic would be tied to a very unfavourable speed/power tradeoff.
As for the nutube rubbish, someone had a vfd factory and saw their market die as customers moved on to more general purpose LCDs. So they went looking for what else the same manufacturing process could make, and they came up with a token valve-y thing. Something that essentially does nowt but make people think they must have valve sound because there's a vacuum thingy glowing at them. Those things didn't catch on for a very good reason. They're rubbish and did nothing useful. With the valve technology of the time - miniature valves made for missile and shell fuze use, we could have got to the moon and back. It might have been more difficult, but not impossible. Compare a nutube to say an ECC82 and the nutube is one mighty step backwards in capability... and it's not one of the cold emitter nano technology spikes things. They only make sense to the owner of the VFD plant. The rest of us can just forget them. David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
24th Nov 2020, 10:30 am | #51 |
Heptode
Join Date: Oct 2011
Location: Culcheth, Cheshire, UK.
Posts: 654
|
Re: Would we have got to the moon with no semiconductors?
I'd forgotten about that Korg nutube device, but that's still not what I'm envisioning 21st Century valves to be.
The nutube is still a glass bubble with wires, uses a heat source, and is specifically aimed at audio applications. I'm thinking more of a digital device, that's what is needed for computing. What is smallest a 'valve' could be made ?. Is there a limit on how close elements can be ?. With micromechanics and nanotechnology I'm fairly certain a 'valve' computer comparable to what Apollo used could be built, if the will was there to do it. Cheers, Buzby |
24th Nov 2020, 11:12 am | #52 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,902
|
Re: Would we have got to the moon with no semiconductors?
I think you'll run into a speed/power per unit of logic tradeoff that is worse than what semiconductors can do. You could likely equal the apollo computer, but a CD player may be out of reach.
David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
24th Nov 2020, 11:28 am | #53 |
Heptode
Join Date: May 2017
Location: Konongo, Ghana
Posts: 516
|
Re: Would we have got to the moon with no semiconductors?
It might be possible with the kind of technology described in the attached article, although designing and producing this technology without the use of semi-conductors would be a challenge in itself.
If you turn the question of TS around (would we have got to the moon with no valves) I think the answer is probably no, because of the need for a traveling wave valve in the transmitters of spacecrafts. See: https://hal.archives-ouvertes.fr/hal-01754885/document
__________________
Robert |
24th Nov 2020, 1:45 pm | #54 |
Nonode
Join Date: Jun 2016
Location: Bognor Regis, West Sussex, UK.
Posts: 2,300
|
Re: Would we have got to the moon with no semiconductors?
Thank you for posting the pdf of "Integrated Vacuum Microelectronics", this is the sort of technology I refered to post 41. Glad I did not imagine it.
Peter |
24th Nov 2020, 2:19 pm | #55 |
Heptode
Join Date: Oct 2011
Location: Culcheth, Cheshire, UK.
Posts: 654
|
Re: Would we have got to the moon with no semiconductors?
Well, it looks like my idea could bear fruit.
That 'Integrated Vacuum Microelectronics' .pdf is describing early work done nearly 20 years ago by researchers. It describes exactly the sort of technologies I thought could be used to build a 'valve' computer, things like cold emitters and planar construction. If that twenty year old vacuum IC research had been continued, what could we be seeing from it now ?. ( Actually, there might be secret military devices already using this tech. Immunity to radiation would be a very attractive feature to have in a nuclear environment. We can only guess. ) |
24th Nov 2020, 4:37 pm | #56 | |
Heptode
Join Date: May 2017
Location: Konongo, Ghana
Posts: 516
|
Re: Would we have got to the moon with no semiconductors?
Quote:
More on this technology: https://repository.tudelft.nl/island...7-4e8a4efdd559 Off topic: My only stay in the UK was in Bognor Regis (1974, holiday with the family, I was 10 at the time).
__________________
Robert |
|
24th Nov 2020, 5:27 pm | #57 |
Dekatron
Join Date: Nov 2003
Location: Heckmondwike, West Yorkshire, UK.
Posts: 9,642
|
Re: Would we have got to the moon with no semiconductors?
The development of all the satellites involved in proving the various technologies involved would have been a lot slower using valves. We might have been getting those small steps any time now!
|
24th Nov 2020, 5:43 pm | #58 |
Guest
Posts: n/a
|
Re: Would we have got to the moon with no semiconductors?
I wonder if using an alpha (positively charged particle) source for the "cathode" of a valve would make a PNP one? It would be prohibitively radioactive though and the idea of space charge disappears.
Using valves would only be a matter of size and power consumption no matter how impractical. I am glad to have started this thread, very interesting. |
24th Nov 2020, 6:02 pm | #59 | |
Heptode
Join Date: Oct 2011
Location: Culcheth, Cheshire, UK.
Posts: 654
|
Re: Would we have got to the moon with no semiconductors?
Quote:
The technical advances needed to achieve this political feat would have still happened, even if the Apollo computer turned out the size of a refrigerator. |
|
24th Nov 2020, 6:05 pm | #60 |
Dekatron
Join Date: Aug 2003
Location: near Reading (and sometimes Torquay)
Posts: 3,099
|
Re: Would we have got to the moon with no semiconductors?
I think people unfairly compare old valve technology with modern semiconductors. If there had been no prospect of semiconductors then it is clear that the massive research effort that semiconductors got would have gone into valve systems - and we would now have Pentium chips using valve technology.
The tiny valves used for the proximity fuses were a product of immediately available WW2 technology. Those valves were old by the end of the war. The next big thing would have been field-emission valves which would have done away with heaters*. But I think that evolution would have been a little slower on digital systems because of the difficulty of making the P-type valve. But perhaps this would have resulted in some entirely novel circuit designs that we now just do not need. BTW I think the "vacuum of space" is generally not a very good vacuum as a valve would see it. * See "Contamination meter No.1" to see just what can be done with heater-less valves! |