Author Topic: A switching power supply in an oscilloscope?  (Read 5738 times)

0 Members and 1 Guest are viewing this topic.

Offline MartiniTopic starter

  • Regular Contributor
  • *
  • Posts: 95
A switching power supply in an oscilloscope?
« on: March 10, 2019, 11:35:20 am »
I'm currently reading Jobs's biography and it says they designed a switch mode power supply like those used in oscilloscopes for the Apple II.

Were scopes using these in the 70's!?
 

Offline tooki

  • Super Contributor
  • ***
  • Posts: 13157
  • Country: ch
Re: A switching power supply in an oscilloscope?
« Reply #1 on: March 10, 2019, 11:41:20 am »
While I actually can't speak to scopes of the era, I can say that Jobs' praise for their SMPS was… exaggerated. He always made it sound as though they were a total innovation, when in fact SMPS's were already in wide use (though not as wide as these days). Apple did innovate a lot over the years, but this instance really is the Reality Distortion Field at work. :P
 

Offline MartiniTopic starter

  • Regular Contributor
  • *
  • Posts: 95
Re: A switching power supply in an oscilloscope?
« Reply #2 on: March 10, 2019, 11:59:06 am »
I was born in 1985 and most power supplies I used when I was young (on my game console and whatnot) were linear. I remember thinking about the progress we had made in the 2000's about the weight and dimensions of the PSU, not knowing what technologies were behind
 

Offline tooki

  • Super Contributor
  • ***
  • Posts: 13157
  • Country: ch
Re: A switching power supply in an oscilloscope?
« Reply #3 on: March 10, 2019, 01:33:10 pm »
Yep, wall warts and the like did change. But larger power supplies had already been switch-mode. Hence why I said that they were already in wide use, but not as wide as today.

Pretty much the only area where large linear power supplies remain common is in better-quality audio gear, where SMPSs can be a bit noisy.
 

Offline Wolfram

  • Frequent Contributor
  • **
  • Posts: 414
  • Country: no
Re: A switching power supply in an oscilloscope?
« Reply #4 on: March 10, 2019, 02:32:06 pm »
I know at least some Tektronix oscilloscopes from that era used a switching power supply. Check out the 7904 for example, which was introduced in 1971. https://lazyelectrons.wordpress.com/2018/11/24/tektronix-7904-oscilloscope-power-supply-repair/
 

Offline IanMacdonald

  • Frequent Contributor
  • **
  • Posts: 943
  • Country: gb
    • IWR Consultancy
Re: A switching power supply in an oscilloscope?
« Reply #5 on: March 10, 2019, 05:39:08 pm »
Many of the early minicomputers such as the PDP11 had SMPS, but they were secondary switching because the power transistors they had in those days didn't handle mains voltages. The switching transistors and rectifiers were germanium, for lower threshold voltage (I guess nobody had thought of the Schottky junction then) Problem with germanium is that if it's heated to more than 50C or so it starts to leak like crazy, so they needed very good cooling. 

The Apollo computer used a similar arrangement, its input being the 28v fuel cell bus.

The BRC 3000 TV chassis was one of the first to use a SMPS, around '74 IIRC, and it was (I am told) based on a PSU from the Lightning military jet. This was primary switching, but since the transistors couldn't handle full rectified 240v mains there was an autotransformer before the rectifier. It was actually a good set for its era, running cooler than most and giving not too much trouble.
« Last Edit: March 10, 2019, 05:44:46 pm by IanMacdonald »
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: A switching power supply in an oscilloscope?
« Reply #6 on: March 10, 2019, 06:09:55 pm »
IIRC the Apple II was the first personal computer to use a switchmode power supply. They were around for a while, but all of the other home/hobby computers of the era had big chunky linear power supplies.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: A switching power supply in an oscilloscope?
« Reply #7 on: March 10, 2019, 07:03:41 pm »
I know at least some Tektronix oscilloscopes from that era used a switching power supply. Check out the 7904 for example, which was introduced in 1971. https://lazyelectrons.wordpress.com/2018/11/24/tektronix-7904-oscilloscope-power-supply-repair/

The 7904 was in the Tektronix 1972 new products catalog but both the 7904 and 7704A were in the 1973 catalog.  However these were the *second* generation of Tektronix oscilloscopes with a switching power supply.

The first was the 7704 released in 1969 which has a completely discrete version of the power supply used in the later models; one of Tektronix's first ICs was the power supply controller used in the 7904/7704A and later instruments.  TekScope Vol. 3 No. 2, Mar 1971 has an article discussing the 1969 power supply and of course details are in the 7704 service manual.  A later article discussed the integrated version used in the 7904/7704A.

The Apple II power supply is a flyback converter unlike the Tektronix series resonant design however both are self oscillating designs where the secondary side regulator blocks the oscillator to control the output.  I was never able to confirm the story discussed here about Holt having designed a switching power supply 10 years earlier for an oscilloscope and found no evidence that he worked for Tektronix or any other oscilloscope manufacturer.  However Steve Jobs said that Ron Holt's switching power supply for the Apple II was *like* one used in an oscilloscope which is consistent with the 7704 design which Holt could have drawn inspiration from.

 
The following users thanked this post: Wolfram

Offline MartiniTopic starter

  • Regular Contributor
  • *
  • Posts: 95
Re: A switching power supply in an oscilloscope?
« Reply #8 on: March 10, 2019, 08:57:12 pm »
Pretty much the only area where large linear power supplies remain common is in better-quality audio gear, where SMPSs can be a bit noisy.
That's why it surprised me they used it in oscilloscopes, I was thinking about the noise induced.


Thanks for the inputs, guys, I learnt something.
 
The following users thanked this post: tooki

Online T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: A switching power supply in an oscilloscope?
« Reply #9 on: March 10, 2019, 10:06:38 pm »
Noise is only due to carelessness -- with a good layout, and with enough filtering and shielding, you can use almost anything.

It helps that early designs were very simple: few transistors, so that the total gain at high frequencies was still only modest.  Transistors weren't terribly fast either, power transistors especially.  This prevents the generation of high harmonics, so that shielding and filtering in the 100MHz+ range isn't needed, or even 20MHz perhaps.  Keeping switching edges low, with resonant designs, always helps, too.

Nowadays, transistors with fT in the GHz are common inside controllers, so that, if nothing else, the controller alone may be responsible for content in the 100s MHz range.  Modern transistors are also freakishly fast, with silicon SuperJunction types having performance in the 100s MHz, and GaN types even faster still!  This requires the designer to be even more careful yet; on the upside, the increase in frequency allows reduction in component size, so that a filter with good stopband attenuation at 100MHz+ doesn't have to be complicated.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline rsjsouza

  • Super Contributor
  • ***
  • Posts: 6106
  • Country: us
  • Eternally curious
    • Vbe - vídeo blog eletrônico
Re: A switching power supply in an oscilloscope?
« Reply #10 on: March 10, 2019, 10:25:52 pm »
Well, Bob Mammano designed the SG1524 during the 1970's and it was released in 1976. That is the staple of switching IC controllers, but it was probably too new (and too expensive) to be adopted in mainstream applications.

I had a 1976 TI30 calculator with a boost converter from 1.2V of its NiCd AA cell to the 9V. Terribly inneficient.

My 1980 TV set from Sharp had switching power supplies as well.
Vbe - vídeo blog eletrônico http://videos.vbeletronico.com

Oh, the "whys" of the datasheets... The information is there not to be an axiomatic truth, but instead each speck of data must be slowly inhaled while carefully performing a deep search inside oneself to find the true metaphysical sense...
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 21226
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: A switching power supply in an oscilloscope?
« Reply #11 on: March 10, 2019, 10:45:12 pm »
I'm currently reading Jobs's biography and it says they designed a switch mode power supply like those used in oscilloscopes for the Apple II.

Were scopes using these in the 70's!?

Yes.

Both to convert from mains voltages to the obvious low voltages, and also to get the high voltages (2.5kV to 20kV) for the CRT.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 21226
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: A switching power supply in an oscilloscope?
« Reply #12 on: March 10, 2019, 10:54:27 pm »
I had a 1976 TI30 calculator with a boost converter from 1.2V of its NiCd AA cell to the 9V. Terribly inneficient.

The earlier groundbreaking HP35 generated 8.2V [for the cathode LED driver], 6V [for the clock driver, the control & timing, the ROMs and the arithmetic registers] and -12V [for the clock driver, the control & timing, the ROMs and the arithmetic registers] from three NiCd cells.
http://home.citycable.ch/pierrefleur/Jacques-Laporte/HP35%20power%20unit.htm

There were many other calculators that had Panaplex displays needing >150V; some of them were portable.

There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 29810
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: A switching power supply in an oscilloscope?
« Reply #13 on: March 10, 2019, 11:16:34 pm »
I'm currently reading Jobs's biography and it says they designed a switch mode power supply like those used in oscilloscopes for the Apple II.

Were scopes using these in the 70's!?
While many were still using linear PSU's for the general circuitry voltages obtaining the higher EHT used on scopes from 70's onward a Colpitts Oscillator was often used on the primary side of EHT generation.
While they are not what we generally now consider as SMPS, they in essence perform the same task, switching a DC voltage to some form of AC to then be transformed to a higher voltage which then rectified becomes the DC EHT that CRO's need for CRT voltages.
In a CRO I had 30V was transformed in this way to -2750V for the CRT and also quintupled for the ~ +12 KV PDA voltages.
This scope was a UK D83 Telequipment which at that time in the early 70's was owned by Tek.
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: A switching power supply in an oscilloscope?
« Reply #14 on: March 11, 2019, 01:01:00 am »
I'm currently reading Jobs's biography and it says they designed a switch mode power supply like those used in oscilloscopes for the Apple II.

Were scopes using these in the 70's!?

While many were still using linear PSU's for the general circuitry voltages obtaining the higher EHT used on scopes from 70's onward a Colpitts Oscillator was often used on the primary side of EHT generation.

While they are not what we generally now consider as SMPS, they in essence perform the same task, switching a DC voltage to some form of AC to then be transformed to a higher voltage which then rectified becomes the DC EHT that CRO's need for CRT voltages.
In a CRO I had 30V was transformed in this way to -2750V for the CRT and also quintupled for the ~ +12 KV PDA voltages.
This scope was a UK D83 Telequipment which at that time in the early 70's was owned by Tek.

I was actually surprised to see any oscilloscopes use a 50/60Hz transformer for the high voltages but this was still common with the lessor manufacturers.  As far as I know, all Tektronix oscilloscopes which used a linear power supply after their first (511) one generated their high voltage as you describe with a separate high frequency inverter.  The 511A used a 2kHz inverter for the high CRT voltages so Tektronix changed this immediately.


 

Offline tooki

  • Super Contributor
  • ***
  • Posts: 13157
  • Country: ch
Re: A switching power supply in an oscilloscope?
« Reply #15 on: March 11, 2019, 05:35:55 pm »
IIRC the Apple II was the first personal computer to use a switchmode power supply. They were around for a while, but all of the other home/hobby computers of the era had big chunky linear power supplies.
Yes, possibly the first personal computer. It's just that Jobs usually worded it as though their hardware engineer had invented something brand new, not simply newly applying it to a particular kind of product.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: A switching power supply in an oscilloscope?
« Reply #16 on: March 11, 2019, 07:16:45 pm »
Well that was Jobs for you, he had some personality traits that I find annoying, although I will agree that the Apple II was a revolutionary machine for many reasons. I never thought much of Steve Jobs when he was alive, but the quality of Apple products took a serious nose dive after he died. My current iPhone is unbelievably buggy.
 

Offline particleman

  • Regular Contributor
  • *
  • Posts: 115
Re: A switching power supply in an oscilloscope?
« Reply #17 on: March 11, 2019, 07:22:57 pm »
Tektronix 485 which was introduced in 1972 also had a SMPS.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: A switching power supply in an oscilloscope?
« Reply #18 on: March 12, 2019, 03:19:52 am »
Tektronix 485 which was introduced in 1972 also had a SMPS.

The 485 used the same topology and switching regulator controller as the 7904/7704A.  So did the late 434.  The early 434 released in 1972 used a discrete off-line switching regulator which I suspect was designed by the Elder Gods; a saving throw against sanity is required to understand it.

The 400 series portable oscilloscopes introduced after the 485 (except the 434) used linear power supplies so I assume this was still more economical when higher power density was not required.

The 468 which required much more power for its digital storage system included just a buck switching regulator for its +5 volt digital supply.  The 7000 series oscilloscope plug-ins which required high power for their digital circuits also included switching regulators but this was more to make up for power distribution limitations in the 7000 mainframes which could not directly supply high power logic in the plug-ins.
 

Offline helius

  • Super Contributor
  • ***
  • Posts: 3688
  • Country: us
Re: A switching power supply in an oscilloscope?
« Reply #19 on: March 12, 2019, 03:42:30 am »
As far as desktop computers, the Datapoint 2200 sold in 1970 used a switching power supply. That was 7 years before the Apple II.
 

Online Zero999

  • Super Contributor
  • ***
  • Posts: 20357
  • Country: gb
  • 0999
Re: A switching power supply in an oscilloscope?
« Reply #20 on: March 12, 2019, 08:54:31 am »
I was born in 1985 and most power supplies I used when I was young (on my game console and whatnot) were linear. I remember thinking about the progress we had made in the 2000's about the weight and dimensions of the PSU, not knowing what technologies were behind
The games consoles of the 70s to early 90s period used relatively little power and back then, switched mode power supplies were only economical at higher power levels. For small power supplies, under 20W or so, it was much cheaper to use a chunky transformer, than expensive semiconductors. In the late 90s, games consoles became more powerful and went to CD-ROMS, which also used more power, switched mode power supplies also became cheaper, so it made sense to move to them.

Nowadays almost all mains power supplies are switched mode, because it's cheap and many countries have environmental legislation which sets limits on minimum efficiency, that can only be met by a switched mode power supply.
 

Offline tooki

  • Super Contributor
  • ***
  • Posts: 13157
  • Country: ch
Re: A switching power supply in an oscilloscope?
« Reply #21 on: March 12, 2019, 06:21:55 pm »
As far as desktop computers, the Datapoint 2200 sold in 1970 used a switching power supply. That was 7 years before the Apple II.
Yes, though I think it'd be a stretch to consider that a "personal computer", in the sense that it was marketed as a mainframe accessory, not a computer for consumers. It does play a neat part of computing history, though!!!
 

Offline helius

  • Super Contributor
  • ***
  • Posts: 3688
  • Country: us
Re: A switching power supply in an oscilloscope?
« Reply #22 on: March 12, 2019, 07:47:36 pm »
Certainly not a computer for consumers.
"Prices for the Datapoint 2200 begin at $6040 with a variety of lease and purchase plans, with worldwide maintenance available."
That's $39,500 in 2018 money.
 
The following users thanked this post: tooki

Offline rsjsouza

  • Super Contributor
  • ***
  • Posts: 6106
  • Country: us
  • Eternally curious
    • Vbe - vídeo blog eletrônico
Re: A switching power supply in an oscilloscope?
« Reply #23 on: March 12, 2019, 08:42:48 pm »
I was born in 1985 and most power supplies I used when I was young (on my game console and whatnot) were linear. I remember thinking about the progress we had made in the 2000's about the weight and dimensions of the PSU, not knowing what technologies were behind
The games consoles of the 70s to early 90s period used relatively little power and back then, switched mode power supplies were only economical at higher power levels. For small power supplies, under 20W or so, it was much cheaper to use a chunky transformer, than expensive semiconductors. In the late 90s, games consoles became more powerful and went to CD-ROMS, which also used more power, switched mode power supplies also became cheaper, so it made sense to move to them.
To me the first striking evidence of the changing times was the advent of switching power supplies used in the Nokia cellphone chargers of the mid to late 1990s (Nokia 6110?). It was certainly not the first switching power supply I have seen, but it was the first time the small bricks were quite liberating given their lightweight and universal input voltage - quite suitable for such sleek portable device.
Vbe - vídeo blog eletrônico http://videos.vbeletronico.com

Oh, the "whys" of the datasheets... The information is there not to be an axiomatic truth, but instead each speck of data must be slowly inhaled while carefully performing a deep search inside oneself to find the true metaphysical sense...
 

Online Zero999

  • Super Contributor
  • ***
  • Posts: 20357
  • Country: gb
  • 0999
Re: A switching power supply in an oscilloscope?
« Reply #24 on: March 12, 2019, 09:01:09 pm »
I was born in 1985 and most power supplies I used when I was young (on my game console and whatnot) were linear. I remember thinking about the progress we had made in the 2000's about the weight and dimensions of the PSU, not knowing what technologies were behind
The games consoles of the 70s to early 90s period used relatively little power and back then, switched mode power supplies were only economical at higher power levels. For small power supplies, under 20W or so, it was much cheaper to use a chunky transformer, than expensive semiconductors. In the late 90s, games consoles became more powerful and went to CD-ROMS, which also used more power, switched mode power supplies also became cheaper, so it made sense to move to them.
To me the first striking evidence of the changing times was the advent of switching power supplies used in the Nokia cellphone chargers of the mid to late 1990s (Nokia 6110?). It was certainly not the first switching power supply I have seen, but it was the first time the small bricks were quite liberating given their lightweight and universal input voltage - quite suitable for such sleek portable device.
Some of them were pretty crappy, with no voltage regulation at all, just an oscillator driving a transformer open loop. I remember cracking one open and finding a very crude blocking oscillator inside. It did the job though and still used less power than an bulky old mains transformer.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf