Author Topic: EEVBlog #261 Marconi  (Read 12990 times)

0 Members and 1 Guest are viewing this topic.

Offline tecman

  • Frequent Contributor
  • **
  • Posts: 434
  • Country: us
EEVBlog #261 Marconi
« on: March 22, 2012, 05:53:33 pm »
Dave, you're slipping. 

In the power supply, the metal band on the transformer is not mu-metal, but copper.  Much more effcetive than mu-metal at high freqs.  Also you pointed to the "output caps, but they are actually the input caps.  Note the 250 volt ratings !

paul
 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 15068
  • Country: za
Re: EEVBlog #261 Marconi
« Reply #1 on: March 22, 2012, 07:12:02 pm »
Dave, do you want some of the optional output protection and conversion adaptor units. I have some that are unused, and will not be using them really. They use a 50ma wire in fuse that does introduce a slight loss, but which protects from feeding back RF into the unit. Cheaper than that output protection built in, and saves you from oopsies.  Converts from N to BNC and such.
 

Offline ei6iz

  • Newbie
  • Posts: 1
Re: EEVBlog #261 Marconi
« Reply #2 on: March 22, 2012, 11:05:27 pm »
Dave,

The 'penetrators' between the lower and upper sides of the RF board are decoupling caps (possibly also with some series inductance ) to keep stray noise out of the RF bits (& stray RF out of everything else  )

The copper tube is Microwave hard-line, used quite a bit in RF test gear. it's big advantage here is that the screen (being solid copper) is superb. Coax, even double screened and semi-rigid types leak too much for this kind of application. 

there are so many screws , RF gaskets and bits of finger stock required because by the time you get to 1GHz a couple of Cm is a significant distance in terms of RF wavelength and RF Impedance to ground climbs very rapidly with distance. an accurate Dynamic range of around 150dB over the full range of 9Khz to 1.2GHz is quite the feat of RF engineering and as a result detail matters ;-)   

I have to say, I'm a fan of Marconi Instruments RF gear (& indeed the racal stuff too)

If you want a decent 10MHz external frequency reference to drive this, look for a 'thunderbolt' on ebay.
Rubidium units are not so good as they can have fairly poor phase noise  performance. The 'Tbolt' locks to GPS so you get the stability along with decent xtal oscillator phase noise  specs. price is right too.. 

If you can get your hands on one, you might wish to do a tear-down on a Racal RA17 receiver. They were the no expense spared pinnacle of Britsh 'hollow state' (tube) Radio receiver design from the 1950's, not bettered in terms of RF performance by solid state until around the mid 1980's 
Takes a bit of work to keep em running like new but when they are in good shape they hold their own in terms of performance.
There are a lot of similarities to the modular design approach and attention to RF details and it would provide an interesting comparison to the modern high end stuff.


regards
Brendan EI6IZ   
 

Offline rolycat

  • Super Contributor
  • ***
  • Posts: 1090
  • Country: gb
Re: EEVBlog #261 Marconi
« Reply #3 on: March 23, 2012, 12:19:17 am »
This was painful to watch.

The manufacturing date of this beautiful signal generator coincides neatly with the last days of GEC under Arnold Weinstock, who innocently handed over the company which he had spent over thirty years building into an engineering giant to the egregious George Simpson in 1996.

Within five years Simpson and his accomplice John Mayo wrecked one of the world's great manufacturing companies with a repulsive mixture of greed and arrogance, backed by the financial Masters of the Universe who, as an encore, did their utmost to destroy the economy of the entire planet.

There's not much electronics left that isn't made in China; almost nothing that's "Made in England", but maybe with inspiration from the boundless enthusiasm of people like Dave we can reclaim some vestige of occidental engineering excellence and start building things of real value again.



 








 

Offline gregariz

  • Frequent Contributor
  • **
  • Posts: 547
  • Country: us
Re: EEVBlog #261 Marconi
« Reply #4 on: March 23, 2012, 01:35:17 am »
Actually the tube is semi-rigid. Hardline is more akin to something like heliax. The penetrators are feed-thru caps/filters as mentioned. What your seeing is a cross between UHF and microwave construction. You wouldnt be able to get away with as much lumped component design in a higher microwave application. What strikes me is that apart from the uC control circuitry, not much has changed in design terms.
 

Online vk6zgo

  • Super Contributor
  • ***
  • Posts: 4794
  • Country: au
Re: EEVBlog #261 Marconi
« Reply #5 on: March 23, 2012, 01:40:50 am »
I want one!!

British Engineering at its best!--Although they did have to put in a bit of quirky design with the cutouts that don't allow you to reach the screws!

Dear old "Macaroni" made some marvellous stuff,as well as some "not so marvellous",& some downright evil!
They were never quite as consistent as HP with their RF stuff,but even the "evil" stuff worked ,& was pretty reliable.
The top level stuff like the one reviewed,could hold its own with anything made,anywhere!

The GEC connection--Never had a lot to do with them,but their RF UHF RT links back in the '60s (first generation solid state) were beautifully constructed.
The East-West Broadband microwave system was originally all GEC valve stuff.

The guys who fixed that system also looked after the STL link for ABW2,& quite a few of the original  Marconi modules in the STL were replaced with GEC ones--The Marconi link was pretty much an "orphan".

The TV Transmitters at ABW2 were Marconi,& remained in service from about 1960 to 1990----not a bad run!

We had an IFR spectrum Analyser at my last job.(Made in South Korea)
It was a very nice unit,but had a few bugs.
One was that a carrier which changed frequency fairly rapidly appeared as multiple carriers---not good when you are looking for a PLL fault.
The other was that you could control it from a PC,but the only way to send displays to the PC was by using a floppy disc---& this thing was new in about 2005!
 

Offline Kozmyk

  • Regular Contributor
  • *
  • Posts: 122
  • Country: wales
Re: EEVBlog #261 Marconi
« Reply #6 on: March 23, 2012, 05:36:17 am »
Hey Dave! That's? a neat little power driver you're using in the vid. What make is it?
 

Online EEVblog

  • Administrator
  • *****
  • Posts: 29483
  • Country: au
    • EEVblog
Re: EEVBlog #261 Marconi
« Reply #7 on: March 23, 2012, 05:38:47 am »
Hey Dave! That's? a neat little power driver you're using in the vid. What make is it?

Just a One Hung Low brand crap cordless screwdriver from Bunnings.

Dave.
 

Offline Kozmyk

  • Regular Contributor
  • *
  • Posts: 122
  • Country: wales
Re: EEVBlog #261 Marconi
« Reply #8 on: March 23, 2012, 06:04:01 am »
Just a One Hung Low brand crap cordless screwdriver from Bunnings.

Dave.
Cheers.
I hadn't noticed them getting so small.
Now that they have, and Li-Ion too, I think I'll get me one.
Bosch IXO4 looks nice with it's Easy Access adapters 3Nm £39 ...
just noticed that the Black&Decker KC460LN has more torques though  5-7Nm £26 ...
Ho hum decisions decisions.  ;)
 

Offline Bored@Work

  • Super Contributor
  • ***
  • Posts: 3932
  • Country: 00
Re: EEVBlog #261 Marconi
« Reply #9 on: March 23, 2012, 06:07:07 am »
British Engineering at its best!--Although they did have to put in a bit of quirky design with the cutouts that don't allow you to reach the screws!

The mechanical engineer was probably coming from the British automobile industry :)
I delete PMs unread. If you have something to say, say it in public.
For all else: Profile->[Modify Profile]Buddies/Ignore List->Edit Ignore List
 

Online EEVblog

  • Administrator
  • *****
  • Posts: 29483
  • Country: au
    • EEVblog
Re: EEVBlog #261 Marconi
« Reply #10 on: March 23, 2012, 06:11:48 am »
I hadn't noticed them getting so small.
Now that they have, and Li-Ion too, I think I'll get me one.
Bosch IXO4 looks nice with it's Easy Access adapters 3Nm £39 ...
just noticed that the Black&Decker KC460LN has more torques though  5-7Nm £26 ...
Ho hum decisions decisions.  ;)

I wanted the Bosch, but found it didn't have variable torque settings, so got the OneHungLow instead.
Turns out the OneHungLow's lowest slip torque setting is still too high for the job of screwing the board on my uCurrent box  :(

Dave.
 

Offline Bored@Work

  • Super Contributor
  • ***
  • Posts: 3932
  • Country: 00
Re: EEVBlog #261 Marconi
« Reply #11 on: March 23, 2012, 06:31:22 am »
This was painful to watch.

The manufacturing date of this beautiful signal generator coincides neatly with the last days of GEC under Arnold Weinstock, who innocently handed over the company which he had spent over thirty years building into an engineering giant to the egregious George Simpson in 1996.

Within five years Simpson and his accomplice John Mayo wrecked one of the world's great manufacturing companies with a repulsive mixture of greed and arrogance, backed by the financial Masters of the Universe who, as an encore, did their utmost to destroy the economy of the entire planet.

I looked at
http://en.wikipedia.org/wiki/Marconi_Company
http://en.wikipedia.org/wiki/Marconi_Communications
http://en.wikipedia.org/wiki/Marconi_Electronic_Systems and
http://en.wikipedia.org/wiki/General_Electric_Company_plc

It is hard to follow how often Marconi was merged, demerged, split and reorganized, and where "the real Marconi" did end up.

Quote
There's not much electronics left that isn't made in China; ",

Looking at all the mess Marconi went through maybe one big mistake was to favour national, British solutions over international, heaven forbid, European, ones. At a time when China was already gearing up and was preparing to steal the lunch of every electronic manufacturer in the world.
I delete PMs unread. If you have something to say, say it in public.
For all else: Profile->[Modify Profile]Buddies/Ignore List->Edit Ignore List
 

Online vk6zgo

  • Super Contributor
  • ***
  • Posts: 4794
  • Country: au
Re: EEVBlog #261 Marconi
« Reply #12 on: March 23, 2012, 07:09:27 am »
I hadn't noticed them getting so small.
Now that they have, and Li-Ion too, I think I'll get me one.
Bosch IXO4 looks nice with it's Easy Access adapters 3Nm £39 ...
just noticed that the Black&Decker KC460LN has more torques though  5-7Nm £26 ...
Ho hum decisions decisions.  ;)

I wanted the Bosch, but found it didn't have variable torque settings, so got the OneHungLow instead.
Turns out the OneHungLow's lowest slip torque setting is still too high for the job of screwing the board on my uCurrent box  :(

Dave.

The old "Wrist Mk1" will probably still do it! ;D
 

Offline Kozmyk

  • Regular Contributor
  • *
  • Posts: 122
  • Country: wales
Re: EEVBlog #261 Marconi
« Reply #13 on: March 23, 2012, 07:25:26 am »
Pologize in advance for the departure from topic.  ;)

When yer MkI Wrist is as decrepit as mine you save it for more delicate work.  ::)
 

Offline rolycat

  • Super Contributor
  • ***
  • Posts: 1090
  • Country: gb
Re: EEVBlog #261 Marconi
« Reply #14 on: March 23, 2012, 07:39:25 am »

It is hard to follow how often Marconi was merged, demerged, split and reorganized, and where "the real Marconi" did end up.


Indeed. By 1996 GEC had swallowed a lot of other famous names, including big chunks of Ferranti, the creators of the world's first commercial computer and more importantly the rather splendid ZN414 radio-in-a-TO-18-can which was one of my very first electronics projects as a schoolboy.

Ferranti also made the early ULAs used by 80s computers like the Acorn BBC micro and the Sinclair Spectrum - more examples of good engineering vanquished by short-sighted management - although of course Acorn gave birth to ARM, who by playing well with the international semiconductor giants are one of the few remaining British success stories in electronics.

I guess its pointless to get misty-eyed over this ancient history. Time to get back to designing new stuff, and maybe hunting down great test gear like the 2023...



 

Offline Kozmyk

  • Regular Contributor
  • *
  • Posts: 122
  • Country: wales
Re: EEVBlog #261 Marconi
« Reply #15 on: March 23, 2012, 08:44:59 am »
I wanted the Bosch, but found it didn't have variable torque settings, so got the OneHungLow instead.
Turns out the OneHungLow's lowest slip torque setting is still too high for the job of screwing the board on my uCurrent box  :(

Dave.
For baby torque jobs I got one of these from HobbyKing.
http://www.hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=7031
It's uncalibrated but there is a scale on it so I cal'ed it myself using a milk bottle of water on the end of a lever.
That was fun ... filling up the water until the clutch slipped then working out the torque each time.
Jeez I hope I haven't lost the result sheet or I'll have to do it all over again  :o
 

Offline Ed.Kloonk

  • Frequent Contributor
  • **
  • Posts: 728
  • Country: au
Re: EEVBlog #261 Marconi
« Reply #16 on: March 23, 2012, 10:54:07 am »
Pardon me barging into this thread but what price tag would such a piece of kit have had on it?

Who would have bought one and for what purpose?

Oh and Dave, I too exclaimed "phworrrrrr" when the lid came off the RF board.

 :)
 

Offline Ed.Kloonk

  • Frequent Contributor
  • **
  • Posts: 728
  • Country: au
Re: EEVBlog #261 Marconi
« Reply #17 on: March 23, 2012, 10:58:45 am »
I nearly forgot to ask what is the beryllium used for?
 

Offline Bored@Work

  • Super Contributor
  • ***
  • Posts: 3932
  • Country: 00
Re: EEVBlog #261 Marconi
« Reply #18 on: March 23, 2012, 12:19:52 pm »
Pardon me barging into this thread but what price tag would such a piece of kit have had on it?

Some of these strange "used test equipment sites" claim the list price was once USD 5775. But one can't really trust these sites.

Quote
Who would have bought one and for what purpose?

That is a little bit like asking who buys screwdrivers and for what purpose. Because such a generator wasn't build for a dedicated purpose. There are other types of generators build for special purposes, e.g. TV pattern generators. But not this one.

Therefore the answer is those who need one for generating the kind of signal the generator can generate. That might be an engineer working on some communication system using such frequencies, or a scientist needing such signals for some experiment.
I delete PMs unread. If you have something to say, say it in public.
For all else: Profile->[Modify Profile]Buddies/Ignore List->Edit Ignore List
 

Online Psi

  • Super Contributor
  • ***
  • Posts: 7241
  • Country: nz
Re: EEVBlog #261 Marconi
« Reply #19 on: March 23, 2012, 12:22:58 pm »
I nearly forgot to ask what is the beryllium used for?

It's got really good thermal conductivity for a non-metal.
Greek letter 'Psi' (not Pounds per Square Inch)
 

Offline Ed.Kloonk

  • Frequent Contributor
  • **
  • Posts: 728
  • Country: au
Re: EEVBlog #261 Marconi
« Reply #20 on: March 23, 2012, 12:32:19 pm »
Ah. Thank you.

 :)
 

Offline markus_b

  • Regular Contributor
  • *
  • Posts: 115
  • Country: ch
Re: EEVBlog #261 Marconi
« Reply #21 on: March 23, 2012, 12:40:57 pm »
Pardon me barging into this thread but what price tag would such a piece of kit have had on it?
Some of these strange "used test equipment sites" claim the list price was once USD 5775. But one can't really trust these sites.
If you look on Ebay then you'll see them starting at $300 and going to $2000, depending on the built-in options (e.g. high-precision oscillator).
Markus

A good scientist is a person with original ideas. A good engineer is a person who makes a design that works with as few original ideas as possible.
 

Online vk6zgo

  • Super Contributor
  • ***
  • Posts: 4794
  • Country: au
Re: EEVBlog #261 Marconi
« Reply #22 on: March 23, 2012, 12:56:17 pm »
I wanted the Bosch, but found it didn't have variable torque settings, so got the OneHungLow instead.
Turns out the OneHungLow's lowest slip torque setting is still too high for the job of screwing the board on my uCurrent box  :(

Dave.
For baby torque jobs I got one of these from HobbyKing.
http://www.hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=7031
It's uncalibrated but there is a scale on it so I cal'ed it myself using a milk bottle of water on the end of a lever.
That was fun ... filling up the water until the clutch slipped then working out the torque each time.
Jeez I hope I haven't lost the result sheet or I'll have to do it all over again  :o

Now THAT'S Engineering!
 

Offline kb2vsq

  • Newbie
  • Posts: 1
Re: EEVBlog #261 Marconi
« Reply #23 on: March 23, 2012, 01:30:58 pm »
Dave --- I disagree with your analysis on the top side.  I think that is a multisection attenuator and that's why you have the multicolo(u)r ribbon going from the RF section to the connector -- as you said, it's DC control.

I would liked to have seen more analysis of the RF sections, but hey, can't have too much fun now...

 

Offline rolycat

  • Super Contributor
  • ***
  • Posts: 1090
  • Country: gb
Re: EEVBlog #261 Marconi
« Reply #24 on: March 23, 2012, 02:25:18 pm »
I nearly forgot to ask what is the beryllium used for?

It's got really good thermal conductivity for a non-metal.

Just to be pedantic, although beryllium does indeed have really good thermal conductivity, it is a metal.

I suspect Psi meant beryllia (beryllium oxide), which also has good thermal conductivity but is an electrical insulator.

 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf