Test Equipment Anonymous (TEA) group therapy thread

[Specmaster]

<snip>
I remember a professor at Imperial College, who also did some industrial design gigs at an art and design college.

He pointed out that if you asked them to design an egg, they would all create something which was the right size and shape. Better designers' eggs would have the right coloration. The best designers' eggs would have the right weight. None of them would have anything inside.

I'll feed my belly before I feed my eye

Don't know if you know this or not, but did you know that Imperial College has service tunnels that run underneath Exhibition Road to the other departments opposite? Another place that has such tunnels is Harrods in Knightsbridge?

[tggzzz]

I remember a professor at Imperial College, who also did some industrial design gigs at an art and design college.

He pointed out that if you asked them to design an egg, they would all create something which was the right size and shape. Better designers' eggs would have the right coloration. The best designers' eggs would have the right weight. None of them would have anything inside.

I'll feed my belly before I feed my eye

That's just the old engineers' canard about design being just about appearances. It isn't. Good design includes ergonomics, suitability to task, and also appearances.

It's a bit like picking a partner. Sure, a pleasing appearance is nice to have, but having a good heart and temperament, and being nice to hold is more important, and great cooking is a plus. But great cooking with a face like a bag of spanners, the temperament of a rattlesnake with a hangover, and the tactile qualities of a leather bag of oobleck is what you'd get if the design task is left to most engineers.

Not always. Modern gliders are the sexiest things in the air. Ditto racing eights on the water.

Designed by physics, thus nature, not by engineers.

Nah - unless you count these as also being "designed by physics", and thereby destroy your contention

The second, a T21, are affectionately known as "barges" due to the shape of their fuselage. They are marginally more aerodynamic than bricks, but I have seen one do a loop over the runway at ~500ft. Must have scared teh bajeesus out of the woodworm

[tggzzz]

<snip>
I remember a professor at Imperial College, who also did some industrial design gigs at an art and design college.

He pointed out that if you asked them to design an egg, they would all create something which was the right size and shape. Better designers' eggs would have the right coloration. The best designers' eggs would have the right weight. None of them would have anything inside.

I'll feed my belly before I feed my eye

Don't know if you know this or not, but did you know that Imperial College has service tunnels that run underneath Exhibition Road to the other departments opposite? Another place that has such tunnels is Harrods in Knightsbridge?

I didn't, but it doesn't surprise me. I've used the tunnel from the Science Museum to South Ken station, and tunnels under the road in the nearby main hospital

I did use the paternoster lift at IC, too many decades ago.

[BU508A]

There is only one true design concept: Form follows function!

That's idiomatic in a purely mechanical world. It's always going to be true of a good spanner, and the inverse "form indicates function" is a natural consequence.

I cannot see what is wrong with that. We all live in a physical world, are part of it and have to obey the laws of physics. These laws form the framework of how something has to function. In addition, there is always a drive to maximise efficiency or, in other words, to minimise the use of energy. In my opinion, the approach "form follows function" follows exactly this premise.

But that all rather falls flat on its face when faced with our favourite subject, test gear. The function of a frequency counter is a mile away from a signal generator, or scope, or SA, or measurement receiver, etc., but they all end up as rectilinear boxes of roughly the same order of magnitude and highly similar appearance. Indeed it's often necessary to understand the language that the scales and controls are labelled in to understand the function of such a device.

So it would appear that there has to exist more than "one true design concept".

Certainly, more than one design concept exists. But to stay with the TE example: let's take an oscilloscope. Its primary task is to visualise time-varying electrical signals. The two most important parameters are the time axis and the amplitude. This is the main reason why the controls for these two parameters are so prominently placed on most oscilloscopes. Anyone who understands what an oscilloscope does will intuitively understand the importance of these controls and use them accordingly. For me, this is an example of "form follows function". If one deviates from this, there must be very good reasons for this different approach. Hameg oscilloscopes, for example, fulfil this "form follows function" approach quite well (imo). How good the technical capabilities are in detail is certainly debatable, but they solve the task of "displaying time-varying electrical signals" quite well and I don't know how it could be done better.

[Cerebus]

I remember a professor at Imperial College, who also did some industrial design gigs at an art and design college.

He pointed out that if you asked them to design an egg, they would all create something which was the right size and shape. Better designers' eggs would have the right coloration. The best designers' eggs would have the right weight. None of them would have anything inside.

I'll feed my belly before I feed my eye

That's just the old engineers' canard about design being just about appearances. It isn't. Good design includes ergonomics, suitability to task, and also appearances.

It's a bit like picking a partner. Sure, a pleasing appearance is nice to have, but having a good heart and temperament, and being nice to hold is more important, and great cooking is a plus. But great cooking with a face like a bag of spanners, the temperament of a rattlesnake with a hangover, and the tactile qualities of a leather bag of oobleck is what you'd get if the design task is left to most engineers.

While a lot of what you say is true, but when it comes to servicing things like TE, cars etc it is also true that they are easier to service if designed by an engineer because they appreciate the need for easy access to the parts that need to be accessed?

That doesn't argue towards them being designed by an engineer, that argues for that aspect being designed by a technician.

[bd139]

There is only one true design concept: Form follows function!

That's idiomatic in a purely mechanical world. It's always going to be true of a good spanner, and the inverse "form indicates function" is a natural consequence.

I cannot see what is wrong with that. We all live in a physical world, are part of it and have to obey the laws of physics. These laws form the framework of how something has to function. In addition, there is always a drive to maximise efficiency or, in other words, to minimise the use of energy. In my opinion, the approach "form follows function" follows exactly this premise.

But that all rather falls flat on its face when faced with our favourite subject, test gear. The function of a frequency counter is a mile away from a signal generator, or scope, or SA, or measurement receiver, etc., but they all end up as rectilinear boxes of roughly the same order of magnitude and highly similar appearance. Indeed it's often necessary to understand the language that the scales and controls are labelled in to understand the function of such a device.

So it would appear that there has to exist more than "one true design concept".

Certainly, more than one design concept exists. But to stay with the TE example: let's take an oscilloscope. Its primary task is to visualise time-varying electrical signals. The two most important parameters are the time axis and the amplitude. This is the main reason why the controls for these two parameters are so prominently placed on most oscilloscopes. Anyone who understands what an oscilloscope does will intuitively understand the importance of these controls and use them accordingly. For me, this is an example of "form follows function". If one deviates from this, there must be very good reasons for this different approach. Hameg oscilloscopes, for example, fulfil this "form follows function" approach quite well (imo). How good the technical capabilities are in detail is certainly debatable, but they solve the task of "displaying time-varying electrical signals" quite well and I don't know how it could be done better.

It depends. Software is different as you can specify concepts in an abstract manor with no major compromises.

If we applied purism like form follows function absolutely then everyone would have to use unix command line tools to edit their photos.

The nuanced position is to meet the user half way. Real quality software goes out of the way to move the problem as close to the user as possible. That software builds trillion dollar companies.

[mnementh]

      

https://www.amazon.com/dp/B09SP54HNW/

Just a quick heads-up; I found these on slAmazon while hunting stick-on feet for a particular project. They look to me like a good generic substitute for the OEM feet on many different models of TE. At $6/doz, pretty reasonable too.

Cheers!

mnem
yes, I've already posted in the buttons, knobs & feet thread.

[Cerebus]

mnem
When I worked as a mechanic, I would swear that automotive engineers hated us fixers... Many times I imagined them standing around the water cooler showing off their latest blueprints: "No, you see there? If we'd put that one bolt in the hinge half a inch further that way, you'd be able to get a wrench on it. This way, you have to take the fender off before you can take the door off.  MUAHAHAHAHA!!!"

Case in point. The offside headlight automatic height adjust on my car is a bit hit and miss. Diagnosis suggests a motor in the headlight assembly is a bit sticky and a spot of grease is needed, OR the wiring to the local controller on the headlight is a bit intermittent and needs the contacts cleaning, drying or something of that ilk. To get to either of those requires taking off the whole front bumper, wheel arch panel and some sundry covers. Actual fix time 5-10 minutes. Fix time with removing and reinstalling all that crap 2 hours if you've done it before, probably 3 for a first time. Consequence: my local and thankfully very trustworthy grease monkeys are going to make a few quid because I just can't be arsed with all that.

Design by monkeys. Malicious monkeys.

[Cerebus]

I could have used such an app last Saturday, if it could warn me about nerds who can jaw the hind leg of a donkey about nothing in particular of interest and a bullshitter of a name-dropper in the bargain 

What are you talking about? I wasn't there, I was meeting a mate at the Groucho Club who works with Stephen Fry. 

[tggzzz]

There is only one true design concept: Form follows function!

That's idiomatic in a purely mechanical world. It's always going to be true of a good spanner, and the inverse "form indicates function" is a natural consequence.

I cannot see what is wrong with that. We all live in a physical world, are part of it and have to obey the laws of physics. These laws form the framework of how something has to function. In addition, there is always a drive to maximise efficiency or, in other words, to minimise the use of energy. In my opinion, the approach "form follows function" follows exactly this premise.

But that all rather falls flat on its face when faced with our favourite subject, test gear. The function of a frequency counter is a mile away from a signal generator, or scope, or SA, or measurement receiver, etc., but they all end up as rectilinear boxes of roughly the same order of magnitude and highly similar appearance. Indeed it's often necessary to understand the language that the scales and controls are labelled in to understand the function of such a device.

So it would appear that there has to exist more than "one true design concept".

Certainly, more than one design concept exists. But to stay with the TE example: let's take an oscilloscope. Its primary task is to visualise time-varying electrical signals. The two most important parameters are the time axis and the amplitude. This is the main reason why the controls for these two parameters are so prominently placed on most oscilloscopes. Anyone who understands what an oscilloscope does will intuitively understand the importance of these controls and use them accordingly. For me, this is an example of "form follows function". If one deviates from this, there must be very good reasons for this different approach. Hameg oscilloscopes, for example, fulfil this "form follows function" approach quite well (imo). How good the technical capabilities are in detail is certainly debatable, but they solve the task of "displaying time-varying electrical signals" quite well and I don't know how it could be done better.

It depends. Software is different as you can specify concepts in an abstract manor with no major compromises.

In theory. In practice it has a lot in common with archaeology, where you keep on discovering another layer of detritus below the last.

If we applied purism like form follows function absolutely then everyone would have to use unix command line tools to edit their photos.

The nuanced position is to meet the user half way. Real quality software goes out of the way to move the problem as close to the user as possible. That software builds trillion dollar companies.

I do use unix command line tools on photos. Either batch jobs, or converting photos of manuals to pdfs that are 1/20th the size and more readable to boot.

Doing the same operations on hundreds of files using a GUI is something only a hipster could conceive.

[bd139]

There is only one true design concept: Form follows function!

That's idiomatic in a purely mechanical world. It's always going to be true of a good spanner, and the inverse "form indicates function" is a natural consequence.

I cannot see what is wrong with that. We all live in a physical world, are part of it and have to obey the laws of physics. These laws form the framework of how something has to function. In addition, there is always a drive to maximise efficiency or, in other words, to minimise the use of energy. In my opinion, the approach "form follows function" follows exactly this premise.

But that all rather falls flat on its face when faced with our favourite subject, test gear. The function of a frequency counter is a mile away from a signal generator, or scope, or SA, or measurement receiver, etc., but they all end up as rectilinear boxes of roughly the same order of magnitude and highly similar appearance. Indeed it's often necessary to understand the language that the scales and controls are labelled in to understand the function of such a device.

So it would appear that there has to exist more than "one true design concept".

Certainly, more than one design concept exists. But to stay with the TE example: let's take an oscilloscope. Its primary task is to visualise time-varying electrical signals. The two most important parameters are the time axis and the amplitude. This is the main reason why the controls for these two parameters are so prominently placed on most oscilloscopes. Anyone who understands what an oscilloscope does will intuitively understand the importance of these controls and use them accordingly. For me, this is an example of "form follows function". If one deviates from this, there must be very good reasons for this different approach. Hameg oscilloscopes, for example, fulfil this "form follows function" approach quite well (imo). How good the technical capabilities are in detail is certainly debatable, but they solve the task of "displaying time-varying electrical signals" quite well and I don't know how it could be done better.

It depends. Software is different as you can specify concepts in an abstract manor with no major compromises.

In theory. In practice it has a lot in common with archaeology, where you keep on discovering another layer of detritus below the last.

If we applied purism like form follows function absolutely then everyone would have to use unix command line tools to edit their photos.

The nuanced position is to meet the user half way. Real quality software goes out of the way to move the problem as close to the user as possible. That software builds trillion dollar companies.

I do use unix command line tools on photos. Either batch jobs, or converting photos of manuals to pdfs that are 1/20th the size and more readable to boot.

Doing the same operations on hundreds of files using a GUI is something only a hipster could conceive.

Hmm... never owned a Mac then. It's far far far more powerful than any of the command line tools.



You can automate most things, including PDF extraction, rerendering, annotation etc. On a schedule, on an event, when a file is dropped somewhere or via voice.

My favourite automation at the moment is this one, which works transparently on my phone, ipad and mac (and probably the watch - haven't tried it yet)...



It can SSH to stuff, run shell commands as well...

Edit: imagine if everyone gave enough of a shit about their user interface to this level.

[mnementh]

Certainly, more than one design concept exists. But to stay with the TE example: let's take an oscilloscope. Its primary task is to visualise time-varying electrical signals. The two most important parameters are the time axis and the amplitude. This is the main reason why the controls for these two parameters are so prominently placed on most oscilloscopes.

Anyone who understands what an oscilloscope does will intuitively understand the importance of these controls and use them accordingly. For me, this is an example of "form follows function". If one deviates from this, there must be very good reasons for this different approach. Hameg oscilloscopes, for example, fulfil this "form follows function" approach quite well (imo). How good the technical capabilities are in detail is certainly debatable, but they solve the task of "displaying time-varying electrical signals" quite well and I don't know how it could be done better.

It depends. Software is different as you can specify concepts in an abstract manor with no major compromises.

If we applied purism like form follows function absolutely then everyone would have to use unix command line tools to edit their photos.

The nuanced position is to meet the user half way. Real quality software goes out of the way to move the problem as close to the user as possible. That software builds trillion dollar companies.

UI design is the tough nut; the bugbear. With some tools it is obvious how it needs to be made because they have to a) do a specific job and b) fit the human hand, so the form refines itself over the years... the required form of tools used by the mind are much more subjective.

This is why I may bitch about the offensively intrusive hoops one has to jump through to use Frustion360 (without paying a bloody fortune for the privilege), but I keep using it. Because somebody bothered to do at least half of that last 20%, so it's the solution that still      me least often.

Under any circumstances; the most important part of that last 20% is not so much what is there as what isn't; the old sculptor's adage of "You start with a chunk of [insert chosen material here] and take away everything that isn't [insert desired object here]..." is so very important.

When I'm doing any job, as I approach completion, I'll take a few moments to survey the result... and it is always the first thing (usually more like a dozen things ) I notice that is what needs to be taken care of before I'm truly done.

I wish more stuff in everyday life received even one short hour of such consideration before being inflicted upon a unsuspecting public at large.   

If even half the things we have to use everyday got half a hour of that sort of consideration, we would save man-years a day of wasted time and frustration.

mnem
 

[mnementh]

...Design by monkeys. Malicious monkeys.

I think I have a new favorite expletive.   

mnem
BWAHAHAHAHAHA!!!

[bd139]

Anyway some TE content. Have reverse engineered the next two gain phases of the TM6 voltmeter. They are fairly simple programmable gain AC amplifiers with a whole bunch of frequency compensation.

Now for the synchronous detector... 

[BU508A]

There is only one true design concept: Form follows function!

That's idiomatic in a purely mechanical world. It's always going to be true of a good spanner, and the inverse "form indicates function" is a natural consequence.

I cannot see what is wrong with that. We all live in a physical world, are part of it and have to obey the laws of physics. These laws form the framework of how something has to function. In addition, there is always a drive to maximise efficiency or, in other words, to minimise the use of energy. In my opinion, the approach "form follows function" follows exactly this premise.

But that all rather falls flat on its face when faced with our favourite subject, test gear. The function of a frequency counter is a mile away from a signal generator, or scope, or SA, or measurement receiver, etc., but they all end up as rectilinear boxes of roughly the same order of magnitude and highly similar appearance. Indeed it's often necessary to understand the language that the scales and controls are labelled in to understand the function of such a device.

So it would appear that there has to exist more than "one true design concept".

Certainly, more than one design concept exists. But to stay with the TE example: let's take an oscilloscope. Its primary task is to visualise time-varying electrical signals. The two most important parameters are the time axis and the amplitude. This is the main reason why the controls for these two parameters are so prominently placed on most oscilloscopes. Anyone who understands what an oscilloscope does will intuitively understand the importance of these controls and use them accordingly. For me, this is an example of "form follows function". If one deviates from this, there must be very good reasons for this different approach. Hameg oscilloscopes, for example, fulfil this "form follows function" approach quite well (imo). How good the technical capabilities are in detail is certainly debatable, but they solve the task of "displaying time-varying electrical signals" quite well and I don't know how it could be done better.

It depends. Software is different as you can specify concepts in an abstract manor with no major compromises.

This is both a curse and a blessing in the case of software.
But here too, imo, the concept of efficiency or minimal use of resources applies. The resources in this case can be, for example: Mouse clicks, number of submenus, etc. In the case of software, it boils down to two things: speed (you want to have a task done as quickly as possible) and ergonomics of use. If, as with gimp for example, you first have to take a course to do simple things like choosing the type of tool or inserting text in the right place with the correct height, etc., then this is not necessarily "form follows function". I used to work with CorelDraw years ago. In terms of ergonomics and ultimately efficiency, CorelDraw was far superior to gimp. I still don't understand why I have to choose "Export" when I want to save a file as .png or .jpg instead of "Save as".

If we applied purism like form follows function absolutely then everyone would have to use unix command line tools to edit their photos.

Unix command line tools are great for automating things. However, the command line fails when you want to specifically change something in a picture, for example. Example: the Sisytek picture on the cup. The tools have to fit the task and should be designed to do that task as well as possible. Otherwise, every problem looks like a nail and every solution like a hammer. And I'm not talking about that bloke who parks his cart in the living room. 

The nuanced position is to meet the user half way. Real quality software goes out of the way to move the problem as close to the user as possible. That software builds trillion dollar companies.

Agreed. But I would like to add: "Form follows function" does not mean that there is only one valid way to solve a task. But this approach helps to find the best or most optimal solutions. Whereby, of course, it has to be clarified again what the best or most optimal is.

[factory]

Another one missed, been a lot of posts in the last five hours.



David

[Specmaster]

mnem
When I worked as a mechanic, I would swear that automotive engineers hated us fixers... Many times I imagined them standing around the water cooler showing off their latest blueprints: "No, you see there? If we'd put that one bolt in the hinge half a inch further that way, you'd be able to get a wrench on it. This way, you have to take the fender off before you can take the door off.  MUAHAHAHAHA!!!"

Case in point. The offside headlight automatic height adjust on my car is a bit hit and miss. Diagnosis suggests a motor in the headlight assembly is a bit sticky and a spot of grease is needed, OR the wiring to the local controller on the headlight is a bit intermittent and needs the contacts cleaning, drying or something of that ilk. To get to either of those requires taking off the whole front bumper, wheel arch panel and some sundry covers. Actual fix time 5-10 minutes. Fix time with removing and reinstalling all that crap 2 hours if you've done it before, probably 3 for a first time. Consequence: my local and thankfully very trustworthy grease monkeys are going to make a few quid because I just can't be arsed with all that.

Design by monkeys. Malicious monkeys.

Yep that what BMW stands for, Bastard Monkey Wankers 

I think my car is no better in that respect either, it has Xenon headlights that self level and also steer with the front wheels and I'm told that similar things have to be performed in order to service them 

[tggzzz]

Haul from the Rugby hamfest. Although I saw him, I think I got there before the amorphous blob that snarfed the HP9100A at last week's FRARS rally.

The Advance [EDIT: OFS2A] OFS2B off-air frequency standard looks like a dead loss. No manual available. The output never syncs, and since there are only two ICs (both 7490), it looks like it presumes Droitwich is at the original 200kHz rather than the current 198kHz. I'll either flip it at a hamfest, or use the mechanics and replace the PLL etc with something of my own design.




The HP654A works reasonably, but could probably do with a little tweaking and switch cleaning. I'm in two minds whether to do that, or to simply flip it "as is", using photos showing intact cal stickers. I haven't seen a sig gen with several output impedances from 50ohm to 600ohms, with switched knobs marked in dBm. Apart from that, I suspect its internals are the same as my HP651B, but with a lower peak output voltage. The output level is easily tweaked, and appears to be within 1sB of where it should be.





The Tek 468 basically works and has the feel of a Tek 465B. Ch2 is finicky in the usual ways: it can work as advertised, but touch the attenuator controls and the amplitude and frequency response goes to hell. Ought to be a standard cleaning operation, but accessing the attenuators is complicated by the wart on top. No, I don't like DM44 variants! I do have a Tek465B which could probably fuflill its remaining function as a parts mule for attenuator components.

The voltage/time measurement only operates on digitised signals, not on the analogue waveforms with DM44 variants. The digitiser is clearly "early" and experimental". 40ns sampling interval, <<100MHz, and a choice of sine or pulse reconstruction filters. The method of moving the cursors is odd, but works well. Haven't assessed the "envelope" mode. Nice to see a "save waveform" mode.

Standard clean 3.5ns risetime without digitiser




Gibbs effect with "sine" reconstruction filter, and asymmetry with "pulse" reconstruction filter. Time is measured between blobs on leading edges.





The Pico TA041 isolating differential probe works nicely up to 25MHz. I'm in two minds whether to flip it or keep it with my other similar probe. The Pico has the advantage of not being /20 or /200, and being battery operated, but the disadvantage of fixed heavy duty probes.

[bd139]

Reasonable haul that. I'm impressed

Was too cheap on the Advantest R3351A at £288. Didn't want to go any higher as it means I have to drive to somewhere south of London to go and get it and I can't be bothered.

[med6753]

The Tek 468 basically works and has the feel of a Tek 465B. Ch2 is finicky in the usual ways: it can work as advertised, but touch the attenuator controls and the amplitude and frequency response goes to hell. Ought to be a standard cleaning operation, but accessing the attenuators is complicated by the wart on top. No, I don't like DM44 variants! I do have a Tek465B which could probably fuflill its remaining function as a parts mule for attenuator components.

You can access the attenuator switch deck covers and do a fair job of cleaning the leaf switches without having to remove the DM44.

[factory]

Haul from the Rugby hamfest. Although I saw him, I think I got there before the amorphous blob that snarfed the HP9100A at last week's FRARS rally.

The Advance OFS2A off-air frequency standard looks like a dead loss. No manual available. The output never syncs, and since there are only two ICs (both 7490), it looks like it presumes Droitwich is at the original 200kHz rather than the current 198kHz. I'll either flip it at a hamfest, or use the mechanics and replace the PLL etc with something of my own design.

Book for the 'B' version seems the same as you describe that one, 200kHz LW, only two of the three ICs installed.

David

[SoundTech-LG]

Nice 468. I don't think mine has the DM44, but not in front of it right now to check...

[capt bullshot]

The Tek 468 basically works and has the feel of a Tek 465B. Ch2 is finicky in the usual ways: it can work as advertised, but touch the attenuator controls and the amplitude and frequency response goes to hell. Ought to be a standard cleaning operation, but accessing the attenuators is complicated by the wart on top. No, I don't like DM44 variants! I do have a Tek465B which could probably fuflill its remaining function as a parts mule for attenuator components.

You can access the attenuator switch deck covers and do a fair job of cleaning the leaf switches without having to remove the DM44.

For the 468, the "DM44" bump doesn't have anything in common with the normal DM44. The digitizer isn't within that space, but on a couple of boards in the main chassis. Afair, the 468 is indeed a 465 with the digital storage added, but required some re-arrangement of its guts.

https://w140.com/tekwiki/wiki/468

[mansaxel]

I do use unix command line tools on photos. Either batch jobs, or converting photos of manuals to pdfs that are 1/20th the size and more readable to boot.

The Vince-fication of pics I link is something along the lines of:

Code: [Select]

convert -strip -resize 50% -quality 60% DSC_1716.JPG measuring-standard.jpeg
(convert the picture by stripping EXIF metadata, resizing to 50%, set JPEG quality to 60% and rename resulting file. ) This will take a 5MiB file and push it to about 400K, depending on motif. )

Granted, the strip-part is because I don't think the web crawlers need to correlate all the pics my phone has taken, has nothing to do with Vince...

[SoundTech-LG]

Something tells me they all came with the DM44 as standard equipment. Am I wrong?