Test Equipment Anonymous (TEA) group therapy thread

[Neomys Sapiens]

Just for completeness (not sure if I did post them back when I got them); that's the one.

[mnementh]

Stop flashing us pics of your peace-pipes, shrew-apostle!

mnem
*toddles off to ded*

[med6753]

Spent hours today of trial and error trying to get all the controls on this Type 1A4 cleaned up and operating smoothly. This plug-in is packed tight around the pre-amp section especially around channels 2 and 3. Very difficult to even see the switch decks to get the deoxit where it's needed.

Finally got the all the controls acceptable. And started assessment of what needed to be calibrated. The gain is fine across all four channels. Once the controls were clean the DC Balance for all 4 fell into place but initially the DC Balance control for channel 4 was jammed and would not turn. Luckily I could get to that pot and disassemble it and free it up. If it had been either channel 2 or 3 forget it.

Compensation: Oddly channels 1 and 4 are excellent. No adjustments required. Channels 2 and 3 need some tweaking. All the compensation controls are accessible on the bottom of the plug-in as typical with most 500 series plug-in's. So Sunday will be compensation day. The Type 547 will be placed on it's side and the bottom plate removed to gain access. At the same time the Type 1A1 plug-in that I repaired last week will get channel 1 compensated again. Once these compensation adjustments are complete all the 500 series plug-in's will be done. 

       

[Zucca]

Just finished to watch "don't look up" movie.
Excellent.

[valley001]

Found this at a thrift store today for $2. 

[med6753]

Found this at a thrift store today for $2.

Put that up for bids and the hp fan boys will be falling over each other and you'll make a killing. 

[valley001]

Found this at a thrift store today for $2.

Put that up for bids and the hp fan boys will be falling over each other and you'll make a killing. 

Ive got to fill out my supply first.    Now I can power my 3450A and 3440A at the same time!

[vk6zgo]

until I realized most of it is probably 75ohms not 50, because among the 150 pieces I got, there are about 30/40 75ohms terminators. So I assume all the adapters and sockets in the lot, are also 75 ohms, bummer     Would have to look at each of them one by one very closely to see if some of them by some chance are 50 ohms. Would be cool because from what I see it's old quality stuff, not modern Chinese junk.  A lot of it is Radiall for starters.

If the centre socket in the female has a cylinder of dielectric around it, it's a 50Ω. If it is just the bare socket, and looking rather fragile, it's 75Ω.

It used to be SOP in the video business (who are those that use 75Ω coax the most) to use 50Ω connectors with 75Ω cable, because the 50Ω is a bit sturdier. They had to stop that when the bandwidth had to increase to accommodate HD-TV, 720p and up.

Beat me to it!
The most likely place to find really old genuine 75 ohm connectors/adaptors is from the baseband /IF part of Microwave comms systems.

I almost bought a R&S SWOB Poyskop at a Hamfest, until I realised it was a 75 ohm one.
Yeah, I know there are workarounds, but that, & lugging it down the road to the car were the things that put me off buying the huge beast.

[mansaxel]

So....... to sum it up, unless proven wrong or better explained by one of you in a few moments :

- Thick white dielectric, or slightly less thick but still chunky =  50 ohms.
- No white dielectric AT ALL,..... or REALLY THIN dielectric = 75ohms.

Basically yes, with the explanation that specifically for video transmission (like 3G-SDI, 1080 uncompressed, which means about 3Gbit.) the standard 75Ω BNC was not clean enough.  Therefore the connector was refined to create an even better connection, specifically for this, and those are the ones without dielectric.  Thus, we've got three steps:

• For analog ~600 lines video, PAL / SECAM / NTSC; 50Ω connector on 75Ω cable.
• For SD-SDI, SMPTE 259M, standard 75Ω through and through; also slightly better cable to get an edge in length where needed.
• For various HD-SDI standards, starting with SMPTE 292M and evolving into SMPTE 424M; new connector (the ultra-clean BNC) and also even better cable.
• The future holds a lot of optical fibre, and also a lot of video over IP networks. Most often it's Ethernet, but it is not necessarily so; except in the end where standards like SMPTE 2110 say it shall be RTP over IP multicast over Ethernet to connect the end node.

For you, things like bulkhead couplers and splitters that are 50Ω, will be most useful.

Cable that's 75Ω with a 50Ω plugs, well maybe... I've got a few of those in my lab, and for non-trivial things (2 meters and audio frequencies for instance) one's hard pressed to notice.

[TERRA Operative]

Vince what is your opinion ...

If we want to replace a knob or button on our vintage instrument, does the button need to look brand new, or does it need to look like other buttons that are "weathered" (model RR term) and used? Which is the perfect choice in your opinion?

 

   Gotta replace them all to keep it all a matching set! 

Or just repair the knob with J-B Weld.

That repair...  is a damn fine compromise.   A little industrial grey spray paint and it would be hard to tell from intact knobs that had been banged about a bit.

No idea how long it'll hold up; there are so many variables with such things. But compared to molding a complete replacement or spending months seeking a cost-effective replacement...? Definitely worth the gamble to see if it holds up.

mnem
*knobber*

Yeah, it seems pretty solid for now, my problem was finding a replacement, these unit's aren't super common and the knob is specific to this gear, so finding a replacement is not so easy.

I'm also anal retentive about making things look 'right', but only up to a point... One day I'll have my resin printer running and will be able to make parts from scratch that are practically indistinguishable from OEM, but that's 'one day'

[tggzzz]

meh. The world is in color; black and white are for people with no imagination. 

8% of men and 0.1% of women have some form of colour blindness.

Being color blind didn't seem to hinder John Dalton's imagination. His primary claim to fame isn't that he was colour blind.

When GeoCities was dominant, a case could be made that colour blindness was beneficial

[tggzzz]

It is a standing joke that the letters to newspapers and other periodicals from nutters are written ALL IN CAPITALS IN GREEN INK. I can tell you from my personal experience of actually sitting, opening and reading the correspondence to periodicals that those kind of formatting quirks are pretty reliable indicators that the correspondent is going to be somewhere on the spectrum from "incoherent monomaniac" to "Erm, I think we'd better call the police". The quirk wasn't always ALL IN CAPITALS IN GREEN INK but it genuinely made up the higher proportion of crank letters than did other creative and unconventional approaches to formatting.

Plus ça change, plus c'est la même chose.

I've always been guided by “Perfection is achieved, not when there is nothing more to add, but when there is nothing left to take away.” ― Antoine de Saint-Exupéry, Airman's Odyssey

[Robert763]

until I realized most of it is probably 75ohms not 50, because among the 150 pieces I got, there are about 30/40 75ohms terminators. So I assume all the adapters and sockets in the lot, are also 75 ohms, bummer     Would have to look at each of them one by one very closely to see if some of them by some chance are 50 ohms. Would be cool because from what I see it's old quality stuff, not modern Chinese junk.  A lot of it is Radiall for starters.

If the center socket in the female has a cylinder of dielectric around it, it's a 50Ω. If it is just the bare socket, and looking rather fragile, it's 75Ω.

It used to be SOP in the video business (who are those that use 75Ω coax the most) to use 50Ω connectors with 75Ω cable, because the 50Ω is a bit sturdier. They had to stop that when the bandwidth had to increase to accommodate HD-TV, 720p and up.

OH !!! so you mean NO white dielectric piece AT ALL ?! 

Lemme see ! Found an article on Wikipedia on BNC plugs... a pic comparing 50 and 75 ohms... pic below.
Hell you are right, ZERO white dielectric piece in the center ! 

I don't understand... from a video a few years back from Shariar, IIRC he said the thickness of the white dielectric bit varied depending on the impedance of the plug/socket.
So I was paying attention to that when looking at the pile of adapters I have in that box.. and indeed noticed some pieces have a very thick white dielectric piece, same as the 50 ohms sockets on my scopes, whereas other pieces have a slightly thinner dielectric, and others have an even thinner, realllly thin dielectric. See picture below where I compare 3 pieces side by side : leftmost is one of the 50ohms adapters I just bought from Farnell. Then center, a 75ohms termination from Tektronix, one can clearly see the white dielectric bit is visibly thinner. Then rightmost, another termination, which is again clearly even thinner still ! Then I searched some more in the pile of bits and found a BNC socket like you describe, with zero white dielectric in the middle... which I photographed next to a 50ohms socket.

WORSE STILL : among the very 50 ohms bits I just ordered from Farnell (made TRIPLE sure when ordering, that it was all 50 ohms not 75...)... I now notice that they do NOT have all the same thickness white piece ! Most of them have a thick piece, but the " Y " adapter has a slightly thinner piece ! Oh no !
I am properly confused now, let me tell ya !!!!! 

What the hell is going on here ?! 

My understanding was as Shariar said that the dielectric thickness and nature/material used, determines the impedance of the connector or cable. so surely if the thickness of the white piece varies, so does the impedance no ?!    Also, if we have a varying thickness among 50ohms connectors... that means that depending on what particular plug you shove into a particular socket... you will end up with an air gap between the two white pieces, a gap whose size varies depending on the plug/socket combo that you happen to be using. Surely an air gap is not good ? Never mind one that you can't even know/control the size of ?!

Oh no.... I am real confused now people.... if I can't even get super basic stuff like this, straight, I will never go very far in my electronics adventures I fear ! 

Wiki article https://en.wikipedia.org/wiki/BNC_connector at any rate confirms what you said... plus video stuff is your line of business or used to be at least as I understand it.... so clearly you know what a 75 ohm socket and plug looks like so I trust you ! 

So that would tend to say that if it has ANY thickness of white stuff, it's 50ohms for sure, I can trust that ? And that it's not important if you mix and match plugs/sockets that have a different thickness of the white stuff...weird but hey.

Wiki article says that 75ohms connectors are good for 2GHz and 50ohms for 4GHz.

Oh boy....


Well any way, if white = 50ohms, then I am lucky and all the stuff I got in that box is 50ohms not 75ohms, so I have hundreds of Euros worth of Radial adapters and sockets, great !!!!      Lots of nice panel mount sockets, some cool 90° angled adapters, Tees and much more.

Oh not this again.
The ORIGINAL 75R BNC had smaller center contacts. It was constant impedance. The problem was it is not intermatable with 50R. Putting a 50R male on a original 75R female damages the female contact so it won't then make proper contact with 75R males. A original 75R male  won't make proper contact with 50R female.
This issue was "fixed" by using the 50R center contacts with cut back insulation. This is inter-matable with 50R. However even two mated new 75R BNCs are not constant impedance. At video and other low frequencies where 75R is mostly used the short length of the discontinuity has little effec on performance.
They also make 75R N types (and C) but these always use thin ceneter contacts. If you come across a N connector with red insulation or painted band inspect it carefully, it may be 75R. I have a R&S 75R SWR bridge that someone ruined by someone (not me) screwing a 50R N onto it.

[capt bullshot]

So....... to sum it up, unless proven wrong or better explained by one of you in a few moments :

- Thick white dielectric, or slightly less thick but still chunky =  50 ohms.
- No white dielectric AT ALL,..... or REALLY THIN dielectric = 75ohms.

This at least is aligned with my observations. If in doubt, I still can use my Tek 1502C TDR to determine the impedance of a cable or connector.

My gut feeling is that the thicker the dielectric the higher the bandwidth. 75ohms bits that have a thin white piece is maybe an attempt to give them a slightly better performance than a regular piece ?

Bandwidth is indirectly affected by the cable / connector construction. The mechanical dimensions plus the dielectric properties of the insulator determine the nominal impedance. The properties of the material (wire and dielectric), especially dielectric and resistive losses, determine the damping of the cable. Usually damping increases with increasing frequency, so "better materials" and construction details have large influence on the resulting bandwidth and damping of a cable. This is more difficult to achieve as the frequency increases.
As far as I remember, a 60Ohm impedance cable, using common (non-expensive) material can achieve the best damping performance, due to some lucky coincidences in construction details (ratio of inner core / shield diameter and amount of used dielectric). High power cables often use air as dielectric and as little as possible non-air material to keep the inner conductor centered in the shield.

[capt bullshot]

meh. The world is in color; black and white are for people with no imagination. 

If it's just meh to you, I suppose you aren't offended too much if I confess: "Usually I skip over your posts because of your creative use of text coloration and formatting".

[Robert763]

So....... to sum it up, unless proven wrong or better explained by one of you in a few moments :

- Thick white dielectric, or slightly less thick but still chunky =  50 ohms.
- No white dielectric AT ALL,..... or REALLY THIN dielectric = 75ohms.

This at least is aligned with my observations. If in doubt, I still can use my Tek 1502C TDR to determine the impedance of a cable or connector.

My gut feeling is that the thicker the dielectric the higher the bandwidth. 75ohms bits that have a thin white piece is maybe an attempt to give them a slightly better performance than a regular piece ?

Bandwidth is indirectly affected by the cable / connector construction. The mechanical dimensions plus the dielectric properties of the insulator determine the nominal impedance. The properties of the material (wire and dielectric), especially dielectric and resistive losses, determine the damping of the cable. Usually damping increases with increasing frequency, so "better materials" and construction details have large influence on the resulting bandwidth and damping of a cable. This is more difficult to achieve as the frequency increases.
As far as I remember, a 60Ohm impedance cable, using common (non-expensive) material can achieve the best damping performance, due to some lucky coincidences in construction details (ratio of inner core / shield diameter and amount of used dielectric). High power cables often use air as dielectric and as little as possible non-air material to keep the inner conductor centered in the shield.

Not sure what you mean by "bandwidth". It's not a normal coaxial cable parameter.
The impedance is set by the ratio of inner conductor OD, outer conductor ID and the dielectric constant of the insulator beween them. This allows an infinite range of dimensions for a giiven impedance.  The size and material of the conductors and insulator also affect the losses. Its all a compromise.  Note that this is nothing to do with frequency. The losses increase with frequency due to resistnce, skin effect dielectric losses. leakage etc. All else being equal larger cables are lower loss. This does not set a "hard" upper limit to the frequency if you can accept the losses.
What sets the upper frequency is "modeing" This is where the ID of outer conductor becomes a waveguide rather than a coaxial cable outer conductor.  This leads to significant losses and reflections. The frequency is not clear cut. It depends on the detail desigin of connectors. For a nominal 7mm cable and connector shield ID (BNC, TNC, N and APC7 are all 7mm connectors) The cut-off is about 18GHz. Some high precision ones will got to 22GHz. 3.5mm connetors like SMA can go to 26GHz. For higher frequencies the SMA-like connectors have reduced sizes e.g. 2.9mm  1.9mm.

[capt bullshot]

Bandwidth is indirectly affected by the cable / connector construction. The mechanical dimensions plus the dielectric properties of the insulator determine the nominal impedance. The properties of the material (wire and dielectric), especially dielectric and resistive losses, determine the damping of the cable. Usually damping increases with increasing frequency, so "better materials" and construction details have large influence on the resulting bandwidth and damping of a cable. This is more difficult to achieve as the frequency increases.
As far as I remember, a 60Ohm impedance cable, using common (non-expensive) material can achieve the best damping performance, due to some lucky coincidences in construction details (ratio of inner core / shield diameter and amount of used dielectric). High power cables often use air as dielectric and as little as possible non-air material to keep the inner conductor centered in the shield.

Not sure what you mean by "bandwidth". It's not a normal coxial cable parameter.

Fully agreed, this isn't a normal cable parameter, and I agree to your further explanations.

I thought of explaining through a very simplified model, why the "thicker" cable might have more "bandwidth" than the "thinner" cable for maybe sub 2GHz applications. Here you've got to deal mostly with increasing damping over frequency, and this could be simplified seen as "the better cable has more bandwidth". Next issue would be losses and distortion due to impedance mismatch, this for sure is valid for some 10MHz and above, depending on cable length.

[Mortymore]

Just finished to watch "don't look up" movie.
Excellent.

 

For those who intend to watch, don't miss the extra scene during credits at the end

[m k]

Thought I score big time, until I realized most of it is probably 75ohms not 50

The light plastic covered one is pretty close to what I have for 50 ohm Thinnet (10base2) and it was 10MHz baseband.
No idea what shape the signal was but amplitude was binary for sure.
There max. segment length was 185m but it wasn't because of attenuation, it was because of the delay.

E,
5<->8

Couldn't find what 30 practical connections per segment actually mean.

[Martin Miranda]

time to fix the boat anchor. 

[mnementh]

meh. The world is in color; black and white are for people with no imagination. 

If it's just meh to you, I suppose you aren't offended too much if I confess: "Usually I skip over your posts because of your creative use of text coloration and formatting".

Not in the least.

Human beings are tool-users; each and every one of us. We all prefer different tools, and we all come to terms with how best to use those tools we choose in our own way.

But since I'm addressing you personally, and I would like for you to read this missive, I'll try and do it normally, just this once.

mnem

[mnementh]

It is a standing joke that the letters to newspapers and other periodicals from nutters are written ALL IN CAPITALS IN GREEN INK. I can tell you from my personal experience of actually sitting, opening and reading the correspondence to periodicals that those kind of formatting quirks are pretty reliable indicators that the correspondent is going to be somewhere on the spectrum from "incoherent monomaniac" to "Erm, I think we'd better call the police". The quirk wasn't always ALL IN CAPITALS IN GREEN INK but it genuinely made up the higher proportion of crank letters than did other creative and unconventional approaches to formatting.

Plus ça change, plus c'est la même chose.

I've always been guided by “Perfection is achieved, not when there is nothing more to add, but when there is nothing left to take away.” ― Antoine de Saint-Exupéry, Airman's Odyssey

Evidently, over the years, the developers of 75Ω BNCs have also followed this directive.

mnem
"La grande roue continue de tourner." ~grand-mère Armantier

[Vince]

@ Coax Gang : thanks chaps for helping clearing everything up ! 

Oh not this again.
The ORIGINAL 75R BNC had smaller center contacts. It was constant impedance. The problem was it is not intermatable with 50R. Putting a 50R male on a original 75R female damages the female contact so it won't then make proper contact with 75R males. A original 75R male  won't make proper contact with 50R female.
This issue was "fixed" by using the 50R center contacts with cut back insulation. This is inter-matable with 50R. However even two mated new 75R BNCs are not constant impedance. At video and other low frequencies where 75R is mostly used the short length of the discontinuity has little effec on performance.
They also make 75R N types (and C) but these always use thin ceneter contacts. If you come across a N connector with red insulation or painted band inspect it carefully, it may be 75R. I have a R&S 75R SWR bridge that someone ruined by someone (not me) screwing a 50R N onto it.

Thanks, that's clear. So if I understand what you said here and in your other post, then the original 75R plug had a thin white dielectric piece, and it got removed altogether later to achieve the 50R mechanical compatibility, since one has to maintain the OD/IN ratio constant for a given impedance. So since the OD of the inner conductor was made bigger to match the 50R one, so they had to make the ID of the outside conductor larger as well, to keep the ratio constant. So they removed the little white dielectric that there was, completely, because that's the only way they could increase the diameter.

Now that I actually know the reason for everything, it all makes sense and will be therefore easy to remember 

So that's cool then, 90% of the stuff I have in that box of random bits, is indeed 50R, Radiall, I am am so happy, excellent score that was 

[med6753]

Wake up this morning to a temperature of -2 C and freezing rain. That's rain that falls as liquid and then immediately freezes upon hitting the ground. Turns everything including the roads into a sheet of ice. The scanner is going nuts with numerous accidents. Road crews are out sanding/salting but naturally can't be everywhere at once.

It's becoming a real clusterfuck out there.  Glad I don't have to go anywhere. It is supposed to go above freezing later today.

Edit....call just came across....even the salt trucks are getting into accidents. 

[mnementh]

   time to fix the boat anchor. 

Reminds me of when I first resurrected my 2230; I think I even had them spread out all over the same dining room table. 

I got my SM from Artek; amazing quality scans on the schematics that scaled up to 2ft x 3ft poster size from Kinko's and still crisp as a new dollar bill.

mnem