Test Equipment Anonymous (TEA) group therapy thread

[tggzzz]

Went to the Chippenham hamfest Radio Rally. Put some treasures in the car, but couldn't be bothered to try to sell anything in a field in the drizzle.

One bloke had a three Tek2712 and two Tek492 21GHz spectrum analysers. One of the former was shown working on a pure sine wave inverter; the display was just about visible. One 492 had a broken blue screen shield and was missing a handle, which is essential for that lump of metal. Said he would sell them for £300, but hadn't when I left.

Didn't get much. A couple of metal enclosures for £1 each, yet another Cropico resistance box (cal ran out in 2014) all resistors measure OK, and some banana plugs.

£300? Some people are dreamers! Wanted to go to this Rally hamfest, but didn't manage to get down there today.

The 492 is a reasonable SA, and a good one is probably worth £300. I have no opinion about the 2712. I doubt many 40yo things sell for £300 at a hamfest.

(See, I too can change your statement without explicitly noting it)

[Squarewave]

Went to the Chippenham hamfest Radio Rally. Put some treasures in the car, but couldn't be bothered to try to sell anything in a field in the drizzle.

One bloke had a three Tek2712 and two Tek492 21GHz spectrum analysers. One of the former was shown working on a pure sine wave inverter; the display was just about visible. One 492 had a broken blue screen shield and was missing a handle, which is essential for that lump of metal. Said he would sell them for £300, but hadn't when I left.

Didn't get much. A couple of metal enclosures for £1 each, yet another Cropico resistance box (cal ran out in 2014) all resistors measure OK, and some banana plugs.

£300? Some people are dreamers! Wanted to go to this Rally hamfest Radio Rally, but didn't manage to get down there today.

The 492 is a reasonable SA, and a good one is probably worth £300. I have no opinion about the 2712. I doubt many 40yo things sell for £300 at a hamfest Radio Rally.

(See, I too can change your statement without explicitly noting it)

There's better to be had for less or even similar money though.

[Cerebus]

Just swapped a third party 'spicy pillow' out of my iPhone SE. That didn't last very long, I put that in August last year. Replaced with what purports to be, and I hope is, a genuine Apple battery.

Anybody want a spicy pillow with about 4Wh charge in it?

[tggzzz]

Went to the Chippenham hamfest Radio Rally. Put some treasures in the car, but couldn't be bothered to try to sell anything in a field in the drizzle.

One bloke had a three Tek2712 and two Tek492 21GHz spectrum analysers. One of the former was shown working on a pure sine wave inverter; the display was just about visible. One 492 had a broken blue screen shield and was missing a handle, which is essential for that lump of metal. Said he would sell them for £300, but hadn't when I left.

Didn't get much. A couple of metal enclosures for £1 each, yet another Cropico resistance box (cal ran out in 2014) all resistors measure OK, and some banana plugs.

£300? Some people are dreamers! Wanted to go to this Rally hamfest Radio Rally, but didn't manage to get down there today.

The 492 is a reasonable SA, and a good one is probably worth £300. I have no opinion about the 2712. I doubt many 40yo things sell for £300 at a hamfest Radio Rally.

(See, I too can change your statement without explicitly noting it)

There's better to be had for less or even similar money though.

Any examples? (My experience is with a 492AP)

[TERRA Operative]

Hmm, I'm sure I'm doing something stupidly obviously stupid..

Anyone know why when I feed a 4V P-P square-ish wave from an OCXO into one section of a 74LS14 Hex Schmitt-Trigger Inverter, I get the following waveform out? I'm looking for a nice snappy TTL square wave...
(Chip supply is exactly 5V, proper decoupling, scoped with 10x probe with low inductance ground connection etc etc)

[tautech]

Try a differential measurement. 

[Zoli]

Hmm, I'm sure I'm doing something stupidly obviously stupid..

Anyone know why when I feed a 4V P-P square-ish wave from an OCXO into one section of a 74LS14 Hex Schmitt-Trigger Inverter, I get the following waveform out? I'm looking for a nice snappy TTL square wave...
(Chip supply is exactly 5V, proper decoupling, scoped with 10x probe with low inductance ground connection etc etc)

DC coupled? 20MHz bandwidth?

[TERRA Operative]

Measurement is DC coupled, no real difference in waveshape if switching to AC coupling. Using 20MHz bandwidth just smooths the waveform (I'm using a Tek TDS220).

I found an older design I used that seemed to work and I put a 1nF cap between the OCXO and 74LS14, followed by a 10K trimpot connected across +5V and GND, with the wiper to the trace between the cap and the 74LS14 to adjust offset etc.
Let me bodge that in and see what happens...

[EDIT] No change, still crap.

[vk6zgo]

<snip>
I was meaning as opposed to modern day stuff that normally just works when you plug it in 

Or not, if it is W10! 

My "you beaut" little nanoVNA arrived recently, & after playing with it for a while, I decided to download one of the software packages readily available on the Internet.

It looks like it is downloading, & even asks you "if you want to open it".
I told it "yes", & it proceeded to make a few beeps & do nothing---rather reminiscent of loading "Frogger" into a C64 from a Datasette, & with less result.

At least the C64 thing would load properly about every tenth time you tried.
Hang on, maybe I have to try this 10 times? 

Is it a genuine nanoVNA or a clone, the clones are known to have issues.

It purports to be "genuine", & is from an Oz seller, for whatever that's worth, as this country has a long history of con artists!

That said, it seems to work OK  "standalone", which is the appropriate mode for how I will use it most of the time.
Galumphing around in the back yard with a laptop "ain't my bag"!

Sometimes, it would be nice to have a BIG display, to be able to set parameters more easily, & to save results in a more normal format, but if it is a problem to do with the nanoVNA  & not W10 or the software, I can live with it.

The software I tried was one of the earlier ones----I know joeqsmith has done a lot of work on an improved version of his own, but I thought I'd just try the "dumbo" one, first!

[mnementh]

   

https://a360.co/3zNKW0M

Currently printing. Dunno how it will look until the first copy cools off.

mnem
 

[Kosmic]

Went to the Chippenham hamfest Radio Rally. Put some treasures in the car, but couldn't be bothered to try to sell anything in a field in the drizzle.

One bloke had a three Tek2712 and two Tek492 21GHz spectrum analysers. One of the former was shown working on a pure sine wave inverter; the display was just about visible. One 492 had a broken blue screen shield and was missing a handle, which is essential for that lump of metal. Said he would sell them for £300, but hadn't when I left.

Didn't get much. A couple of metal enclosures for £1 each, yet another Cropico resistance box (cal ran out in 2014) all resistors measure OK, and some banana plugs.

£300? Some people are dreamers! Wanted to go to this Rally hamfest Radio Rally, but didn't manage to get down there today.

The 492 is a reasonable SA, and a good one is probably worth £300. I have no opinion about the 2712. I doubt many 40yo things sell for £300 at a hamfest Radio Rally.

(See, I too can change your statement without explicitly noting it)

There's better to be had for less or even similar money though.

Any examples? (My experience is with a 492AP)

£300 is actually not a bad price for a working 21GHz spectrum analyzer. I actually sold couples of Tek 2794 (a bit better than the 492) for £1200.

[capt bullshot]

Hmm, I'm sure I'm doing something stupidly obviously stupid..

Anyone know why when I feed a 4V P-P square-ish wave from an OCXO into one section of a 74LS14 Hex Schmitt-Trigger Inverter, I get the following waveform out? I'm looking for a nice snappy TTL square wave...
(Chip supply is exactly 5V, proper decoupling, scoped with 10x probe with low inductance ground connection etc etc)

I wouldn't expect anything better from a '74LS' chip. Try some faster CMOS ones.

[BU508A]

Curious Marc got his hands on a HP 711A 500V PSU from the early 60's.
And he replaced some caps (recapping porn for med6753  ).

[bd139]

Just swapped a third party 'spicy pillow' out of my iPhone SE. That didn't last very long, I put that in August last year. Replaced with what purports to be, and I hope is, a genuine Apple battery.

Anybody want a spicy pillow with about 4Wh charge in it?

Unless you bought it from Apple it’s probably not genuine. I just get them to do it at the apple store. If they break it they give you a new phone. That happened once 

[bd139]

Hmm, I'm sure I'm doing something stupidly obviously stupid..

Anyone know why when I feed a 4V P-P square-ish wave from an OCXO into one section of a 74LS14 Hex Schmitt-Trigger Inverter, I get the following waveform out? I'm looking for a nice snappy TTL square wave...
(Chip supply is exactly 5V, proper decoupling, scoped with 10x probe with low inductance ground connection etc etc)

74LS14 is specified as 22ns rise time. It’s coming out at 19ns there which is about 1/5 of the cycle time so that’s about right.

Try 74HC or even better a 74LVC1G14. The later will give you 2.2ns approx rise time. That of course then brings more bandwidth problems to the table.

[Robert763]

Went to the Chippenham hamfest Radio Rally. Put some treasures in the car, but couldn't be bothered to try to sell anything in a field in the drizzle.

One bloke had a three Tek2712 and two Tek492 21GHz spectrum analysers. One of the former was shown working on a pure sine wave inverter; the display was just about visible. One 492 had a broken blue screen shield and was missing a handle, which is essential for that lump of metal. Said he would sell them for £300, but hadn't when I left.

Didn't get much. A couple of metal enclosures for £1 each, yet another Cropico resistance box (cal ran out in 2014) all resistors measure OK, and some banana plugs.

£300? Some people are dreamers! Wanted to go to this Rally hamfest Radio Rally, but didn't manage to get down there today.

The 492 is a reasonable SA, and a good one is probably worth £300. I have no opinion about the 2712. I doubt many 40yo things sell for £300 at a hamfest Radio Rally.

(See, I too can change your statement without explicitly noting it)

There's better to be had for less or even similar money though.

Any examples? (My experience is with a 492AP)

I don't know of any 21 GHz or even 18GHz SA that can be reguarly had for less than £300 never mind are better than a 49x. Well, the ancient Polrad things turn up sometimes but they are generally only good for parts. Even a working HP 141T + 8555A is likely to fetch £300 and no way is it better in any respect than a 492.
I paid £600 for a 492BP which I thought was a good price. Mind you it is in perfect condition and came with cover, diplexer and 28.5 GHz mixer. The 492BP is a bit rare and hits a sweetspot in perfomance / features / reliabilty as long as you don't need 10 Hz RBW.

[tggzzz]

Hmm, I'm sure I'm doing something stupidly obviously stupid..

Anyone know why when I feed a 4V P-P square-ish wave from an OCXO into one section of a 74LS14 Hex Schmitt-Trigger Inverter, I get the following waveform out? I'm looking for a nice snappy TTL square wave...
(Chip supply is exactly 5V, proper decoupling, scoped with 10x probe with low inductance ground connection etc etc)

TTL (including H, L, S, AS, ALS, F TTL) outputs (and inputs) are asymmetric. Input voltages only have to be below 0.8V and above 2.0V, and the output transitions only have to be snappy 0.8-2V. TTL outputs rarely go much above 3.5V, and counter outputs can and do legitimately have a lot of "ripple" on the high output voltage.

The various "C" TTL families have symmetric outputs; modern variants fall into the "damn-fast" category (I've managed to arrange 74LVC1G gates driving 50ohms with ~300ps transition times.)

[Zenith]

Went to the Chippenham hamfest. Put some treasures in the car, but couldn't be bothered to try to sell anything in a field in the drizzle.

One bloke had a three Tek2712 and two Tek492 21GHz spectrum analysers. One of the former was shown working on a pure sine wave inverter; the display was just about visible. One 492 had a broken blue screen shield and was missing a handle, which is essential for that lump of metal. Said he would sell them for £300, but hadn't when I left.

Didn't get much. A couple of metal enclosures for £1 each, yet another Cropico resistance box (cal ran out in 2014) all resistors measure OK, and some banana plugs.

I don't think much is sold at rallies for over £250 a pop.

What with one thing and another I missed it. It's small but it's been OK and worth the trip when I've been there in the past.

FRARS and Rugby are next for me. Rugby has been pretty good. I used to go up the A429 (Fosse Way) which was a nice drive, but they've turned it into one big speed trap.

[TERRA Operative]

Hmm, I'm sure I'm doing something stupidly obviously stupid..

Anyone know why when I feed a 4V P-P square-ish wave from an OCXO into one section of a 74LS14 Hex Schmitt-Trigger Inverter, I get the following waveform out? I'm looking for a nice snappy TTL square wave...
(Chip supply is exactly 5V, proper decoupling, scoped with 10x probe with low inductance ground connection etc etc)

TTL (including H, L, S, AS, ALS, F TTL) outputs (and inputs) are asymmetric. Input voltages only have to be below 0.8V and above 2.0V, and the output transitions only have to be snappy 0.8-2V. TTL outputs rarely go much above 3.5V, and counter outputs can and do legitimately have a lot of "ripple" on the high output voltage.

The various "C" TTL families have symmetric outputs; modern variants fall into the "damn-fast" category (I've managed to arrange 74LVC1G gates driving 50ohms with ~300ps transition times.)

Looks like I can get TC74VHCT14AF off the shelf in Akihabara, which are good for 5.0ns.
https://toshiba.semicon-storage.com/info/docget.jsp?did=18349&prodName=TC74VHCT14AF

I might go get a fistful of those tomorrow and see how they go.

[tggzzz]

Hmm, I'm sure I'm doing something stupidly obviously stupid..

Anyone know why when I feed a 4V P-P square-ish wave from an OCXO into one section of a 74LS14 Hex Schmitt-Trigger Inverter, I get the following waveform out? I'm looking for a nice snappy TTL square wave...
(Chip supply is exactly 5V, proper decoupling, scoped with 10x probe with low inductance ground connection etc etc)

TTL (including H, L, S, AS, ALS, F TTL) outputs (and inputs) are asymmetric. Input voltages only have to be below 0.8V and above 2.0V, and the output transitions only have to be snappy 0.8-2V. TTL outputs rarely go much above 3.5V, and counter outputs can and do legitimately have a lot of "ripple" on the high output voltage.

The various "C" TTL families have symmetric outputs; modern variants fall into the "damn-fast" category (I've managed to arrange 74LVC1G gates driving 50ohms with ~300ps transition times.)

Looks like I can get TC74VHCT14AF off the shelf in Akihabara, which are good for 5.0ns.
https://toshiba.semicon-storage.com/info/docget.jsp?did=18349&prodName=TC74VHCT14AF

I might go get a fistful of those tomorrow and see how they go.

That 5ns is propagation delay, not transition time. The transition time isn't specified, and will be strongly dependent on load is being driven.

Be careful if you want to drive a 50ohm load, e.g. a transmission line or input; it would probably be better to have several parallel "output+series resistor" to keep the per-gate output current relatively low.

[tggzzz]

Went to the Chippenham hamfest. Put some treasures in the car, but couldn't be bothered to try to sell anything in a field in the drizzle.

One bloke had a three Tek2712 and two Tek492 21GHz spectrum analysers. One of the former was shown working on a pure sine wave inverter; the display was just about visible. One 492 had a broken blue screen shield and was missing a handle, which is essential for that lump of metal. Said he would sell them for £300, but hadn't when I left.

Didn't get much. A couple of metal enclosures for £1 each, yet another Cropico resistance box (cal ran out in 2014) all resistors measure OK, and some banana plugs.

I don't think much is sold at rallies for over £250 a pop.

What with one thing and another I missed it. It's small but it's been OK and worth the trip when I've been there in the past.

Agreed on both points.

FRARS and Rugby are next for me. Rugby has been pretty good. I used to go up the A429 (Fosse Way) which was a nice drive, but they've turned it into one big speed trap.

Fosse Way is still nicer than the M5/M42, and I'n happy it isn't so much of a death trap. Looks like a similar time/distance to RATS and FRARS, dammit!

Sounds like we aren't too far apart; I'm near Brizzle.

[ch_scr]

Hmm, I'm sure I'm doing something stupidly obviously stupid..

Anyone know why when I feed a 4V P-P square-ish wave from an OCXO into one section of a 74LS14 Hex Schmitt-Trigger Inverter, I get the following waveform out? I'm looking for a nice snappy TTL square wave...
(Chip supply is exactly 5V, proper decoupling, scoped with 10x probe with low inductance ground connection etc etc)

TTL (including H, L, S, AS, ALS, F TTL) outputs (and inputs) are asymmetric. Input voltages only have to be below 0.8V and above 2.0V, and the output transitions only have to be snappy 0.8-2V. TTL outputs rarely go much above 3.5V, and counter outputs can and do legitimately have a lot of "ripple" on the high output voltage.

The various "C" TTL families have symmetric outputs; modern variants fall into the "damn-fast" category (I've managed to arrange 74LVC1G gates driving 50ohms with ~300ps transition times.)

Looks like I can get TC74VHCT14AF off the shelf in Akihabara, which are good for 5.0ns.
https://toshiba.semicon-storage.com/info/docget.jsp?did=18349&prodName=TC74VHCT14AF

I might go get a fistful of those tomorrow and see how they go.

That 5ns is propagation delay, not transition time. The transition time isn't specified, and will be strongly dependent on load is being driven.

Be careful if you want to drive a 50ohm load, e.g. a transmission line or input; it would probably be better to have several parallel "output+series resistor" to keep the per-gate output current relatively low.

If you look at page 4: V(OH) at 8mA load - they specify at least 3.94Vout @ 4.5Vcc - a 0.56V drop implying an output resistance of [0.56V/8mA]=70R (or lower). This kind of calculation gave me so far "close enough" values to make an estimate on how many gates to put in parallel / how much series resistance is needed to get to 50R. In this case two gates with 33R in series give 100R, parallel them after the resistors and it should give a nice square into 50R.

[Specmaster]

Todays eBay fail, seller describes this as "TTI 1604 digital multimeter Digital Oscilloscope"

https://www.ebay.co.uk/itm/224955531830

[Specmaster]

Possible bargain for someone today https://www.ebay.co.uk/itm/185519656143

Error 1 on this instrument appears if it detects an error on POST, like corrupt calibration data, normally caused by the RAM battery failing. This meter can be calibrated via the front panel. It is a good performer, 5.5 digit with a 210,000 count.

[bd139]

Yeah saw that one and considered it. Have too many meters already