Test Equipment Anonymous (TEA) group therapy thread

[Robert763]

A return loss bridge or directional coupler allows you to make reflection measurements e.g. SWR. The SA/TG combination only makes transmission measurements between the two ports. It will tell you the gain or loss through a device against frequency.
They are good for adjusting / testing filters, lowlevel amplifiers etc.

Adding a return loss bridge allow you to test single port devices like antennas and dummy loads. It is built into your 8712ET
A HP 8502A or 85044A Transmission/Reflection Test Set is a deluxe version and quite small. Don't buy a B version they are 75R

Robert.

[Peter_O]

snip:
A "Magic Eraser" cleaning sponge can work well on this sort of mark.

A magic eraser with IPA removed the numbers with a few light wipes completely.  This, after a cotton cloth with IPA didn't do anything at all.

Thank you for reminding me!   

[Robert763]

Do you have a return loss bridge or does the spectrum analyzer have 1 built in?

A scalar network analyzer can't measure phase.  So it can't transform the measurement into the time domain.  And since there is no way of knowing the magnitude of the difference of 2 numbers based on their magnitudes, it can't compensate for the cables either.

That is not quite correct. You don't need phase information to convert between frequency and time domains e.g. FFT. What a scalar analyser can't do is differentiate between resistive, inductive and capacitive reactance.
The OP's SA/TG has a normalise function so can compensate for cable reflection test sets etc.

[44kgk1lkf6u]

Is that the same as normalization where I connect the TG and RF input, set my freqs and push the normalization button and then add my DUT

They are not exactly the same.  Whether either one is good enough depends.  If you are only interested in the bandwidth of a filter then you can probably get by without normalization.

The scalar network analyzer normalizes the transmission loss by dividing that with the component by that without the component, or subtracting the log.  If the ports on the instrument are perfectly matched and the cables only dissipate but do not reflect, then this method gives the right answer.  When the connectors reflect some signal, how much signal gets reflected depends on what is connected.  This causes some error.

You don't need phase information to convert between frequency and time domains e.g. FFT.

How?  I thought FFT involved addition.

[watchmaker]

Just for giggles, here is the Tracking Gen on a 30mHZ BPF.  Also, I looked up the NSN cost on this.  $40K!!!  I feel like a thief!!

[rdenney]

The latest acquisition of the T&M variety was an HP350D attenuator. USPS claims it was delivered last Thursday. I'm sure the are right, but I'm also sure that where they delivered it was not at my address.

Aaaaand, of course, USPS doesn't even allow one to open a tracking request until 7 days after the promised delivery data, notwithstanding that a package claimed as delivered might be far more easily found if sought after immediately. I'm going to try to reach the local postmaster today by actual telephone to see if he can at least look up the GPS location where the package actually was located when it was scanned at delivery.

All the delivery services truly do suck at one time or another, but there's something to be said for the UPS habit of photographing the item at the delivery location. USPS could learn a thing or two from that.

Rick "usually USPS doesn't report delivery when it wasn't delivered--they just lose it and go silent" Denney

The HP350D did finally appear--right where my wife had put it the day it arrived. Sigh.

It works fine, or at least it did after I cleaned the rotary switches.

And I finished putting the Thaikits CH-012 curve tracer together. Pics:

I started with the power cord connector. I used the milling machine to cut the rectangle for that device. The one I bought has an integral EMI filter comprising a common-mode filter followed by a differential filter, which I think added about a dollar to the price I paid for it. It also had a fuse holder and a proper power switch, so it was a useful package. I used to hate things like this (and meter openings) until I added a Bridgeport mill to the shop.


The kit came with a little linear +/- 15VDC power supply--I had to buy a transformer for it. I located the transformer centrally so that wouldn't be too heavy on one end. I thought the cheapie Amazon box I bought was abundantly large, but I never seem to buy boxes that are too large even when I intend to. Here is is after the PCB's are mounted and the power supply hooked up, but before making the front panel.


Again, I used the mill to precisely locate the holes on the front panel. (The plate was screwed down to the piece of wood used as a backing fixture when I cut the openings.)


I used coaxial cable to wire up the X and Y connectors, and insulated connector bodies to isolate the chassis ground of the PCB from the case. I made the front panel by drawing it up in Corel Draw, printing, and then laminating it in heavy plastic. I glued that to the metal panel using 3M Type 77 Spray Adhesive. I had space on the front panel which I used to summarize the info on the sparse downloaded material that accompanies the kit, so I wouldn't have to look that up. The small switch on the right was part of a half dozen of them that I bought from Digikey, but these were the worst junk switches I've ever seen. Just terrible. The switch would melt if the soldering iron spent more than about three seconds on the tab, the tabs weren't tinned effectively and I had to flux the crap out of them to get the solder to stick, and if you tighten the clamp nut more than about finger tight it will pull the threaded barrel right out of the switch body. I was able to use one of the six I bought--the other five ended up in the trash in various states of destruction. The other switch was cheapie I bought at Radio Shack probably 20 years ago that I had lying around--I wish I had two. I used a 4-position header socket, similar to what the kit-maker used on the PCB, which is great for inserting a small transistor like the 2N4401 shown here. I'll make a set of pigtails with clips for larger transistors.



Here's the back with the power plug:


I get that these are not much more use than showing pretty pictures on scopes. But I'm curious as to what you guys would interpret about this transistor knowing the currents that were pulled through it (shown on the front panel) and the traces on the scope. I've put cursors at the four corners to show the voltages. Lower left is zero volts on both X and Y. Upper right is 8.9V on X and 44 mV on Y. I'm thinking this means the hFE gain is about 260, which pretty closely matches the data sheet. (I would have used my Tektronix analog scope for this, but the XY setting on the selector switch seems to be in need of cleaning.)


Rick "comments welcome" Denney

[Robert763]

Anybody ever used an Infratek power analyser? Specifically a 106A. I made an offer on one earlier and it was accepted  Looks very nice. Single phase model DC to several kHz. Has PC control. Manual and software avilable from Infratek. Looks like it is still available. Looks like a better interface for harmonics than my Voltech PM1000.

[jimolson]

I just stumbled onto this Danameter thread from last year.  I can shed some light on the topic.  I am a Dana Laboratories engineering alum from 1974 when the Danameter 2000 first went to production.

DanaLabs was my first paying job out of my undergraduate electrical engineering program.  I worked on the 5900/6900 high precision instruments under John Brady and Dick Payne, but my lab bench was adjacent to John Crosby's where all Danameter engineering took place.

About 10 years older than me, John Crosby had come to Dana from Beckman Instruments. He spent part of his later career at Keithley Instruments. John is one of the most brilliant engineers I met during my career.

The "Damn Nameter 2000" (as John called it) was a frustrating project because it was truly state-of-the-art and had the high warranty return rate to prove it.  It operated for months on a single 9V cell when competing handheld DMMs ate their batteries like candy. I recall that the plastic cam that pressed PCB-mounted switch contacts had a lot of early field failures, driven by users who measured high voltages in the ohms mode. Overload protection is hard to implement in a voltmeter.

The Danameter project was doubly frustrating because the company was losing gobs of money on it due to inexperience selling a commodity item through distributors.  In 1974 the average price of Dana's instruments was about an engineer's monthly salary.  The Danameter was priced at $200--a rounding error to company accountants. 

Crosby once said that it would have been cheaper and more rewarding for Dana to send [a $100 bill in 2025 dollars] to a prospective customer looking to purchase a Danameter.

The unit contained CMOS ICs custom-designed for Dana by Micropower Systems of San Jose.  I remember that an engineer from Micropower would frequently fly down to Irvine from SJC on Air California in the morning, work with Crosby all day in our lab, and return to SJC at the end of the day.

Danameters were built in an area of Dana's Irvine manufacturing building set aside for the product.  The Vietnam War was going on and the assembly line was staffed by newly-arrived Asian immigrant women who chatted in their native tongues while stuffing PCBs.

Much of Dana's sales were to the Pentagon or to military subcontractors building precision electronic devices, so the economic downturn that followed the end of the war in 1975 devastated the company.  In the late 1970s (or early 1980s?) the company was sold to a European T&M maker, Racal.

Danameter: a great product from a great little company!  At my midwestern wife's insistence, I quit Dana in 1976 and we returned to her native Indiana.

[KrudyZ]

Anybody ever used an Infratek power analyser? Specifically a 106A. I made an offer on one earlier and it was accepted  Looks very nice. Single phase model DC to several kHz. Has PC control. Manual and software avilable from Infratek. Looks like it is still available. Looks like a better interface for harmonics than my Voltech PM1000.

(Attachment Link)

Bane of the TEA addict,
the accepted low ball offer...

[Robert763]

Anybody ever used an Infratek power analyser? Specifically a 106A. I made an offer on one earlier and it was accepted  Looks very nice. Single phase model DC to several kHz. Has PC control. Manual and software avilable from Infratek. Looks like it is still available. Looks like a better interface for harmonics than my Voltech PM1000.

(Attachment Link)

Bane of the TEA addict,
the accepted low ball offer...

Indeed. I have a 210kg pallet arriving on Thursday. Trouble is it's only one item 
Not really TEA. It's a laser based fluorescent scanner for cromatography gels and phosphorescent radiation detection screens.
I don't even have equipment to move it. It will be scrapped where the delivery driver drops it.
It has 3 nice lasers, red green and blue in it but one or more may be faulty. also photomultipliers with EHT PSUs and a big X-Y stage with linear bearings. Main reason for purchase is the linear bearings that I want for a mechanical project. Biggest issue with scrapping it is the case and covers are made from high density rigid polyurethane foam. This is tough, bulky and hard to dispose of.
I offered 1/10 of the asking price. Delivery was more than purchase. This was a $220,000 machine 20 years ago. The company I worked for back then made machines using the same technology. 

[Peter_O]

snip:

Might be as simple as a bad connection for the sense wire(s).

And as simple as that it was.    Thx!

Meanwhile I've rewired the transformers from 220V to 240v and calibrated the baby, which went fine.

The only issue left is the left ampere meter damping switch, which seems not to switch off the damping all the time, resulting in a 2s average reading, but with no impact on the CV or CC regulation speed. I'll care for that later.

[Peter_O]

What about the power consumption of these linear lab power supplies?

Here's a quick test to compare the TTi PL330 QMD with a Siglent SPD 3303X-E, both 30V 3A max per channel.

Output V	Output A	Output W	Input W TTI	Input W Siglent
0 V	0 A	0 W	20 W	11 W
1 V	0.1 A	0.1 W	25 W	14 W
10 V	1 A	10 W	71 W	41 W
20 V	2 A	40 W	120 W	86 W
30 V	3 A	90 W	173 W	151 W

[Cyclotron]

Well, I have a shot to join the 8.5 club.  This Keithley 2002 comes to me with "No Comm Link".  The CPU board checks out for the most part that I can tell so I'm going to stuff some EPROMs with the latest firmware and cross my fingers that will solve the issue. Have to wait for my EPROMs to come in though.

I had debated even removing the cover over the analog board, I'm glad I did. It's pretty nasty given how clean they are supposed to be.  So, since I will remove that board for a good cleaning, I figure I might as well replace those Caps. given the habit of these things leaking on the Keithley series.

[d-el]

Congratulations! It looks fresh  . I tried to win it at auction too

[Cyclotron]

Congratulations! It looks fresh  . I tried to win it at auction too

I assumed I had other volt nuts as competition for the unit. I just finished up a good year so I rewarded myself with a project.
I hope it turns out well.  We should all get together like a labor union and vote on who gets the next one so we don't drive up the cost for each other.
 

[Robert763]

The Infratek 106A Power  & Harmonics meter turned up. It's very nice. condition is like new.
It came with a homemade IEC 60360 "PC plug" breakout box that is positively lethal. Earth pin mssing from the inpt connector ane the output to the meter is on unshrouded 4mm plugs 
Works fine Shown with a switchmode PSU as a load.

[TERRA Operative]

Had a HP 8447E delivered today that I picked up for a reasonable price on a whim.
It's essentially a Micro Circuits amplifier and power supply in a HP box, but will be useful here and there. Calibrating and adjusting some of my RF gear calls for this sort of thing.

Besides one little paint chip on the front near the HP badge it's in really good condition. There's nothing really to do to it besides check the power supply capacitors (only two electrolytics that I suspect will test perfectly fine). There isn't even a single Rifa capacitor and the BNC jacks are still shiny as new. Almost boring really.



I also picked up a P6251 1GHz differential probe, but it's faulty. Luckily the seller sold it with a guarantee, so it'll be sent back next week. No huge loss though, I already have two with all standard accessories (including storage cases) and almost all optional accessories anyway.

[m k]

It came with a homemade IEC 60360 "PC plug" breakout box that is positively lethal. Earth pin mssing from the inpt connector ane the output to the meter is on unshrouded 4mm plugs 

Pro stuff, no doubt.

[Robert763]

The HP 8447 range of amplifiers used a range of amplifier modules from various manufactures. That one looks like an HP rather than a Mini-circuits one. I got fed up waiting for one to turn up at a decent price so built my own with a Mini-circuits ZFL-1000H amp, torodial mains thansformer and a Hammond extruded case.

[Cyclotron]

Well, I have a shot to join the 8.5 club.  This Keithley 2002 comes to me with "No Comm Link".  The CPU board checks out for the most part that I can tell so I'm going to stuff some EPROMs with the latest firmware and cross my fingers that will solve the issue. Have to wait for my EPROMs to come in though.

I had debated even removing the cover over the analog board, I'm glad I did. It's pretty nasty given how clean they are supposed to be.  So, since I will remove that board for a good cleaning, I figure I might as well replace those Caps. given the habit of these things leaking on the Keithley series.

And I replaced the EPROMs with the latest code and no dice, still "No Comm Link". I pulled the DALLAS EEPROM DS1243Y+120 and it booted up. I had already acquired a new DS1248Y-70+ to upgrade the memory to MEM2 so I wrote the serial number in ASCII to the begining of the chip and used the guide on XDEV to init the BB-RAM.  It runs and takes measurements but It does not like the EEPROM.

On boot I see errors:
956, Hardware error
511,
513,
514,
518,
515,
512

Running built in test gives only one error which is for the 102.1 EEPROM error.

I assume all the boot errors are related to the 102.1 and for some reason the system can't write properly or read from the EEPROM.
Based on what I've read, the init option should write all the missing data to the EEPROM but for whatever reason, either it doesn't do that or the system can't read the EEPROM.

I have replaced all the CAPs on the boards with NICHICON YV series. I replaced the 4 on the CPU board too as at lease one post indicated that a 47uF cap threw a 102.1 error.

I have cleared the original DS1243Y and swapped it back in with the same errors.  If anyone has thoughts, I welcome them. I guess I'll go start a repair thread and do more research on the 102.1 error.

[wkb]

Had a HP 8447E delivered today that I picked up for a reasonable price on a whim.
It's essentially a Micro Circuits amplifier and power supply in a HP box, but will be useful here and there. Calibrating and adjusting some of my RF gear calls for this sort of thing.

Besides one little paint chip on the front near the HP badge it's in really good condition. There's nothing really to do to it besides check the power supply capacitors (only two electrolytics that I suspect will test perfectly fine). There isn't even a single Rifa capacitor and the BNC jacks are still shiny as new. Almost boring really.



I also picked up a P6251 1GHz differential probe, but it's faulty. Luckily the seller sold it with a guarantee, so it'll be sent back next week. No huge loss though, I already have two with all standard accessories (including storage cases) and almost all optional accessories anyway.

Ohh... it sports a 2N3055 ❤️

[KungFuJosh]

Ohh... it sports a 2N3055 ❤️

Sinisa gets around. 

[tautech]

Ohh... it sports a 2N3055 ❤️

Sinisa gets around. 

And has for many decades. 
He looks much younger in real life. 

[BatGS]

Greetings: Mr Factory/ BatGS:
   Confirming this unit you speak of and included the great images is a THM-565?
   I recently acquired the Mac ET-2020 version you spoke. The shop mechanic told me it functioned then quit after he found the batteries corroded. He cleaned the holder and it still did not function... No surprise there. We know what happened. He said it was buried in the shop many years untill he was cleaning up, which is when it was offered to me... "cheap" so he said. Me, being an electronics tech and service trucks, autos, marine, Etc .. it caught my eye. No time for research, figuring it a simple repair I dropped him from $500 down to $300. Stupid me impulse buy. Well, maybe not all stupid. I have a nice $300 case full of accessories and a tough looking paperweight.
   Now my premise - I read your entries and viewed images. The PCBAs seem to match as do the parts so few I checked. Prior this site I diagnosed it a power supply issue.Your idea of the caps I'm was part of my repair idea. I am concerned of the replacement 1174 regulators. The little information I found has different part #s as replacement. I prefer a DIP as I hate SMD stuff. I have various data sheets on the 1174 however there is so many versions. Did you replace the regulators? If yes, which part #s and source (reference). I realize your 1174 # will vary a little because of the DIP package I choose. I saw it allows Eml from members here. U can just make sure U have EEVBLOG + your username / BatGS in the subject line as protocol proof not spam. If other members have suggestions Follow same Subject line protocol. Thx. BatGS. Mac ET-2020 / THM565.

[TERRA Operative]

You will need to replace all the capacitors inside your THM565 / Mac ET-2020 for a start, and the voltage regulators too if you are unlucky and the bad caps popped them (sounds likely in your case).

Here is the complete list of what I used in my THM565's.

1x	Memory backup supercap	https://www.digikey.com/en/products/detail/elna-america/RSCXJ3345R5F05013U/1658303
1x	Input board	https://www.digikey.com/en/products/detail/chemi-con/APXG250ARA220ME61G/3983377
1x	Input board	https://www.digikey.com/en/products/detail/panasonic-electronic-components/EEH-ZA1V100R/10292372
4x	Power supply section	https://www.digikey.com/en/products/detail/nichicon/GXC1V151MCW1GS/24420287
2x	LCD Panel	https://www.digikey.com/en/products/detail/panasonic-electronic-components/EEU-FR1H100B/9921042
2x	Voltage regulators	https://www.digikey.com/en/products/detail/analog-devices-inc/LTC1174CS8-PBF/889843

Or these voltage regulators if you find them in stock somewhere, they are higher input voltage rated versions (18.5V instead of 13.5V) if you want a bit extra resilience to accidental over voltage, but it's not strictly needed.
https://www.digikey.jp/en/products/detail/analog-devices-inc/LTC1174HVCS8-PBF/888672

If you replace the supercap, double check polarity as the pinout may be reversed on the new capacitor!