Test Equipment Anonymous (TEA) group therapy thread

[DC1MC]

This one also went up for Saskia 

https://www.ebay-kleinanzeigen.de/s-anzeige/akai-gx4000d-coil-recorder-ausgezeichneter-zustand-funktional/2113559697-168-8068

[AVGresponding]

I was limiting the list intentionally at 3-terminal ones. Otherwise I would have not only included the L200, but also the LM104, LM106, LM2991, LR8K4, SG1501, SG1532, ....
LM150/250/350 is a 3A adjustable between the 117 and the 138.

That's an interesting one!
I've attached the datasheet of it. Is there a similar one of the SG1501A from Linear Technology?

And for entertainment purposes, I've added an issue of Electronic Design from 1973. Please see page 46 for the SG1501A.
Spoiler: it contains also a lot of Tektronix stuff fun. 

Well, I tried but I then saw it has a size of more than 56MByte and the forum is limited to 5MByte only.
So, here is the link: http://bitsavers.trailing-edge.com/magazines/Electronic_Design/Electronic_Design_V21_N14_19730705.pdf

I found page 23 to be surprisingly relevant too.

Other highlights for me were the CROM, never heard of that before, the HP "Snap-On" system, surely they had to change the name of that for copyright purposes, and the ads for thumbwheel selector switches from no less then three different manufacturers.

[Kosmic]

the 100MHz bandwidth is optional. So they ship all the hardware for 100MHz (including the probes) and try to encourage you later to unlock the feature.

Just unlocked mine. Work fine at 100MHz now. The 24M Memory was already unlocked. It's a nice little scope for 65$  Think I'm going to sell the 200MHz Owon I have here.

Last time I measured it like 3 - 4 years ago the unlocked "100 MHz" BW was actually ~ 130 MHz. Be interesting if you would check it too. 

With a fast pulse I got 243MHz / 1.44ns rise time (to be taken with a grain of salt).
With a signal generator I get to the -3db point at 177MHz.

Hardly likely, recheck your work.....if it's for the 100 MHz Rigol.

It's not like the setup is super complex. Tested with a IFR 2025 starting at 0dBm (632mVpp) 10MHz. at 177MHz I reach around -3bBm (447mVpp).

[xrunner]

It's not like the setup is super complex. Tested with a IFR 2025 starting at 0dBm (632mVpp) 10MHz. at 177MHz I reach around -3bBm (447mVpp).

Pretty simple yep. I just re-checked mine it's -3 dB @ 140 MHz using the same mV readings you did. Looks like they made some changes ...

(Using Agilent 8648A)

[m k]

Who knew, who should have known.

[capt bullshot]

Who knew, who should have known.

At least, it looks quite similar to mine. But wasn't able to recognize because of the different case.

[25 CPS]

Today, I decided to do some "ham Technicianing"!

I scratched my bum, donned my bib & braces overalls, picked a few fleas out of my beard, & headed off to the "shack".
Once there, I let off a stentorian fart, (my dog sniffed the air, & quickly vacated the scene), settled down on the long suffering "operator's chair", moaned loudly, & set to work.

The"patient" is an elderly Yaesu FT707 transceiver.

This rather nice looking device has a strange fault, with both the "carrier" control which adjusts output power in the CW (morse) mode, & a similar one in the SSB mode associated with the "mic gain" control.
In both cases, the control seems to go from "zero" to "flat out", if the control is moved, which makes the controls, & hence, the radio useless.

My main fear was that both pots were faulty, & initial tests, looking back at them, seemed to bear this out, with the resistance between the moving contact & the chassis, measured with my Fluke 77, jumping around as the pots were turned.
This was worrying, because they are pretty much "unobtainium".

After digging deeper into the manual I downloaded, I found several pages with better representation of the circuitry involved, so I resolved to try again, today.
This more thorough check still showed the resistance reading "varying all over the place", but, knowing that the "77" was a bit ancient (& digital) , I decided to call upon an even older device-------my HP 410C!

On Ohms, the 410C showed the resistance varying smoothly on both pots, ----What the hell?
Trying the Fluke again, showed the previous situation.

Slowly, my ancient solder smoke addled brain found the answer ---"Dawk!" the Fluke was auto ranging, hence the weird readings!!
Pushing the range hold button changes the readings whilst turning the pots  to something similar to that of the 410C!

Definitely, a "Trap for Old Players"!

If the pots are ok, what's your next suspect?

[AVGresponding]

Here's one for the HFM?!? file:  https://www.ebay.co.uk/itm/304515527042

[capt bullshot]

Digi:



https://www.campaignasia.com/article/case-study-digi-improves-its-brand-perception-in-malaysia/218443

Mess:



Digimess:

[AVGresponding]

They are what's left of the Grundig TE division afaik



EDIT: They could definitely have come up with a better name.



EDIT2: The seller has quite a few similarly madly priced listings, with only 3 feedback score? Hmmm.

[mnementh]

I also got a high voltage transfo to rewire and was planning to do that by hands 

Funny that nobody made a cheap CNC winding machine. I might look into that.

There are a number of 3d printed ones out there. Usual threaded rods, steppers, extrusion + 3d printed parts.

This one looks very interesting, but as of the last time I checked a while back he hadn't really posted any further details.


-Pat

Yeeeah... that there is quite a project. Easily a month's worth of work, maybe more. But obviously, it's a substitute for a tool that's pretty much unobtanium for hobbyist use.

mnem
 

[bd139]

That digimess stuff is Chinese rebrand garbage. Not even decent Chinese garbage but rock bottom stuff 

[factory]

Apparently they had to raise the price to make back for their losses on the $82 ones.

Same link but different seller. This one is from "Electro-Meters Co. Ltd.." and I bought from "Spreetail".

The product description is great.  Only good for 50mhz, oh well at the probes have three times the bandwidth. 

Brand:Riorand; the ds1054z series is the new economical digital oscilloscope to meet the customer's application demands with its innovative technology, industry leading specifications, powerful trigger functions and broad analysis capabilities. 50 mhz bandwidth with a 7-inch display and ultravision technology as well as a host of options. Add the optional analysis, decoding, deep memory, and integrated 2 channel waveform generator for a powerful 4 channel scope at an exceptional price. Standard probes: rp2200 150mhz bw passive probe:4 sets.

David

the 100MHz bandwidth is optional. So they ship all the hardware for 100MHz (including the probes) and try to encourage you later to unlock the feature.

Just unlocked mine. Work fine at 100MHz now. The 24M Memory was already unlocked. It's a nice little scope for 65$  Think I'm going to sell the 200MHz Owon I have here.

I was referring to the mhz typos on the bigriver selling site (that I rarely ever use), also like the error with the brand name. 


David

[mnementh]

VOLTAGE REGULATORS...

Some thoughts:   with modern components, you won't find any numbering system worth remembering.

At a certain age of the components, just like the stuff you're dealing with right now, one could find quite an amount of generic numbers like 7805 and everyone knew it's a 1.5A 5V regulatur in TO220 case, as long as it itsn't. One often could easily identify manufacturers and their product lines by the prefix letters, while the postfix (or middle) letters or the first digit of the number identified special characteristics of a device.

A certain number of these jelly bean devices still exist, are produced and used today.

Most young engineers don't even know about the above, they tend to find a special purpose IC for every purpose, often recommended by the manufacturer or their representatives. Their main focus is to sell the newer parts, since they all are more expensive than ye olde jelly beans. Except you buy a gazillion of them, like automotive manufacturers.   So, today there simply exist way too many different devices to have any kind of intuitive numbering system.

For the same reason - you don't have any switching regulator - I can't recommend getting a switching regulator that would be a good choice to keep some in stock. The old ones are poor in todays metrics, or quite complicated to use or both - while the modern ones have way to many variants, each of them designed for a small scope of applications.

In contrary, keeping some generic 78xx and 79xx stock is useful, because they can easily used in lab prototypes or bodged repairs.

Precisely the point I was trying to make here:

VOLTAGE REGULATORS

Thanks people for your feedback. I have now enough information to confidently, expertly organize all my regulators : I will just shove them all into the same drawer and call it a day ! 

I do the same. I have one whole plastic shoebox that is nothing but VR ICs sorted by form-factor into baggies and salvaged or cheap buck, boost, & buck-boost DC-DC converter modules. I treat them all as electronic Legos.

mnem
https://youtu.be/pNP5lFsZDQ8

Between the LM1117A-xx family of LDOs and the 78/79xx family of linear regs, you can whip out almost any form of temp power supply in 30 minutes, and with the grenade-proof design of the 78/79xx in particular, you can be pretty confident it'll be clean and stable with just a couple brute-force filtering caps. Plus, they are easily misused to serve as a current regulator as well. Winner, winner, pork-chop dinner.



Many of the current Chinesium adjustable buck converters can be treated similarly... add a couple reasonably large electrolytics (they usually already have some MMLCCs at VIN & VOUT) and you have a reasonably clean adjustable power source for almost any project, plus the efficiency is plenty high for battery operated projects or ones where your original source of VIN cannot afford the waste of linear.

Search term Mini360 or MP1584 0r MP1584EN will yield oodles of hits; usually ~80c each in Qty 10.

There is a new family of these DC-DC converters based on the MP2307 and MP2338; I have years and years of experience with the MP1584EN which shows it to be very durable in harsh, even abusive and very broad operating conditions.

What I did notice on the newer designs is A) Tiny SOT583 package vs SOIC8, 2) Lower operating frequency - 650Khz & 450Khz vs 1.5Mhz for the MP1584EN, and C) Output voltage is much narrower range than the MP1584EN.

For these reasons (and the fact it is a known device) I still seek out the MP1584EN, myself; particularly the D-SUN branded ones. MP1584EN datasheet is attached below.

Then, of course, even these parts are being deprecated by Monolithic Power; the current parts are MP2338 & MP2393, which are not yet common in these little Chinesium bricks.

I still avoid the LM2596-based buck converters... I've read elsewhere that there's been yet another round of the fakes that fail at much less than rated load, and yup... they still fail short VIN--->VOUT. 

No doubt somebody unearthed a warehouse full of the original fake LM2596s from back when, and decided it had been "long enuf" that they should be able to unload them instead of dumping them in the incinerator as they should have been years ago.

https://www.monolithicpower.com/en/mp2307.html

https://www.monolithicpower.com/en/mp2338.html

https://www.monolithicpower.com/en/mp2393.html


mnem
Monolithic Power is just monolithic.

[TERRA Operative]

DOPAMIIIINNNNNEEEEEE!

I now have a fully working TDS794D!
I tracked down two bad acquisition RAM chips, U405 and U413 on channel 2. Replaced them with parts pulled from a scrap board (a smashed unit with a physically damaged acquisition board, well and truly for pulling parts..)

Now it passes all self-tests and no spurious spikes on the display. Yaaaaaaaaayyyyyy! 


Next step, find probes at a not-insane price.....

[Vince]

Oh no...... oh help me..... it's JUNK CONSUMER ELECTRONICS REPAIR TIME again !!! 

A friend just bought me this compact  "retro looking" stereo system. These things are trendy I am starting to gather. It's a good friend and I owe him much, so I must absolutely go the extra mile to fix this thing.
Plus, it comes from his wife's work place. she works in my village's retirement home. She uses that thing to entertain the elderly. They are not allowed to use the thing, normally it's her that operates the stereo. Also, it appears it still available for sale currently, and for whooping 120 freaking Euros ?!  That's a lot for something that's worth 2 Euros tops to me... if even that.
So, a nearly new 120 Euros unit, that's extra motivation for this rescue mission.

She said that the problem is that when she puts a USB stick with MP3 files on it, it doesn't work anymore. Just nothing happens, as if she hadn't inserted the stick. I confirm that.
There is also an SD card reader with the same purpose... so I put some MP3 files on an SD card and... same result, nothing happens.

The controls aren't exactly intuitive, to me at least.. I never find consumer electronics intuitive, there UI never "speaks" to me somehow 

So to make 100% sure it was not an operator error, I searched for the user manual... no luck.
It's sold under the distributor/ retailer's own "brand" : " essentielb ". No typo. the trailing 'b' stands for  " Boulanger ", the name of the distributor. Model " Madison BT ".
while failing to find a user manual for it, Google did find something else that was interesting : the manual for a JVC RD-F327 that, although has slight UI difference : top row of buttons merged with the bottom row, plus they added a remote control capability, that's about it. Clearly the same unit. So the manual says that in order to read MP3 files from an SD card or USB stick, all you have to do is select SD or USB, simple enough, then insert the media and that's all, it should start playing music automatically, no need to press any button.

So clearly, it doesn't work.  The fact that both the SD card and USB don't work, would suggest that the problem is common to both, rather both having their own failure. So obviously they have in common the CPU.. which runs fine since the stereo works just fine others than USB and SD cards. So either some I/O in the CPU are fried somehow. Or maybe the SD and USB are not wired wired straight to the CPU and there is some discrete components involved. Problem is, no schematic.
Oh forgot to say. USB and SD connectors are soldered directly to the main CPU board, close to the CPU. So there is no interconnect problem here.
Also, the USB stick I used happens to have an LED built into it, that flashed when there is activity.. and that LED is dead as a dodo.

So I am thinking that the problem might be that the stick and SD card are simply not detected at all and the CPU is therefore not even trying to communicate.
But  again I don't know how these things work and no schematic, so no idea what to probe, what supporting external components there might be around these connectors, aside from the CPU per se.

Luckily the main board is fairly easy to remove / work on, once you remove the turn table assembly and the CD player transport, and cut some cable ties and remove lots of plugs... so at least I can do a close inspection of the board, try to trace signals, scope things, inspect solder joint, solder stuff.... I can work on that board. So it's not totally hopeless.

Below a High-res pic of the board, 5MP, for the good cause, in case someone catches on something. I asked for a derogation for this one 5MP pic, of course. Still I compressed it, so "only" 1MB or so.

Any help welcome... if you can find schematics for this thing, or know how a USB connector and SD card might be wired to a CPU... what supporting discrete components that might need, that might fail... I am all ears.

In the meantime I will stat digging...

Power supplies : unit gets 12V DC from an external brick. Then there is a tiny power board at the back of the rear panel that carries a 7805. The 5V and 12V are then sent to the main CPU board. On that board I see another regulator, a 78M06. So the board has 5V, 6V and 12V.

I checked the main 12V supply at the barrel, it's fine. Need to check 5V and 6V now, but I am not expecting them to be bad... at least not the 5V or else the freaking CPU would not even work.

CPU BTW is an " ALI   M5675 A1 ".

I will search for the datasheet see how if they give a typical wiring schematic for the SD card and USB... but I doubt it would involbe an additional 6V supply. So I don't think it's as easy as a dead supply. I fear a fried I/O in the CPU, hence not repairable...


Thanks for your attention, for the one or two that read all that.

[mnementh]

Congratulations!

But you have to wonder .... how come the price?

Best never to ask questions like that 

Kosmic doesn't have to ... he's come out a winner.


Me...?  Just curious.  Could be anything from "fallen off the back of a truck" to a surprise inclusion in a bulk acquisition ... or maybe a part of a "Storage Wars" episode?

I had another thought last night...
Fulfillment by Amazon with the seller making a typo in the price.   ???

Yeah, that was my guess too, hence trying to buy 4 of them. 

mnem
*flippity...?* [DENIED!]

[bd139]

Seriously? £80?

https://www.ebay.co.uk/itm/384932494888

[mnementh]

Oh no...... oh help me..... it's JUNK CONSUMER ELECTRONICS REPAIR TIME again !!! 
(SNIP)

I'd start looking at the DC rails for noise first; a cheap cap not doing its job anymore. Then anything in the serial data between the USB port/SD card and the CPU. Don't assume it connects direct; there may be a inverter (as simple as a couple transistors) or a level-shifter in those data lines.

mnem
Virtual

[bd139]

Oh no...... oh help me..... it's JUNK CONSUMER ELECTRONICS REPAIR TIME again !!! 
(SNIP)

I'd start looking at the DC rails for noise first; a cheap cap not doing its job anymore. Then anything in the serial data between the USB port/SD card and the CPU. Don't assume it connects direct; there may be a inverter (as simple as a couple transistors) or a level-shifter in those data lines.

mnem
Virtual

Looking at the lack of ESD protection I bet the USB controller die has a hole in it.

[Carl_Smith]

So for months now I have been watching ePay for a good deal on a Tek THS720 or 730 scope.  I want one with isolated channels so I can watch both windings of a bipolar stepper motor or put one channel on a motor on an H-bridge while monitoring logic that drives the H bridge on the other channel.  So today I came across this listing.  Looks like it comes with a rather interesting power line monitoring probe. 

https://www.ebay.com/itm/115387055667

[mnementh]

Digi:

mnem
No relation.

[Vince]

DOPAMIIIINNNNNEEEEEE!

I now have a fully working TDS794D!
I tracked down two bad acquisition RAM chips, U405 and U413 on channel 2. Replaced them with parts pulled from a scrap board (a smashed unit with a physically damaged acquisition board, well and truly for pulling parts..)

Now it passes all self-tests and no spurious spikes on the display. Yaaaaaaaaayyyyyy! 

Wow that was quick !!  Your intuition was spot on then, you are getting good at this !

But the report lacks detail : how did you find the bad chips ? Brute force, replacing them one by one till the fault cleared  ? Or you did some more troubleshooting to try to identify the bad chips, if so how did you proceed ? Maybe just look at the data bus of each RAM chip searching for the ones that are stuck on FF, simple as that maybe ?

Congratulations in any case !

[Vince]

Between the LM1117A-xx family of LDOs and the 78/79xx family of linear regs, you can whip out almost any form of temp power supply in 30 minutes, and with the grenade-proof design of the 78/79xx in particular, you can be pretty confident it'll be clean and stable with just a couple brute-force filtering caps. Plus, they are easily misused to serve as a current regulator as well. Winner, winner, pork-chop dinner.



Many of the current Chinesium adjustable buck converters can be treated similarly... add a couple reasonably large electrolytics (they usually already have some MMLCCs at VIN & VOUT) and you have a reasonably clean adjustable power source for almost any project, plus the efficiency is plenty high for battery operated projects or ones where your original source of VIN cannot afford the waste of linear.

Search term Mini360 or MP1584 0r MP1584EN will yield oodles of hits; usually ~80c each in Qty 10.

There is a new family of these DC-DC converters based on the MP2307 and MP2338; I have years and years of experience with the MP1584EN which shows it to be very durable in harsh, even abusive and very broad operating conditions.

What I did notice on the newer designs is A) Tiny SOT583 package vs SOIC8, 2) Lower operating frequency - 650Khz & 450Khz vs 1.5Mhz for the MP1584EN, and C) Output voltage is much narrower range than the MP1584EN.

For these reasons (and the fact it is a known device) I still seek out the MP1584EN, myself; particularly the D-SUN branded ones. MP1584EN datasheet is attached below.

Then, of course, even these parts are being deprecated by Monolithic Power; the current parts are MP2338 & MP2393, which are not yet common in these little Chinesium bricks.

I still avoid the LM2596-based buck converters... I've read elsewhere that there's been yet another round of the fakes that fail at much less than rated load, and yup... they still fail short VIN--->VOUT. 

No doubt somebody unearthed a warehouse full of the original fake LM2596s from back when, and decided it had been "long enuf" that they should be able to unload them instead of dumping them in the incinerator as they should have been years ago.

https://www.monolithicpower.com/en/mp2307.html

https://www.monolithicpower.com/en/mp2338.html

https://www.monolithicpower.com/en/mp2393.html


mnem
Monolithic Power is just monolithic.

Thanks for the report, looks like these little modules are worth having indeed, will add that to my shopping list. 

[mnementh]

Oh no...... oh help me..... it's JUNK CONSUMER ELECTRONICS REPAIR TIME again !!! 
(SNIP)

I'd start looking at the DC rails for noise first; a cheap cap not doing its job anymore. Then anything in the serial data between the USB port/SD card and the CPU. Don't assume it connects direct; there may be a inverter (as simple as a couple transistors) or a level-shifter in those data lines.

mnem
Virtual

Looking at the lack of ESD protection I bet the USB controller die has a hole in it.

Looking again at the SD socket, it appears it is directly connected to the CPU through a pretty simple array of buffer caps/resistors. While it's possible one of those failed without killing the associated serial data pins on the CPU, unfortunately unlikely.

I've seen similar failure in the card reader/USB on numerous Digital Photo frames, a lot of which use the ALi CPUs cuz cheap as chips... and rarely not terminal. Unsure if poor design or poor implementation (ie not using a level-shifter when should be, etc).

mnem