Test Equipment Anonymous (TEA) group therapy thread

[factory]

It's bad, very bad, mainboard has 20+ surface mount ICs and many other SM parts, all freshly marinated in Yuasa juice. 

Pictures tomorrow....

David

[TERRA Operative]

Anyone who has any Yokogawa gear, or interest therein, I've just set up a shiny new groups.io for Yokogawa stuff.

https://groups.io/g/Yokogawa-Test-Equipment

I just expanded the group to include a few more Japanese brands, because why not?

https://groups.io/g/Yokogawa-Hioki-Anritsu-Kyoritsu

At this point you might as well call it something generic like "Japanese TE" so that you don't have to rename it every other months when adding another brand 
It would also be easier to remember the name of.

Yep, now it's https://groups.io/g/Japanese-Test-Equipment/

Forgot take pictures yesterday, there is a lot of green crud on the connector and what might have been a display contrast adjustor on the side. 

https://www.eevblog.com/forum/testgear/test-equipment-anonymous-(tea)-group-therapy-thread/?action=dlattach;attach=1905480

https://www.eevblog.com/forum/testgear/test-equipment-anonymous-(tea)-group-therapy-thread/?action=dlattach;attach=1905486

https://www.eevblog.com/forum/testgear/test-equipment-anonymous-(tea)-group-therapy-thread/?action=dlattach;attach=1905492

Inside is quite a mess, death by Yuasa  , hopefully I can fix it, guessing all the pots & switches are trash, some track repair needed at the edges.
Can't seem to get the mainboard to separate from the display board, looks to have lots of surface mount bits underneath. 

https://www.eevblog.com/forum/testgear/test-equipment-anonymous-(tea)-group-therapy-thread/?action=dlattach;attach=1905432

https://www.eevblog.com/forum/testgear/test-equipment-anonymous-(tea)-group-therapy-thread/?action=dlattach;attach=1905438

https://www.eevblog.com/forum/testgear/test-equipment-anonymous-(tea)-group-therapy-thread/?action=dlattach;attach=1905462

https://www.eevblog.com/forum/testgear/test-equipment-anonymous-(tea)-group-therapy-thread/?action=dlattach;attach=1905474

More pictures below, couldn't find any internal pictures of the S3 last night, Dataman don't give out circuit diagrams either. 

https://www.eevblog.com/forum/testgear/test-equipment-anonymous-(tea)-group-therapy-thread/?action=dlattach;attach=1905561

David

We need a vomit emoji....

[tautech]

We need a vomit emoji....

Here, borrow this one:
🤮

[vk6zgo]

OK OK....

I wonder what the heck was running at 18GHz back in 1968 ?! 

I mean today we have wireless this and that, GPS, cell phones, Wifi and XYZ, that run in the GHz zone... but back in 1968, what area of technology was running at 18GHz ?
Experimental voodoo magic hidden in secret underground labs ?
Hugh tech Fundamental physics experiments  ? Particle physics ? some types of radars ?

It's startling that one was able to do stuff at nearly 20GHz in the '60s...

Point to point microwave radio bearers were quite common, though most commercial ones were a tad lower in frequency.
The military probably happily inhabited the area around 18GHz, both for their own links & other more nefarious uses.

The Biggie would be NASA.

[Robert763]

That's a fair price, the DVM module goes for at least half that price on it's own.
Last one here that wouldn't count required the multiplier board adjusting, exactly the same as CuriousMarc found.

David

That S3 is dead as a doornail. If the corrosion has gone that far the PCB will be toast.

[m k]

Based on pictures the board seems be pretty fine.
Maybe it has been cleaned so thoroughly that corrosion has had clean tips to start away from the board.

[factory]

It's bad, very bad, mainboard has 20+ surface mount ICs and many other SM parts, all freshly marinated in Yuasa juice. 

Pictures tomorrow....

David

It's grim on the other side... 





Closer view showing the only custom part on the mainboard, also one part came off with the tape.



Display board, probably a standard off the shelf part (I've not checked this).



Texas instruments keypad, looks non-serviceable.

That S3 is dead as a doornail. If the corrosion has gone that far the PCB will be toast.

Well the board is about the right size to have come out of a toaster. 
I couldn't find any internal pictures on the web the Dataman S3 two days ago, now there are some at least.

David

[Vince]

Ah crap, that side of the board looks "great" as well..... hum hum...

Still, although it it's a lot of green, from your pics it does not look like it's that bad...

If you don't want it anymore I will be happy to give it a go, send it my way 

The component that got stuck on the tape, did yo manage to recover it so you can at least know what type of component it is, and try to meausure its value so you could replace it ?
Or maybe it's still possible to solder it back to the board...

I would just clean all the green, reflow all the IC's, buzz all the IC pins to make sure they are still connected to their tracks, and power up the thing...see what happens and go from there.

Keep the faith David...

Well I guess it all depends how easy these devices are to find, and how much money they fetch if they are working. I have no idea.
If one can find one at every street corner, working, for 50 quid,  then I guess it's not worth spending too much time on it, but in any case I would at least spend half a day on it just in case you get lucky.

If you give up on it, you can send it my way, I would be pleased to give it a go as well 

[factory]

Cleaning would reveal how many traces are eaten away, would need to find out what is best to use on NiCd pee, note the display board vias are not coated, probably mostly open circuit.

Detached component is still on the tape, shown one of the attached pictures, the pads came away with it.

The S3 doesn't seem that common, none for sale ATM, only two completed listings at approx £40 each, neither shown powered up (but no visible corrosion), the S4 seems more common and makes more £.

I think I was very lucky with those two 1990s Toshiba laptops, neither had any serious corrosion from the three batteries in each, OK one needs rework to the screen board, which might be related to capacitor pee.

David

[Vince]

Oops, sorry somehow I didn't click on the tape picture....   

OK so that's great news then, since the pads came off the board then the component itself is physically intact so you can reuse it, that's fantastic.

Google found this article that lists methods to clean the corrosion for various types of batteries :

https://www.thespruce.com/how-to-clean-battery-corrosion-5073244#toc-how-to-clean-nickel-cadmium-battery-corrosion

For NiCd they say : ".... with dry baking soda to neutralize the potassium hydroxide in the corrosion. ".

[factory]

Thanks Vince, we seem to have some bicarbonate of soda (aka baking soda) gathering dust in the kitchen.

I should probably have put the tape picture inline.

David

[Robert763]

Not sure how bicarbonate of soda will neutralise potassium hydroxide as they are both alkaline.....
In the aircraft battery shops we used bicarbonate of soda in the lead acid shop to neutralise electrolyte (sulfuric acid) and acetic acid (vinegar) to neutralise potassium hydroxide electrolyte in the NiCad shop.

Robert.

[factory]

Just had a closer look at the cleaning article, showing the corroded batteries in a camera, they are alkaline AA from Pairdeer, someone at work uses these exact batteries, definitely not Nicd.
Then at the end of the cleaning article some Nicd go in.
Very odd, so baking soda for alkaline? And vinegar for Nicd.

David

[tautech]

@TERRA Operative
Buddy member @Bravo acquired a TDS5054 with the view of convincing it to become 5104 to extend capabilities beyond his SDS2504X Plus.

Tek Wiki
https://w140.com/tekwiki/wiki/TDS5054

It's a PC based scope as you probably know and his initial plans are to backup the HD onto a SSD.
Are HW mods required to extend BW and a model # change ?
Can you offer any further guidance for improvements ?

[TERRA Operative]

Unfortunately I haven't got to the TDS5000 etc series yet, they still command a good price here in Japan so I haven't acquired any yet...

[Vince]

It's junk repair time again this week-end... yesterday a friend bought me his bathroom radiator, a large one with electronics : comes with a remote control bearing an LCD display and 2 dozen buttons.
IT's just a freaking radiator FFS.... 

Anyway. He bought it 15 years ago so we thought... again some old filter cap that needs to be replaced. Indeed it needed to be replaced, was completely dead, the chinese component tester had a hard time even figuring this one out.

Sadly the repair was unsuccessful because there was a bigger, more fundamental, and rarer problem : unlike all oher radiators I fixed, this one had a much more luxurious / expensive power supply design : normally it's just a zener diode with big dropper X2 film cap, but not here. Here we have a nice molded transformer, and full diode bridge, and wow, no less than 4 opto-couplers, 2 relays and two big ass SMD transistors to drive their coils. Way beefier than I have ever seen elsewhere.

The micro is located on a remote board, the front panel board, that carries the buttons and LEDs and buzzer.

Anyway, to cut a long story short.... the transformer has just one secondary, marked 24Vac, but I measured it at nearly 40 Volts ! 
The filter cap I measured very shy of 50 Volts across it ! It's rated also at 50V.... not good eh ? This powers the relay coils, which are rated at 24V not 48....
Then from that, they derive 5V for the micro. Only way I see they could be doing that, was a tiny TO-92 package but... it was not a regulator, it was a BC337 NPN transistor... which means they are doing the regulation "by hand", and there must be a 4.3V Zener diode next to this transistor...

Anyway, the transistor was dead and was sending 15V to the micro... yes, it was not happy about that....

So basically, somehow the transformer failed, produced a much higher voltage than was planned, and it blew everything downstream.

Radiator is 15 years old but still produced to day yet... yet the boards are not available ?! Is this a joke ?! What kind of a manufacturer is that...
So I declared this one dead, and gladly so because it takes so much space my bench is unusable for anything else, so it better get the hell out of my house ASAP. Friend said he would pick it up today 

Anyway, it's the first time I see these small molded transformer fail, is that a thing, or am I just unlucky ? 

[factory]

Doubt the transformer has failed, more likely the failure of the rest of the circuit means there is no load on the TF.
Having no load on a TF you should always measure a higher voltage.

David

[Vince]

I don't kn,ow... it's such a tiny transformer, and there is not much of a load to speak of : when no LED's are lit and none of the relays (and associated opto-couplers) are energized, there is just the load of the micro running, not much then... so the transformer is supposed to be happy with a minuscule load.
Plus, at the beginning of the troubleshooting, there was still some life in this radiator : filter cap was not yet 100 dead, mcro was stillrunning, LED's lighting up, relays clickuing... and it made no difference to the output voltage of the transformer.
It should be able to produce 24DC even when idling, for the relay coils are rated at 24V... not 30 never mind 50V...

[Robert763]

The small moulded transformers are a "thing". It is impossible for a transformer to provide too high an output voltage unless the turns ratio changes. The only way this can happen, without wiring errors, is shorted turns. However that also caused excessive current normally resulting in overheating.
However as David said the voltage will rise off-load. This is the regulation. thes small transformers generally have poor regulation. This example:
https://www.farnell.com/datasheets/3967072.pdf
Has +25% regulation (1.25 times). This gives an AC voltage of 30V and DC of 43V so your measured values are not unreasonable. Most likely the 5V regulator died and killed the controller. But as you know a custom programmed controller in unobtanium except from the manufacturer or a donor unit.

Robert.

[tautech]

I always thought PCB mount transformers were a fine space saving idea until I found one with its pins pulled when the welder it was in fell off its trolley. Even the reasonably beefy zip tie around it didn't save it and probably added to the almost impossible to find myriad of cracked SMD solder fillets on the PCB it was mounted to.
Replacement transformer sourced from RS and the 3rd time back for intermittent fault and reflowed every SMD joint finally fixed the bugger. 

Now gimme one that bolts to a chassis thank you !

[Vince]

Thanks Robert, so you agree with David, adn that datasheet thank yo for it, agrees too... boy 25% extra voltage with no load that's quite something, good to know.
35% loss as well !!    Those small transformers are horrible... but I guess horrible losses are not a big deal with tiny transformers like this, since the absolute  amount of wasted power is still very small.
I guess.

However as I explained in my previos message, I don't understand : if they rely on the minuscule idel load of this circuitry (CPU + tiny/low power 5V regulation circuitry) to drop massively the output voltage of the secondary the.... when the compnents that actually draw a lot of current (relays coils, LEDs, buzzer, LEDs inside the opto-couplers that drive the coils ...) kick  in, then the transformer voltage should completely collapse....

So you got me curious. I don't want to die stupid. So I powered the thing up again... I measured again the output of the transformer, then the rectified voltage across  the filter cap.
It's worse than yesterday : today I get 45Vac at the secondary and 60Vdc at the cap.  This is getting out of hand...

THEN... I created I soldered a resistor straight at the secondary pins of the transformer, to simulate a load : I put a 1K resistor, so should drazwx way more mA than the micro and 5V regulations circuitry used to draw.... yet, the secondary didn't drop very significantly at all. It dropped, sure, but nowhere near the magnitude we want : I got 36.8VVac and 53Vdc ! Still way, way, waaaay too much.

So I don't know.... 

[ch_scr]

Even if the controller is dead, the heating element is still fine? Why not add an external one? The generic chinese ones are 15€ now, e.g. https://www.ebay.de/itm/155785494151. Way cheaper than a new heater?

[factory]

It looks as if it might have a fan too, nearly every heater I been asked to fix has the fan seized, from dried out oil in the bearings.
The last one I failed to fix, as I damaged the motor windings opening it, found out the hard way the rotor only comes out one way, it was only a few years old and had a transformer based PSU too, it's the first I've seen that uses PTC heaters too. It's still here awaiting scrapping, been too busy with other things....

Pictures are of an identical one I did successfully fix, notice the motor stator windings sticking out at the bottom, ready to get damaged by the rotor if removed from the wrong side.

David

[Vince]

EDIT : I experimented somre more. I added more resistors in // to increase the load :

So with x2 1K in  //, doubling the load, 500R then, it drops at bit but still not a lot.

However if I put 100R in // with the 1K, it drops a lot, I get 14Vac !

So about 100R load.  Tranny is advertised as a nominal 24V DC and 1.9VA
so I am trying to pull almost 3 times the rated power... so getting well below the nominal voltage makes sense, I understand, so it adds up here.

so, I lightened the load : needs to be more than 100R and less than 500R (because that didn't drop it enough).
So I added 500R in // with the 1K I had soldered at the beginning. So about 333R then, if I get my math right...
If I do that, I get 25.7ac, almost nominal, yeah ! ... and that gives 34.5Vdc once filtered.

So... OK, you convinced me. You can get 24Vac out of the thing, tranny is not dead... however to get that 24V you need the loa dto be quite precise / controlled, there is a very narrow window 
OK OK... I guess it's not a problem since the 5V regulator will compensate for the varying secondary output, and as for the relay coils rated at 24Vdc, having 35Vdc at idle is not a problem since they are not connected / energized at that time... when they do get energized, the voltage will drop much closer to thei 24V they expect.

So overall... I guess the most likely scenario is what Robert said ? The 5V regulator failed, pass tranny failed / BC336 failed and sent 15V to the micro...

[Vince]

Even if the controller is dead, the heating element is still fine? Why not add an external one? The generic chinese ones are 15€ now, e.g. https://www.ebay.de/itm/155785494151. Way cheaper than a new heater?

The heating element I am not unsure of its health as it measured at 350 ohms or so, hence barely 200Watts... but it's rated at 750W.
As David said, there is a blower as well (which appears to spin, but very slowly, so I am concerned), and a fancy remote control.

I just found out that the manufacturer DOES sell a "solution" to fix this radiator..... they don't sell the boards per se, no, they sell a complete kit to replace the entire assembly at the bottom of the radiator : all the plastics, blower, electronics, new remote control, the whole business... at 150 Euros.  Once you add shipping, that's half the price of a  new radiator... plus the risk that it might now work... some people on forums bought this kit and it was defective, some features didn't work.

http://www.piecesdetacheeselec.com/16040-kit-de-souflerie-soufflerie-blanche-universel-pour-radiateur-%C3%A0-tubes-ronds-aurelia-delonghi.html


So overall it's best to start from a clean sheet (it's 15 years old as well), and get a new one, with a warranty and more life left in it. The old one can be kept for spares...