Tek 2445 with bad channels 2-4 - likely causes?

[uoficowboy]

Hi - I'm looking at buying a Tek 2445 with known bad channels 2-4. Any idea what would cause that to happen? Bad hybrids? How likely is the problem to be fixable? I'm worried about picking it up and finding that the only way to fix it is to get some unobtanium parts.

[c4757p]

Bad how? It could be simple or the scope could be toast.

[David_AVD]

Maybe assume the worst and don't pay more than what it's worth as single channel unit?   

[uoficowboy]

Bad how? It could be simple or the scope could be toast.

Not sure - I'll try to find out.

I ran across this article: http://www.ebay.com/gds/buying-an-oscilloscope-on-ebay/10000000001568756/g.html which specifically says to avoid broken 2445 scopes. Interesting.

The same seller also has a 485 that won't even power up (fan spins - nothing else) that they want to get rid of. I'm thinking there might be more hope for that unit.

[c4757p]

I love "won't power up" units. Power supplies are easy.

I'd say "avoid broken 2445" is only good advice if you can't figure out what is wrong with it - and in that case it's good advice for anything. Those custom ICs are hard to replace, but they're also reasonably robust.

[edavid]

Saying that the 3 channels are bad doesn't help very much... what is wrong with them?  There's not much they share, so it's tempting to say the problem is the channel switch, but who knows.

The ebay advice you quote is obsolete... 2445s and 2465s have dropped greatly in price, and many parts units and parts are available.  It really just depends on the price and how much you enjoy fixing things.  However, you don't want a scope that displays "TEST 04 FAIL XX" at startup, because that means it needs a full calibration, which is difficult.

On the other hand, the Tek 485 power supply is pretty frustrating to deal with, so I'd be more hesitant about that one.  In general the 485 is much harder to work on than the 2445.

[BravoV]

  However, you don't want a scope that displays "TEST 04 FAIL XX" at startup, because that means it needs a full calibration, which is difficult.

+1

A full blown calibration at Tek 244x/246x model might cost so much thats not even funny, that cal alone worth a decent brand new Chinese DSO. 

Unless you can get it at very cheap, imo, its not worth it. Yes, I own a Tek 2465B. 

[edavid]

  However, you don't want a scope that displays "TEST 04 FAIL XX" at startup, because that means it needs a full calibration, which is difficult.

+1

A full blown calibration at Tek 244x/246x model might cost so much thats not even funny, that cal alone worth a decent brand new Chinese DSO. 

I don't think it's quite that bad, but I'd expect to pay around $150 for a new NVRAM and hobby grade calibration, which is enough to buy another 2445, so not cost effective.  Of course you can do it yourself if you have the equipment.

[c4757p]

246x might be tricky because of the higher bandwidth, but calibration for 2445(A/B) isn't too bad. Obviously you need some equipment that many hobbyists are unlikely to have (precise function generator - or a very stable one that can be set against a good multimeter and will not drift, RF generator, fast-rise pulse generator, and a second oscilloscope to aid in esp. the horizontal section), but if you can find someone to let you use theirs for a while the procedure isn't horrible. The tolerances aren't all that tight in most places. They call for some pretty specialized equipment but a lot of it isn't strictly necessary if you understand the point of the exercise.

Obviously if you want a traceable calibration you're up the creek without a paddle, but it's an oscilloscope, not a high precision multimeter.

[uoficowboy]

  However, you don't want a scope that displays "TEST 04 FAIL XX" at startup, because that means it needs a full calibration, which is difficult.

+1

A full blown calibration at Tek 244x/246x model might cost so much thats not even funny, that cal alone worth a decent brand new Chinese DSO. 

Unless you can get it at very cheap, imo, its not worth it. Yes, I own a Tek 2465B. 

I've never had a scope calibrated -  is it hard to do yourself, and is it really needed?

[KJDS]

I've never had a scope calibrated -  is it hard to do yourself, and is it really needed?

I've never had one calibrated either. I've connected up a signal generator to make sure a scope is doing what it should before i make a real measurement and that's good enough for me.

[c4757p]

I've never had a scope calibrated -  is it hard to do yourself, and is it really needed?

It's not hard if you have the equipment and the service manual. The manual will usually just walk you through the process - the hardest part is keeping track so you don't get lost! The adjustment section of the Tek 2445A manual (don't have the manual handy for the plain 2445) is 21 pages long, and a couple of those pages are tables with a dozen or so steps each. It's easy but tedious.

No, it's usually not needed if the scope is working fine. I'm fairly obsessed with keeping all my equipment in the best shape I can manage, so I like it to all be perfectly calibrated, but that's a bit more along the lines of "masturbating to error measurements" than "necessary". YMMV, of course, with a used scope that may have God knows what problem.

Obviously the qualifications for "having the equipment" increase with the capabilities of the scope. I've done numerous 20 MHz clunkers with no more equipment than any average hobbyist might have (a couple without even the manual), but I had to improvise a few techniques and instruments to do a 2445A.