Oscilloscopes to stay away from? (spec: Tek 2337 question)

[c4757p]

On the 2465, U800 is the primary problem.  There is a design with smd parts that's available that more or less is a drop in replacement.

Indeed, there is! Damn you, I've already got too many projects piling up! I'd love to try to have a go at this and see if I can do something about the couple problems it supposedly has.

Why the hell would he have blurred out the part numbers in the photo?? They're all in the schematic...

[w2aew]

By the way, does anyone know a good online reference with release dates, manufacturing history, specs and possibly some words about the major Tektronix models? (Like the differences between a 2336 and a 2337, etc.)

http://w140.com/tekwiki/wiki/Main_Page

You might also be interested in this archive of older Tek product catalogs - from 1962 to 1981:
http://bitsavers.informatik.uni-stuttgart.de/pdf/tektronix/catalog/

[PaulAm]

Somebody mentioned the 5000 series.  These are mostly low bandwidth scopes, but they were like the 7000 series with bunches of plugins.  I think most of the mainframes were in the 15MHz BW range, although one went to 50MHz and there was one that had digitizing ability.  I think these are most useful for either audio work or if you have a specific plugin (eg spectrum analyzer or curve tracer) that doesn't need a high bandwidth scope.  Working with high speed logic we tend to forget that there are lots of applications for a low BW scope.

[kfitch42]

Lol it's one of these military ones, right? Looks pretty cool, but I think I'll stay with the civilian models.

Yup, got a navy sticker on it. I mainly went with it for the price. It was definitely treated like a bit of military gear, hence the coax falling of the bnc connectors.

[kfitch42]

By the way, does anyone know a good online reference with release dates, manufacturing history, specs and possibly some words about the major Tektronix models? (Like the differences between a 2336 and a 2337, etc.)

Here is one that is specific to the 233x series:
www.buy17.com/TEK/PDF/tek2336.pdf?

[rf-loop]

The 2465A was probably the best general-purpose analogue scope ever made.

I'd concur with that. They are still highly sort after on ebay.
Don't forget the 2225 is you are doing low level signal work. The 500uV/div sensitivity is pretty much unmatched by any other scope?
http://republika.pl/rubi/oscyloskopy/tek/t.pdf

Dave.

With  Tek2465 (and other 2400 models)  there can use CH1 - CH2 cascading and get 200uV/div
There is CH2 output what give 20mV/div out. (20mV to 1M and 10mV/div to 50ohm.  This can connect to CH1 and get 200uV or 400uV/div. (using 1M or internal 50 ohm impedance impedance )

I have not seen data about cascaded BW/Risetime/noise and I have not measured it.
But B model specs for 2mV/div is 350MHz and  5mV and more, 400MHz

[Kennynva]

Well I seen alot of Tek scopes on here..but not ONE mention of my Tek 2431L,,,,,Is this some kind of a weird scope, in between a 2430..??

[edavid]

Well I seen alot of Tek scopes on here..but not ONE mention of my Tek 2431L,,,,,Is this some kind of a weird scope, in between a 2430..??

You're right, it's not very common.

It was a cost reduced version of the 2432A.

[rstofer]

Most of my FPGA projects run at 50 MHz or so.  To see a decent square wave, I need at least 250 MHz to show the 5th harmonic and that's kind of a minimum if you want to see something resembling square.  The 7th and 9th really help but the 9th requires a 450 MHz bandwidth.

i bought a used Tek 485 about 12 years ago for about $200;.  I got lucky and the scope still works well.  I bought a Rigol DS1054Z for the measurement and decode features (plus 4 channels).  But when the image requires bandwidth, the 485 is the scope to use.

If my FPGA projects get up to 100 MHz or higherk I don't have enough bandwidth, even with the 485.

[tggzzz]

As a beginners CRO with 20MHz BW you won't go far wrong with one of the following:

They won't go right with such a tool. They are only suitable for audio and mechatronics applications. They are useless for digital circuits built using any technology since the mid-70s!

When you have a problem with digital circuits there are three things you need to check:

• signal integrity: are the edges monotonic, are there overshoots/undershoots, is there ground bounce. The period is irrelevant. 100MHz is just about adequate for original TTL; modern TTL requires >=300MHz
• runt pulses: especially near reset inputs and in asynchronous circuits. 100MHz is just about about adequate to spot 5ns runt pulses
• setup/hold violations: a common source of problems. Look at the T_su and T_h spec of any logic family!

And a scope is the wrong tool for RF applications, other than seeing if a signal is present. RF requires sensitivity and linearity; scopes fail both of those.

[tggzzz]

i bought a used Tek 485 about 12 years ago for about $200;.  I got lucky and the scope still works well.  I bought a Rigol DS1054Z for the measurement and decode features (plus 4 channels).  But when the image requires bandwidth, the 485 is the scope to use.

Good choices

The 485 is fairly easy to repair; I've done two. In one a custom IC was faulty, but a little understanding allowed me to repair it with external resistors The only penalty was that the risetime was ~1.25ns rather than 0.9ns.

[rstofer]

i bought a used Tek 485 about 12 years ago for about $200;.  I got lucky and the scope still works well.  I bought a Rigol DS1054Z for the measurement and decode features (plus 4 channels).  But when the image requires bandwidth, the 485 is the scope to use.

Good choices

The 485 is fairly easy to repair; I've done two. In one a custom IC was faulty, but a little understanding allowed me to repair it with external resistors The only penalty was that the risetime was ~1.25ns rather than 0.9ns.

Let's see, 12 years, $200 so about $17 per year.  I think I got my money's worth!

If it breaks, it goes to Electronic Waste Disposal.  I'll just buy another.  Maybe a 485, maybe something with a bunch more bandwidth.  I clearly won't get another 12 years out of a scope (just turned 71) but something will work out.

485's are all over eBay for prices between $200 and about $650 for a unit that is refurbished and calibrated.  Maybe do a little dollar cost averaging to make the numbers look better.  One $200 scope plus one $600 scope over 12 years plus, perhaps, 5 years so $800 over 17 years or around $50 per year ($4 per month).  That makes sense!  Can I cook the books, or what?

The problem with used scopes is that you are buying a 'pig in a poke'.  You have no idea whether the unit works properly or even if it can be fixed.  Assuming you can afford to fix it.  We had one fellow on the forum go down this path and he wound up with 2 defective scopes.  He would have been money ahead to just buy the DS1054Z and be done with it.  Shipping alone is enough to make me think twice.  Given this attitude, it's a wonder I bought the 485 at all.  I probably wouldn't have if the 1054 had been around at the time.

There are some great scopes on the used market.  I'm not going to recommend any of them.  I'm out of the recommending 'used' anything business.  Inevitably, it is a newcomer to the hobby with limited ability to repair a defective unit trying to get a scope for no money.  It is almost guaranteed to end badly.

[danmcb]

I've had three old Tek's. I have a 475 which still works great, but I don't really use. Don't want to get rid of it though, because it's just such a beautifully made thing.

I have a 2225 which is my workhorse - i do a lot of audio and it is great for that. But it had an intermittent fault which the seller hid. Turned out to be a transistor deep inside with a leg that was physically broken. It was attached to the metal can, so that was easy to repair - once I found it.

I have also had a 5103 that I got for next to nothing with an HT fault. I was happy because it had a differential amp, amongst other things. But it turned out to be a dog to work on, and I got shot of it in the end.

So like everything second hand, you can run into issues. Doesn't have to be Tek though - Hitachi, Hameg and Philips also made some great ones. They are cheap enough second hand, if you get a bad one, oh well, better luck next time.

[j ferguson]

I replaced the dallas nvram on my 2445b. and the smd capacitors which had just started to leak, no trace penetrations.  I was very careful in solder sucking but did disturb a trace running past one of the pins but it is still sound.

Now I get flashing lights on control board when I turn it on, but it never gets to self-test.  I tried replacing the dallas nvram but no joy. 

if i do the delta v- delta t slope button push, no effect.

my garage has room,  so I could buy another 2445b, or maybe i should go to a 2445 and get away from the smd problem.

what would you do?

[FlyingHacker]

By the way, does anyone know a good online reference with release dates, manufacturing history, specs and possibly some words about the major Tektronix models? (Like the differences between a 2336 and a 2337, etc.)

http://w140.com/tekwiki/wiki/Main_Page
Or
http://barrytech.com/tektronix/tek.html

Are both great resources for us Tek fans.

[FlyingHacker]

I replaced the dallas nvram on my 2445b. and the smd capacitors which had just started to leak, no trace penetrations.  I was very careful in solder sucking but did disturb a trace running past one of the pins but it is still sound.

Now I get flashing lights on control board when I turn it on, but it never gets to self-test.  I tried replacing the dallas nvram but no joy. 

if i do the delta v- delta t slope button push, no effect.

my garage has room,  so I could buy another 2445b, or maybe i should go to a 2445 and get away from the smd problem.

what would you do?

Join the TekScope yahoo group. Some of those guys are on here too, but they seem to know everything about old Tek gear and how to fix it.

[j ferguson]

I do follow the tekscope list-serv.  I suppose I could have asked the question more clearly.  I have a problematic but otherwise very nice 2445b.  But it doesn't work for reasons not clear.  I could buy another 2445B and might find a deteriorating situation withe the A5 SMD caps AND get to do another  NVRAM ....

OR
 
I could buy a used 2445 and avoid the SMD problems. 

is the 2445 a better bet as far a maintainability over the next ten years or so?  Or is this a silly question?

[edavid]

is the 2445 a better bet as far a maintainability over the next ten years or so?  Or is this a silly question?

Sure, a 2445 (or 2465) has fewer things to go wrong.  Not a huge difference though.

Re: the 2445B, did you double check all the internal cables and make sure they were plugged in properly?

[j ferguson]

i only removed the A5 board to fix the smd caps and remove the dallas nvram, install a socket and the FRAM I replaced the Dallas with.  When it didn't work I replaced the Dallas NVRAM but it didn't make any difference.  it seems possible that one of the ribbon cables may have gone intermittent.  there are also some shakey looking cables which are part of the GPIB option. I guess it would make sense to see if all of these have conitinuity and aren't sorted.  the advantage of getting another 2445b in addition to all the parts is that if it is working now, i can use it to see if my A5 is actually ok.

thanks for the idea.

john

[peter-h]

Just a little update of an old thread, since the situation with these Tek scopes has changed a bit.

The hybrid modules seem plentiful on Ebay, and quite cheap (tens of $) presumably due to plenty of scopes being parted out. A good 2465B may fail but isn't usually expensive to repair.

A late-model 2465B will be mostly as good as new inside.

At work, we spent 4k on a Le Croy 4Gsps digital scope which blew up (some BIOS error message) after 1.5 years. It was done in warranty but if/when it blows up again it will be cheaper to buy a 15 year old Tek digital scope which was 15-20k $ new!

[JimS]

On the 2465, U800 is the primary problem.

I have not seen U800 be bad on any of the 2465B/2467B scopes I have worked on.
I have replaced some U650 display sequencers because the bandwidth was around half of what it should be.
Other than that, is leaking caps in the power supply and occasionally the scale lamps are burned out.
Nice scopes as long as the CRT is not in too bad a shape.
I prefer the 2467B for what I need a analog scope for.

[peter-h]

There are big changes in the quality of the components used in the latest 2465B etc models, compared to ones made say 15 years earlier.

The last ones were made 1994-1996 or so.

I have been in electronics design and production since 1978 and things had really improved in the 1990s, with really good quality SMT components, including the miniature electrolytics. I have many thousands of products out there with the same components and most of them are still working. The actual MTBF numbers we are seeing are of the order of 500+ years. With SMT, things improved a lot because the components had to withstand crazy soldering temperatures, so the chemicals had to be improved.

When I read reports of the WIMA polyester film caps needing to be changed, I simply do not believe them. I have used those caps since the 1970s (they were the primary ways of getting a cap in the say 1000pF to 1uF range; nowadays one would use ceramics if possible) and never saw one fail.

Old electrolytics do dry up and fail, certainly, and in these scopes these will be found mostly in the power areas. But again the most recent scopes should be ok for many more years.

OTOH the very recent, say 1995, examples go for a lot more money. I have just bought a 2465DVS for GBP 250 and it is fully working and a superb scope, but that one will be from c. 1985.