Tektronix 2235A owners, any potential/serious problems for this model ?

[BravoV]

Such as known weak parts or anything else I should aware of ?

I've been offered a working Tektronix 2235A locally here, and I will be allowed inspect and test briefly on spot which is an advantage.

Just curious what kind of common "potential" serious problems that are plagueing this particular model that I should be aware of ?

Weird unobtainium hybrids chip ? Customized hard to find cooling fan replacement ? or any others ?

Really appreciate any advises or suggestions for this particular model, especially on what are the priority check lists if any when I'm going to test and inspect it physically before closing the deal.

Thanks in advance.

[SeanB]

Hybrids and fans yes, but they are reasonably reliable. The capacitors are known to fail and probably will have to be replaced art some time or the other.

[David Hess]

Hybrids and fans yes, but they are reasonably reliable. The capacitors are known to fail and probably will have to be replaced art some time or the other.

I would not consider the capacitors failure prone so much as wear out prone which is understandable given their age and that pretty much identifies the largest issue with this oscilloscope and its cousins; they are old.

I do not know of any specific issues with the 2235A or the 2235 series for that matter.  The common issues that come mind include the power supply capacitors wearing out and CRT DC restorer failures although no more often than other CRT oscilloscopes.

Since they share a lot of common parts and a lot of 2235 series oscilloscopes were sold, parts availability is not generally a problem.  I think this carries over to the 2235A although I have less detailed documentation on it.

[BravoV]

Hybrids and fans yes, but they are reasonably reliable. The capacitors are known to fail and probably will have to be replaced art some time or the other.

So its the fan, damn.

Regarding hybrids, I guess I will have to gamble on that, while for common electrolytic caps like David mentioned, I don't consider them as problem as once they're aged enough, that it self warrants for a replacement.

Thanks.

[David Hess]

Hybrids and fans yes, but they are reasonably reliable. The capacitors are known to fail and probably will have to be replaced art some time or the other.

So its the fan, damn.

Regarding hybrids, I guess I will have to gamble on that, while for common electrolytic caps like David mentioned, I don't consider them as problem as once they're aged enough, that it self warrants for a replacement.

Thanks.

Fans are just another one of those parts subject to wearing out like electrolytic capacitors.  I have noticed that preemptively oiling their bearings makes a huge difference in operating life.  The Tektronix custom fans seem to last longer than the more modern box fans that replaced them.

Many or most or all of the custom parts will be shared with the other 60 MHz and 100 MHz oscilloscopes in that series including the 2213A, 2215A, 2220, 2221, 2221A, 2224, 2230, 2232, 2235, and 2236.  They all use discrete CRT amplifiers although that does not necessarily make replacement semiconductors any easier to find but I have never seen failures there.

[BravoV]

Fans are just another one of those parts subject to wearing out like electrolytic capacitors.  I have noticed that preemptively oiling their bearings makes a huge difference in operating life.  The Tektronix custom fans seem to last longer than the more modern box fans that replaced them.

Many or most or all of the custom parts will be shared with the other 60 MHz and 100 MHz oscilloscopes in that series including the 2213A, 2215A, 2220, 2221, 2221A, 2224, 2230, 2232, 2235, and 2236.  They all use discrete CRT amplifiers although that does not necessarily make replacement semiconductors any easier to find but I have never seen failures there.

No, what I meant is the customised Tek fan that can not be replaced with ordinary standard sizes fan.

There is a thread mentioned in this forum, that the owner had to rewind the fan motor cause its not using industry standard fan. 

Edit : Found it -> https://www.eevblog.com/forum/testgear/tektronix-2445-repair/

[David Hess]

Fans are just another one of those parts subject to wearing out like electrolytic capacitors.  I have noticed that preemptively oiling their bearings makes a huge difference in operating life.  The Tektronix custom fans seem to last longer than the more modern box fans that replaced them.

Many or most or all of the custom parts will be shared with the other 60 MHz and 100 MHz oscilloscopes in that series including the 2213A, 2215A, 2220, 2221, 2221A, 2224, 2230, 2232, 2235, and 2236.  They all use discrete CRT amplifiers although that does not necessarily make replacement semiconductors any easier to find but I have never seen failures there.

No, what I meant is the customised Tek fan that can not be replaced with ordinary standard sizes fan.

There is a thread mentioned in this forum, that the owner had to rewind the fan motor cause its not using industry standard fan. 

Edit : Found it -> https://www.eevblog.com/forum/testgear/tektronix-2445-repair/

Oh, ya.  Tektronix made their own custom fans which are better than most box fans but like any mechanical device they do wear out.  When I do major maintenance on my Tektronix oscilloscopes, I oil and adjust the fan bearings.

They *can* be replaced with box fans on the 22xx series if necessary with a little bit of mechanical work.

[SeanB]

Ther Tek fans can be dismantled, and if you strip and oil it ( I did some other equipment fans that were not exactly designed to be repaired, but did it anyway as they were not easy to get even from the manufacturer) with a good quality synthetic oil ( Mobil 1 works well as a lube, a single bottle will last a lifetime here, use the 5W version) it will give a much longer service life.

[BravoV]

Yeah, but its better to have a standardized fan like other old Tek scopes, so finding the replacement is no brainer.

Also the offered price is about $75, what do you think ? Remember, I will be allowed to do brief test and external inspection.

[Wh1sper]

Yeah, but its better to have a standardized fan like other old Tek scopes, so finding the replacement is no brainer.

Also the offered price is about $75, what do you think ? Remember, I will be allowed to do brief test and external inspection.

Take it and RUN!
I payed about 150$ for my 2232  and I#m satisfied

[edavid]

I see a lot of them with broken/missing knobs, and broken timebase switch assemblies.  The knobs are surprisingly expensive.

$75 is meh in the US, I forgot what your secret location is.

To Wh1sper, you can't compare this to a 2232, that's a way better (and heavier) scope.

[David Hess]

I see a lot of them with broken/missing knobs, and broken timebase switch assemblies.  The knobs are surprisingly expensive.

$75 is meh in the US, I forgot what your secret location is.

To Wh1sper, you can't compare this to a 2232, that's a way better (and heavier) scope.

I agree with Wh1sper; at $75 it is a steal even in the US although I bought my 2230s for not much more than that.  The only better model unless you want storage is the 2236 because of the timer/counter which is really handy and does double duty reading out the sweep delay.

The list of oscilloscopes I gave earlier including the 2232 are all basically the same design.  The 60 MHz versions lack temperature compensation for the frequency and phase response limiting them to lower bandwidth.  The single timebase versions lack the second timebase.  The storage versions have a storage board and readout support.  The 2236 has a timer/counter board.

The last gasp of this series was the 2221A, 2235A, 2236A, and late model 2232 which together replaced all of the other versions in 1990 or 1991.  I think the 2235A and 2236A were only produced for 2 years while the 2221A and late 2232 lasted until 1994.

[spec]

...
The last gasp of this series was the 2221A, 2235A, 2236A, and late model 2232 which together replaced all of the other versions in 1990 or 1991.  I think the 2235A and 2236A were only produced for 2 years while the 2221A and late 2232 lasted until 1994.

Hi David.

My Gould OS series scope became terminally unwell in a house gas explosion so I've been looking for a replacement. I didn't for a moment imagine that a Tek scope- like we had at work- would be within my grasp because of the initial cost and the fear that if anything went wrong it would be a total write-off due to the complexity of the circuitry and custom components. But your informative posts and the others on this forum have made me reconsider- thanks for that!

I'm now fancying a 2235 or 2235A but can't find what the difference is between the two. I do know that the 2235A has a revised front panel but it appears to be only cosmetic. Can you or anyone else on the forum advise?

I don't really want storage, digital, counters etc, on KISS grounds, although they would be nice, but a high bandwidth would be the cherry on the cake, say 400MHz. Are the high frequency scopes more complex/ difficult to maintain. Would you also be able to give some guidance on a suitable model?

Comments would be appreciated.

[tekfan]

The difference between the A and non A versions is an improved power supply in the A version.
There's a factory mod for the non A version which involves replacing a transistor with one with better specs and adding a few diodes and resistors.

I have the non A Tek 2236 which is exactly the same as the 2235 but with an added digital counter/timer/multimeter board.
The power supply is exactly the same as the 2235 and I haven't had any problems with it, despite the power supply loaded even more than the 2235. The fan is not exactly quiet but the bearings are still quiet. Most of the noise is actually due to airflow.

The 2236 and lower series are all off the shelf components. The only custom component are 2 eproms for the CTS board for the 2236.
I've heard that the FET output transistors for the vertical deflection amp are replaceable with newer FET equivalents but they have to be matched to compensate for temperature drift. Apart from that everything else are standard components. Nothing custom.

It's nice to open the scope about once per year to blow out the dust (especially around the PS) and oil the fan. That should prevent overheating.

The 2240, 2245 and more advanced models use quite a few custom Tek hybrids which you can only get from a suitable donor scope (if those aren't broken as well).

All the higher bandwidth scopes have at least a few custom Tek amplifier ICs inside. Even the old 350MHz bandwidth Tek 485 shares the same Tek HF amplifier IC's as the Tek 7000 series plugins.

[spec]

Thanks Tek for your fast and succinct reply. I have read your other posts around this topic area and have found them very helpful and well informed too.  As a result of your comments I'm now leaning towards a 2236. Presumably, if something goes amiss with the CTS board, in principle anyway, you just loose the fancy functions but the 2236 would still operate effectively as a 2235.

[tekfan]

In theory it should be possible yes. The delayed timebase won't be calibrated though because instead of using a counting dial on the delay time potentiometer it uses the counter display to show you the exact delay time. But not a very big deal. The delayed time base is just an added bonus and not something completely necessary.

The 2236 is a great scope for field work. No need to carry a separate multimeter around. It has a built in 3 3/4 digit meter that's fully isolated from the scope chassis and even does True RMS measurements up to about 200kHz. Diode and continuity check (although a bit slow, but latched). The voltmeter can be coupled directly to the CH1 input and can measure DC and AC (true RMS). The gated timer is useful  if you want to measure the exact frequency of a signal riding on top of a signal with a different frequency. All in all a very powerful scope.
I'm not saying the 2235 or the lower models aren't any good. They're very nice scopes, but the 2236 is probably the most useful and feature packed of all of them.
The only thing I sometimes find lacking in the 2200 series scopes is a trigger high/low rejection filter. But the auto trigger mode works brilliantly most of the time.

If for any reason you need the EPROM images for the CTM board I'll be happy to provide them.

[David Hess]

The only thing I sometimes find lacking in the 2200 series scopes is a trigger high/low rejection filter. But the auto trigger mode works brilliantly most of the time.

Some models include this.  The 2221A, 2224 (?), 2232, 2235 (option 1), 2235A, and 2236A include LF and HF reject on the A trigger.  The 2220, 2221, and 2230 have HF reject on the A trigger.

Something I never noticed before is that the 2235, 2235A, 2236, and 2236A also support trigger view.

[BillWojo]

Do any parts interchange between the Tek 2445A and the 2236 scopes? I picked up a 2445A for 25.00 because it fails on start up with Test 4 Fail 10, looks like it means EAROM Interface fault, check U2118, U2208, U2108 and Q2025. I can clear the message and get nice traces on all 4 channels using the calibration test point.
Today I was at the local scrap yard and saw a Tek 2236 laying in the pile. Front was covered with mud and the back looks like it took a hit, round metal cover missing and big dent on the top of the back. I could probably get it for a buck or two a pound.
Worth picking up?
Thanks
BillWojo

[markce]

if something goes amiss with the CTS board, in principle anyway, you just loose the fancy functions but the 2236 would still operate effectively as a 2235.

You could also lose part of the delay function of the time base, it's not a complete 2235 without the CTS board.

The Tek fan is not really a problem. I replaced one with its driver box in a 2236 by a standard Papst part, which was also lower power and less noisy (audible and electrical).

The 22xxA has a bunch of improvements, input amplifiers are lower drift, trigger amplifier changed from discrete to  custom IC, lots of part changes. Trigger functionallity is changed as well.

[David Hess]

I'm now fancying a 2235 or 2235A but can't find what the difference is between the two. I do know that the 2235A has a revised front panel but it appears to be only cosmetic. Can you or anyone else on the forum advise?

The difference between the A and non A versions is an improved power supply in the A version.
There's a factory mod for the non A version which involves replacing a transistor with one with better specs and adding a few diodes and resistors.

The 22xxA has a bunch of improvements, input amplifiers are lower drift, trigger amplifier changed from discrete to  custom IC, lots of part changes. Trigger functionality is changed as well.

The differences between the 2235/2236 and 2235A/2236A are minor.  I know we tracked down one recently that involved the A versions of these oscilloscopes using a thin film network for the focus divider chain instead of a bunch of discrete resistors which are a known reliability problem but easy to repair.  The largest difference in reliability may be simply do to age; the A versions were available in 1990 and 1991 while the non-A versions were available from 1984 to 1989 so could be up to 7 years older.

The differences between the 2213/2215 and 2213A/2215A are significant in all respects except for their look.  They are very much different designs and I would avoid the 2213 and 2215.

The four channel 22xx oscilloscopes had A and non-A versions as well but I think they are all very similar.

I don't really want storage, digital, counters etc, on KISS grounds, although they would be nice, but a high bandwidth would be the cherry on the cake, say 400MHz. Are the high frequency scopes more complex/ difficult to maintain. Would you also be able to give some guidance on a suitable model?

Keep in mind that high bandwidth trades off easy probing.  High impedance probe bandwidth assumes a terminated 25 ohm source impedance which is unrealistic in most applications where the probe's input capacitance will limit bandwidth.  Low impedance probes solve this at the cost of additional restrictions but you still need to minimize ground length or even use coaxial connections at the probe tip.

High bandwidth oscilloscopes are often more difficult to maintain simply because their are fewer of them in the wild.  They are also more difficult to calibrate if necessary.

I have a couple of high bandwidth, 400 to 500 MHz, 7000 series analog oscilloscope mainframes which only support high impedance probes up to 200 MHz but even when restricted to lower bandwidths, they have the advantage of producing brighter and clearer displays then my slower analog oscilloscopes.  I suspect the 24xx series of analog oscilloscopes have a similar advantage over the 22xx analog oscilloscopes when used at lower bandwidths.