Did I score, or not? 2465a

[Phaedrus]

I made an offer on a Tektronix 2465a. It was listed at $295, I offered $200 and to my surprise was accepted. Unit is claimed all channels tested for display, comes with manual and two probes. With shipping & tax it came to $237.

I didn't think much of it, until I checked fleabay afterward and saw that most other 2465as run $350+. $500 for a calibrated one (which this is not). Next cheapest one on offer is $229 and that's with a bad Ch.1.

It should arrive some time this week. Assuming it's in good working shape, I think that qualifies as a pretty decent score. It will definitely be a welcome addition to the lab, assisting the TDS-3014C I'm using now.

PS, what are some good tests to run to get a feel for function and calibration?

[linux-works]

200 is a great deal for that!  congrats.

show us inside views of the boards once you get it.  read up on the mods for caps and stuff that need replacing.

if you don't have a funct gen yet, that's next on your list.

[Phaedrus]

It definitely is. I'm still reading up on models though. What are some good ones to look out for under $200? I'd like at least 20MHz.

This is what I've got in the lab at work. We have a focus on power electronics: PSUs, fan controllers, etc. and some minor digital stuff.

Tektronix TDS-3014C
Tektronix 2465a (new, mine)
Fluke 87-V
Fluke 115
BK Precision 8514 1200W DC Load
2x BK Precision 8540 150W DC Load
Extech 380820 1400VA Power Analyzer
Brand 20-1850 1800VA Power Analyzer
Weller WES-51 Solder Station

[linux-works]

I am a fan of hp funct gens.  so many to choose from!  but the older ones often top-out at 10 or 15mhz, max.

my favorite is this one: hp 33120a.

its not cheap, but I like the fact that I can change freq's down to the 1hz level and it will generally stay there, too (pretty stable).  direct digital display and the ability to take a 10mhz lab clock input and sync to that.   one thing I don't like is that the triangle stops at 100khz but the sine/square go to 15mhz.  strange.

still, a very likeable unit with reasonable wave fidelity.

[Phaedrus]

The unit arrived today, seems to be in pretty good nick. Couple oddities: the manual it came with wasn't for the scope, but for a GPIB plug-in. However, no GPIB plug-in is installed. Weird. I have the manual in PDF form though, so no big deal.

The more annoying thing is that Vdiv doesn't seem to go below 20mV. The dial is labeled down to 2mV, but I can't seem to figure out how to get it lower than 20mV. Maybe it's in the manual.

Overall pretty satisfied though!

[casinada]

Probably you have a Tektronix 10X probe connected to the input of the scope. Without any probe connected you should be able to go down to 2mv per division and if you cascade channel one to channel 2 through the back of the scope you should be able to measure even smaller signals (/10)
If it shows 20mv/div without probes, something is not right.

[Phaedrus]

That was it. Disconnected the probe and it went down to 2mV; and when I connected a 1x probe it stayed there. Thanks!

[casinada]

There is a pin on the connector of the probe that touches the outside ring of the BNC connector that enables the 10X multiplier

[EEVblog]

Yes, excellent price!
Assuming it's fully working, a bargain.

[breadbox]

A couple of months ago I got myself through a long and tedious work project by placing optimistically low bids on various members of the 2465 family. I kept losing but it was a good distraction. One day I won an auction for a 2465a, $200 shipped. Giddy with joy I looked up the item description and found that somehow I'd overlooked a key sentence: 'The display doesn't seem to be working quite right.'  Well, good, I thought, a little repair project to look forward to after the work slog. Then I did some googling to see what could go wrong with a 2465a display and learned about the failure-prone U800 horizontal amplifier. Just last week I saw a salvaged one listed on ebay for $195 plus shipping, offered by a fellow who takes the trouble to explain how ebay and paypal fees work and how he's really making only $169 on the deal, and don't forget, 'Part of that $169 pays for the surgical skill that does not lose any of the IC's 24 legs.' He adds: 'If you figure out any other way I get to keep $169 that costs you less, I'll accept.' I wonder how many under-the-table sales are facilitated by ebay.

When you are waiting for a package that will contain non-returnable, non-refundable proof of your own folly, you start to hope that it will be lost in transit. Mine wasn't. When it arrived I cleared some space on the desk and held a ceremonial powering-on--for closure, as they say. Readout, cursors, and a trace appeared, and after about a minute they started to move. But their motion was not the slow sideways drift that signals U800 failure. It began as a tremor, grew into a flicker, then escalated into a rapid oscillation. On the vertical axis. Things started to look up, so to speak.

Meanwhile, the transformer on the high-voltage board that supplies the CRT was rattling like a joy buzzer, keeping time with the jittery display. Here's where my luck really started to change. I know one rule for troubleshooting electronics, namely: Um, maybe it's a bad electrolytic capacitor? And happily this turned out to be the only rule I needed. After studying the schematics I headed straight for the filter cap on the unregulated -15V line leading to the high-voltage board (a journey that took much longer than expected) and tested it for ESR using another scope and a function generator. Bingo. Replaced it a few days later, putting an end to the song and dance.

[Phaedrus]

Well I do seem to be having a couple issues in fact. Fortunately I don't seem to be having the side-to-side motion of the U800 failure. But Channel 1 seems to be extremely fuzzy. I don't have a function gen, so I measured the output ripple of a PSU. Channel 2 shows the same ripple waveform as the digital scope, with a nice intensity graded fuzz around it for the high frequency noise. However, Channel 1 with the same probe measuring the same thing shows just fuzz. The amplitude is about right, but there's no visible waveform. Oddly, now it's started so that Channel 2 also shows only fuzz, unless Channel 1 is also turned on. There also seems to be some vertical ghosting of waveform peaks.

Nothing's ever that easy, is it?

I'll post a video of what's happening with the screen later.


EDIT: I did look at U800s on Ebay, and there are a ton of them with missing pins from China. Why is that even a thing? It's not that hard to desolder a 24-pin DIP, what, are they tearing these things out with pliers?