(Tektronix TDS 460A) 100Ms/s Oscilloscope still good enough for modern stuff?

[jbennet-t]

I just picked up a used Tektronix TDS 460A oscilloscope for relatively cheap (around 85$ after shipping). To summarize, it's a digital scope (circa 1996), a 400Mhz, 4 channel model, with a 100Ms/s sampling rate. My question is if a 100Ms/s sampling rate is still good enough for modern applications.

I know 400Mhz is plenty, but I've heard that high bandwidth won't do you a whole lot of good if sampling rate is poor. For modern beginner scopes, 1Gs/s seems to be the more of the norm, or at least somewhere in that ballpark.

Scope PDF for reference: https://assets.tequipment.net/assets/1/26/Documents/TDS400_UserManual.pdf

I'm still relatively new to electronics (2nd year computer/robotics engineering student), and my main uses would be on personal projects, such as working with microcontrollers, servos, PIDs, ... and typical signals for these applications (PWM, I2C, serial, etc.). This will be my first scope. I'm also planning to pick up a usb logic analyzer for digital analysis.

Apologies if this is a silly question, and for any formatting errors. Thank you for your time!

[snoopy]

For the higher sweep rates this scope relies on equivalent time sampling to accommodate the higher bandwidth. This means whatever you are measuring should be repetitive so the scope can build up the waveform information from repetitive sampling the same signal. For standard micro work typically your requirements would be much lower bandwidths and sample rate and the scope would be using real time sampling so this scope should be more than adequate for your needs although it does not have any serial decoders etc but still a scope like this with 4-channels is much much better than some of the ebay toy scope offerings so you should be happy with this although I have heard some problems with leaky capacitors and power supply problems.

cheers

[vk6zgo]

For the higher sweep rates this scope relies on equivalent time sampling to accommodate the higher bandwidth. This means whatever you are measuring should be repetitive so the scope can build up the waveform information from repetitive sampling the same signal. For standard micro work typically your requirements would be much lower bandwidths and sample rate and the scope would be using real time sampling so this scope should be more than adequate for your needs although it does not have any serial decoders etc but still a scope like this with 4-channels is much much better than some of the ebay toy scope offerings so you should be happy with this although I have heard some problems with leaky capacitors and power supply problems.

cheers

Most such older DSOs also have a problem with long sweep lengths, as the sample rate reduces, to such a point that displays of signals like analog video (which has components up to 5MHz) at field rate, or in some cases even line rate, are unusable, due to aliasing.

Trying to look for mains hum on pulse trains with fairly high PRFs is also problematical.

[T3sl4co1l]

I don't have a problem with my TDS460 (not A; no floppy, 350MHz, and needed a recap).  EQ acquisition is ponderously slow, but it's rare that I'm looking at a fast signal that is not repetitive in some way, so it gets there eventually.

Bigger gripe is basic usability; Tek menus in the 80s and 90s were TERRIBLE.  Well organized, but SLOW.  If you have the choice between a contemporary HP/Agilent (546xx series and the like) of equivalent specs, I recommend it over Tek.  Their menus and displays are real time.

Most modern offerings are at least as good, with poorly organized menus being the most common failing.  Quirky acquisition is sometimes a thing, too (e.g., step response of entry-level Teks, bandwidth control, voltage range and PLL quality, etc.).  Acq and display is almost always better than of the olden days, with real time readouts, "DPO" effects and such being the norm.  Not to mention more pixels and modern interfaces.  And protocol analysis, and...

Tim

[james_s]

I don't really find the menus to be slow, I have a TDS340 that is my "portable" scope and while it's not fancy by modern standards, given it's ~25 years old I think it works quite well and is easy to use. $85 seems like a great deal for a TDS400 series in working condition, there's nothing new in the <$100 range that isn't a toy.

[DaJMasta]

If it's an analog scope with digital sampling features, it can probably do what's needed in most cases, but if it's entirely digital.... your realtime bandwidth is limited to 50MHz (Nyquist limit of 100MS/s), so you'll only be able to use any higher bandwidth modes in sampling mode, which requires a repetitive signal, so you lose out on most of the responsiveness and much of the utility a scope with the same bandwidth and enough sampling rate to actually digitize it in realtime.

So basically, if it can be used normally as an analog scope without messing with the digital system much, it will probably be ok as long as the UI and such is responsive (I don't have personal experience), but if it is limited to digital-only operation, I'd avoid it.  Only going to be useful at low bandwidths for many scope applications and will probably be more tedious to use at higher bandwidth operating modes.

[David Hess]

My question is if a 100Ms/s sampling rate is still good enough for modern applications.

100 MS/s is more than enough for a majority of applications.

I know 400Mhz is plenty, but I've heard that high bandwidth won't do you a whole lot of good if sampling rate is poor. For modern beginner scopes, 1Gs/s seems to be the more of the norm, or at least somewhere in that ballpark.

This DSO uses random equivalent time sampling resulting in a sample rate of up to 50 GS/s on repetitive waveforms.  With such a high sample rate, there is no nonsense with sin(x)/x reconstruction errors or aliasing.

Most such older DSOs also have a problem with long sweep lengths, as the sample rate reduces, to such a point that displays of signals like analog video (which has components up to 5MHz) at field rate, or in some cases even line rate, are unusable, due to aliasing.

These were not very good DSOs for video applications although better than some because of their longer standard and optional record length.

The loss in sample rate at slow sweep speeds is somewhat ameliorated by peak detection which will capture peaks down to 10 nanoseconds at any sample rate.

[snoopy]

For the higher sweep rates this scope relies on equivalent time sampling to accommodate the higher bandwidth. This means whatever you are measuring should be repetitive so the scope can build up the waveform information from repetitive sampling the same signal. For standard micro work typically your requirements would be much lower bandwidths and sample rate and the scope would be using real time sampling so this scope should be more than adequate for your needs although it does not have any serial decoders etc but still a scope like this with 4-channels is much much better than some of the ebay toy scope offerings so you should be happy with this although I have heard some problems with leaky capacitors and power supply problems.

cheers

Most such older DSOs also have a problem with long sweep lengths, as the sample rate reduces, to such a point that displays of signals like analog video (which has components up to 5MHz) at field rate, or in some cases even line rate, are unusable, due to aliasing.

Trying to look for mains hum on pulse trains with fairly high PRFs is also problematical.

That's why Tek came out with the Digital Phosphor scopes. However I recently had a situation where the DPO feature of the Rigol 1054Z scope was more of a hindrance than a saviour when it came to actual measurements and I could not find a way of disabling it and turning on waveform averaging. I needed to measure the phase response of a class-D amp with quite a bit of carrier signal and this is where my Tek TDS784A saved the day with it's waveform averaging ability. DPO is probably good for observing a waveform and measuring stuff like amplitudes but it's probably not so great for measuring detailed timing in the presence of a lot of unwanted noise.

cheers