First oscilloscope

[wizard69]

Nigel has some good points here!!!    You don't need a lot of capability for audio work.   Further there are tons of used equipment out there that is more than good enough.   The last thing you want to do, be it a business or hobby, is to burn a lot of cash on instruments and lab setup.   The trick is knowing what you need and where to get it.

I have got by with a 20MHz analogue scope for about 10 years now in my own business.
I designed stuff upto about 20MHz so its fine for that.
I do a lot of audio and that is we  lwithin its limits.
Paid £40 off ebay for it as its second channel isnt working properly.

You really need to think about what you are going to use a scope for.
Thats will give yo ua bandwidth and scope type to go for.

My sig gen is about 10 years old to oand paid £20 for that on ebay.

Fancy soldering station ? No, £10 soldering iron from Maplin that goes on forever and a tip lasts about a year.

[No.Mad]

Hi Chris,

Two models you pointed out are quite good for any starting hobbyist, but I wouldn't recommend for repairing stuff as bandwidth sometimes could be insufficient.

There are few problems with older scopes: obsolete or proprietary parts, mostly uses CRT displays which use high voltage and when it's gone, it's gone.
Advantage of older scope is surely service manual available in pdf that you could download from net and high reparability - even after considering obsolete parts.

Try to look around on forum searching for models you are interested in, there should be at least one post with review

On side note: I have unused GW Instek GDS-1052U oscilloscope, 50MHz, 250MSps. It sits in box, collecting dust as I upgraded to older Agilent MSO (I needed higher bandwidth and digital probe in scope). It has everything original, in perfect working condition and in small size. I could sell it to you for £200+P. Hit me PM if you are interested.

Regards,
No.Mad

[rstofer]

Nigel has some good points here!!!    You don't need a lot of capability for audio work.

As long as audio is the end game, bandwidth isn't really important as long as it is somewhat more than the audio spectrum - up to about 20 kHz.  But there may be some high frequency oscillations that need to be seen to be dealt with.  10 MHz will show these oscillations in most cases.

But just as soon as you hit digital, bandwidth requirements increase; quickly.  I have a DS1054Z unlocked for 100 MHz operation.  The other day I was looking at a clock signal generated by an FPGA project.  I wanted to be certain it was the proper frequency so I planned to use the scope's internal frequency counter.  This worked fine and measured about 60 MHz, as I expected.

But the waveform looked like a distorted sine wave.  Well, duh!  The square wave would be composed of the odd harmonics of 60 MHz so the 3rd harmonic would be around 180 MHz, the 5th around 300 MHz and the 7th around 420 MHz and, in my view, a square wave doesn't look good without at least the 9th harmonic, possibly more.  My poor little 100 MHz scope was totally inadequate.  Even my 350 MHz Tek 485 would only display up to the 5th harmonic.  I would need a 500 MHz scope to get a reasonable representation.

Had I bought the 200 MHz Siglent I have been lusting over, it might have been able to display the fundamental plus the 3rd harmonic - still inadequate if I were trying to look at signal integrity.

Fortunately, all I cared about was the frequency of the waveform, not the waveform display.

In my view, these low bandwidth analog scopes are useless for modern digital projects.  That's why I got rid of my 10 MHz Heathkit scope and moved to the 350 MHz Tek.

An FPGA output at 60+ MHz is unusual for me.  I am normally working at 10 MHz and below for things like SPI.  As a result, my 100 MHz scope will display the 9th harmonic and that 200 MHz Siglent would be even better.

So, if the goal is to stay forever and always in the audio spectrum, almost any scope is good enough.  But just as soon as you start playing in the digital sandbox, bandwidth requirements explode.

https://www.mathworks.com/help/matlab/math/square-wave-from-sine-waves.html

The real spec to worry about is the scope's risetime.  But this is tied to the bandwidth on the nameplate by a constant of 0.35 so, if you have have the bandwidth, you will have the risetime and vice versa.  Any discussion of 'equivalent time sampling' is omitted...  It's just voodoo and requires a repetitive signal.

RT = 0.35 / BW  where RT (risetime) is in nanoseconds and BW (bandwidth) is in gigahertz.  For 100 MHz you get RT = 0.35 / 0.1 = 3.5 ns.  Check your scope manual and see what you find.

[akimpowerscr]

you forget to mention something very important! probing frequencies of 200 and more Mhz is not as simple as you seem to present it.

In the past, Tektronix made all is input amplifiers with bandwith higher than 200Mhz only with 50R input resistance.

There was a good reason for this: to probe high frequencies is quite difficult : you need an expensive active probe to not disturb the circuit you are probing.

Passive probe have a too large input capacitance, which mean a very low impedance at high frequencies.

So, for a beginner, and even for an hobbiest, bandwith of 350Mhz does not matter....

It's like to buy a car with maximum speed of 250 Km/h when you live in a country where speed is limited to 120Km/h !!!!

[george.b]

you forget to mention something very important! probing frequencies of 200 and more Mhz is not as simple as you seem to present it.

In the past, Tektronix made all is input amplifiers with bandwith higher than 200Mhz only with 50R input resistance.

There was a good reason for this: to probe high frequencies is quite difficult : you need an expensive active probe to not disturb the circuit you are probing.

Passive probe have a too large input capacitance, which mean a very low impedance at high frequencies.

So, for a beginner, and even for an hobbiest, bandwith of 350Mhz does not matter....

It's like to buy a car with maximum speed of 250 Km/h when you live in a country where speed is limited to 120Km/h !!!!

Umm... no. As hinted by rstofer, if you're measuring a square wave at, say, 50MHz, which is well within the domain of hobbyists, then to get a remotely passable representation of it, you'll need at least 250MHz bandwidth. Passive 10Mohm, 500MHz bandwidth probes do exist, so yeah. Dunno what you're going on about.

Also:

I would like to add some information that might be useful.

For the repair of analog synthesizers, it is sometimes necessary to adjust the frequency of oscillators.

We could do this by using a function generator with digital reading and using the XY fonction of the oscilloscope.

With the Lissajous pattern, you can get a precise adjustment but it is quite touchy for frequencies of 1Khz and higher.

Another method would be to use an inexpensive vintage frequency counter or a cheap kit from ebay.

I use an old HP5315A ... The frequency counter does not need to have high accuracy or a high frequency bandwith.

One advantage that the HM204 has over the HM203 is that it has a Y output on the rear panel.

This makes it possible to connect the frequency counter to it and to be able to constantly read the frequency of the signal displayed on the screen.

Or just get a DSO, which will show you the frequency of the signal displayed on the screen without need for other equipment?

[akimpowerscr]

Let's see for example :
N2873A Passive Probe, 10:1, 500 MHz, 1.3 m

Key Features & Specifications
500 MHz bandwidth
10:1 attenuation ratio
9.5 pF input capacitance
10 MΩ input resistance (when terminated into 1 MΩ scope input)
400 V CAT I and 300 V CAT II maximum input voltage
Scope compensation range: 10 – 25 pF

At 350Mhz, 9.5 pf has an impedance of 47.9R.....When probing, you are disturbing the circuit with a capacitive impedance of 47.9R at 350 Mhz and this may change the thrue waveform.

This is the reason why it is recomanded to use active probe to probe such high frequencies.

Umm... no. As hinted by rstofer, if you're measuring a square wave at, say, 50MHz, which is well within the domain of hobbyists, then to get a remotely passable representation of it, you'll need at least 250MHz bandwidth. Passive 10Mohm, 500MHz bandwidth probes do exist, so yeah. Dunno what you're going on about.

Really ?

As hobbyist you often need to measure the risetime of a 50Mhz square wave ?
For sure, you have to spend your time in a very specialized field of electronics .....

[rstofer]

As hobbyist you often need to measure the risetime of a 50Mhz square wave ?
For sure, you have to spend your time in a very specialized field of electronics .....

These days there is nothing specialized about a Digilent Basys3 FPGA board with an onboard oscillator running at 100 MHz.  The 60 MHz was created by an internal clock circuit and reduces the 100 MHz input to 60 MHz for internal use.  Believe me, there is nothing sophisticated about this.  If it were, I wouldn't be working on it.

But, as I said, I wasn't interested in the waveform for this particular application, I wanted to verify the frequency.  Seeing the waveform just reinforced how limited my scope actually is.  OTOH, I can't afford a 500 MHz DSO so what I see is what I get.  It's just a hobby and only 1 of several.

There's quite a difference between a 10 MHz analog scope and a 100 MHz DSO.  Just in terms of 9th harmonic values, the 10 MHz scope is limited to around 1 MHz while the 100 MHz DSO is limited to around 10 MHz.  There are signals all over the uC world that are in excess of 1 MHz but not so many that are higher than 10 MHz.  Sure, they exist and with these 600 MHz uCs they are becoming fairly common but, so far, they don't outrun my 350 MHz Tek 485.  But they will...

Again, if the end game is just audio, ignore all these comments.  Almost any scope is good enough.  The problem I see is that the hobby tends to morph and the next thing you know, you're playing with an Arduino and lusting for the speed of a Teensy 4.1.  Or you decide to play with FPGAs...

[george.b]

Whether using a passive probe will introduce problems with capacitive loading or not depends on what you're measuring. It doesn't sound fair to me to dismiss a whole class of measuring apparatus because in some cases they may not suffice.

Really ?

As hobbyist you often need to measure the risetime of a 50Mhz square wave ?
For sure, you have to spend your time in a very specialized field of electronics .....

Yes, really. The year is 2020, and FPGAs, multi-hundred-MHz microcontrollers and other fairly high-speed digital electronics are not exactly far from the grasp of hobbyists anymore.

[james_s]

I have a 1GHz scope and the bandwidth is nice to have, but honestly I could get by with a 20MHz analog scope for 90% of what I use a scope for. Bandwidth is like money, you can never have too much, but how much you really "need" depends on your goals. IMO the only reason to consider an analog scope anymore is if you find one really cheap, and they do turn up. You might even find one for free, I've given and received a few of them over the years, some that worked just fine. Ultimately any scope that even mostly works is better than no scope at all, which is what I had for the first decade or so of my electronics hobby.

[jonslab]

Go with a DIY one

[Chrishas]

First of all thanks everyone for all the valuable information and taking the time to reply. I've been looking at the suggestions provided and Charlotte's thread has been very educational and I'm still going over all the info.
I also understand that my concern for probes was invalid.
I'm keeping an eye out for a good deal and so far I see Hamegs come up more often than other brands although most are HM203s and usually around £100.
There's a few Tektronix but the only one which I saw was tested was a TDS320 on an auction but it doesn't seem to have great reviews.
I've also seen a Goldstar 9020A for £25 which I may be able to checkout in person. What's the opinion on those and is it worth it in terms of reliability, reputation? This is much cheaper to what I've seen so far and with the location it's very tempting. If I did check it out what are some things to test on the spot?

[Mr. Scram]

Go with a DIY one

I did that and regretted it. I saw various things that started making sense after I bought a basic but "real" oscilloscope.

[george.b]

I'm keeping an eye out for a good deal and so far I see Hamegs come up more often than other brands although most are HM203s and usually around £100.
There's a few Tektronix but the only one which I saw was tested was a TDS320 on an auction but it doesn't seem to have great reviews.
I've also seen a Goldstar 9020A for £25 which I may be able to checkout in person. What's the opinion on those and is it worth it in terms of reliability, reputation? This is much cheaper to what I've seen so far and with the location it's very tempting. If I did check it out what are some things to test on the spot?

The TDS320 isn't a bad scope, if you can get it for cheap. It's a lot better than my first DSO (a Tek 2430A). It is old though, like all the options you present. It's got just 1k points acquisition memory, but meh, that's a lot more than none at all, and 500MSa/s isn't too shabby. It has some modern conveniences like automatic waveform measurements - no FFT, though.
I have a Leader-rebadged Goldstar, not a 9020A, but a similar 7020A. Its HV transformer failed; I found the whole thing kinda shoddily built overall. All phenolic single-sided board construction inside. £25 is alright, I think, if it's in good condition. I mean, it's still a real oscilloscope, albeit a basic one.
Be aware that any of them could work for many years or a couple of days before malfunctioning, and you may have to service it eventually.

Go with a DIY one

I did that and regretted it. I saw various things that started making sense after I bought a basic but "real" oscilloscope.

Those DIY scopes are little more than toys.

[james_s]

I've had a few TDS300 series scopes, they're decent instruments, just old. The thing to watch out for with them is the DS1644 which holds the calibration constants, I tried to back that up on two that I had and in both instances it got corrupted somehow and calibration costs more than the scope is worth generally. The TDS400 series is better in the regard that it stores the calibration in EEPROM but it's also quite an old instrument. Decent if you can get it cheap but a lot of times people ask inflated prices due to the Tek name.

[rsjsouza]

I've also seen a Goldstar 9020A for £25 which I may be able to checkout in person. What's the opinion on those and is it worth it in terms of reliability, reputation? This is much cheaper to what I've seen so far and with the location it's very tempting. If I did check it out what are some things to test on the spot?

The Goldstar you mentioned is one of the opportunities for a cheap and very basic oscilloscope. I would definitely schedule a visit and check it out.

Several common sense aspects to look for:
- see if the screen is in good shape and not slanted, has severe distiortions around a certain point. These could be signs of mistreatment that are affecring the tube.
- not a problem per se, but check if the waveforms are aligned with the horizontal axis. If not, then locate the rotation adjustment and see if you can put them horizontal. If you can't, this may be a sign of mistreatment on the tube or simply a failing circuit. This is usually good to see after the oscilloscope is on for 10+ minutes.
- If the scope has probes, plug each channel to the 1kHz square wave compensation output and check the output for frequency or amplitude deviations (you need to count divisions on the graticule). Rotate the various controls looking for mechanical jamming or lockups or electrical noise when rotating them. Pay special attention to the V/div and ms/div and see how the waveform (or just the GND signal) behaves.
- Do you have a multimeter with a built in signal generator or maybe a simple capacitance checker? If so, that is an interesting way to show a waveform that can go higher than 1kHz. Otherwise, assemble a small high frequency NAND oscillator (or use a crystal oscillator)  ircuit and see how the oscilloscope shows it. For a square wave, don't go over 3 or 5MHz.
- take a small battery where you know the voltage (using a DMM) and measure it with the oscilloscope. You can know its voltsge and if the AC mode is working (in this setting it should show zero volts)

Others may have additional tips. Good luck!