In my recent technology report "The Ultimate Test Drive: High-Octane Oscilloscopes" , I got a chance to write about digital sampling oscilloscopes (DSOs). These instruments are truly amazing today because they not only make all the basic measurements, but also perform all sorts of other useful test, troubleshooting, and measuring functions, thanks mainly to the amazing analysis software available.
After I wrote the report, however, it dawned on me that I forgot to mention analog oscilloscopes. It is probably just as well, as I suspect that most of you only use digital scopes and that analog scopes are history. But digging around in the analog scope world recently convinced me that these scopes, while they may be in the minority, are still alive and well. There are more than a few of them around in labs and shops everywhere. Here is what I found out:
There is something to be said about looking at a signal directly on an analog scope. What you see is what is really there. In a digital scope, the signal is sampled, stored, processed, and then reconstructed on an LCD screen for viewing. What you see is a highly modified version of the actual input. So just how close is the actual signal to what you are seeing?
I realize that if the input bandwidth, sampling rate, ADC resolution, and processor speeds are great enough, then what you see is pretty close to the actual. If it were not, we would not be using these instruments. But such a process seems to be a great deal of signal processing just to look at a signal. It is expensive to do it this way, and for many routine applications, a digital scope is probably overkill. DSOs are also very expensive. If you are just doing basic troubleshooting, you probably do not need as much scope horsepower in the modern DSO.
Where analog scopes really still shine is in looking at mostly analog signals. If you are doing audio or video work, an analog scope will probably still tell you all you need to know. Modulated signals also show up better on an analog scope. Low speed digital signals are also OK on an analog scope. But if you are working in the realm of all digital signals, especially at very high data rates, digital is the only way to go. You probably need a logic analyzer. If you are using complex modulation and communications protocols, DSOs are a must.
In asking around about analog scopes, the following opinions emerged:
- Analog scopes are easier to use: I tend to agree with this opinion, but then again, maybe because I am just used to using an analog scope. DSOs, especially the older ones, are a bear to use. Like most digital devices, you end up spending lots of time playing with buttons and menus. I hate that. But I do have to say that the newer DSOs are a bit easier to use than the old ones. But they are still a challenge just because they do so much more. Remember, with a DSO you are dealing more with software than hardware.
- Analog signals display better on a crt: I also agree with this. They are clear and crisp, thanks to fine focus on the crt. You can really see the details better.
- Digital scopes can give false readings: Aliasing can sometimes be a problem. You really need to know what you are looking for and looking at on a DSO. With practice, you get used to the DSOs.
- The dynamic range of an analog scope is greater: This is probably true. The digital resolution of the typical DSO is 8 bits, and that does limit the range. For some analog analyses, the analog scope is better.
- I am used to analog and don't really need to change This is an old way thinking, but it is also so very common. I would have to say, don't knock it (a DSO) until you have tried it—especially the newer models. Yet, so many applications and design projects just do not justify the magnificence of a DSO. Old analog scopes do just fine, thank you.
- Analog scopes are cheaper: Boy, is that ever true. You can get a 100 MHz dual trace scope for less than $1000. There are no DSOs that cheap. A good DSO will set you back mucho thousands of dollars. Most wide bandwidth, fast sampling DSOs are over $10,000. And there are really high-end models over $100,000. But in defense of the DSO, these things do so much more. You really do get what you pay for. But if you really don't need it, why do it?
As for me, I still do a bit of design and use an analog scope. It is an old Tektronix 100 MHz unit I got used. It works great. I did have to buy a new set of probes from Tektronix, and those cost me more than the scope. But you can't not have the right probes and expect the scope to work right. Anyway, I am happy with it and it is fine for the work I do. I also experienced the fussiness of the early digital scopes when I bought 16 of them from Tektronix for my employer back in the mid-1990s. I hated these things. But I must say that after many hours of use, you do get used to the arcane menus and other control idiosyncrasies. These units had a bandwidth of 100 MHz and had the FFT feature, which I learned to love for certain measurements. The new DSOs are much better and do so much more. But, you really need to be able to justify their sky-high prices.
Incidentally, there are quite a few companies making analog scopes. The big U.S. manufacturers, like Agilent, Lecroy and Tektronix, retired their analog scope lines a while back. But you can still get many of them on the used market. Tons of used instruments are available—even on eBay. New analog scope makers include B&K Precision, EZ Digital, Instek, Iwatsu, Kenwood, Leader, and Protek. I may have missed some, but they are out there if you are still interested.
So how do you feel about analog scopes? Send me a note at [email protected]. And do answer the simple question below.
Which of the following best describes how you feel about oscilloscopes?
- I prefer analog scopes, they are adequate for my needs, and I will never change.
- When digital scopes come down in price, I will buy one.
- Use one of the newer DSOs with deep memory, color LCD, and analysis software, and you will never go back.
- Any good lab should have both.