When do you need a 4 channel scope???

[RissViss]

Hello everyone.

Have been reading posts and watching Dave on youtube. This statement in one way or another comes up " Get a scope with 4 channels " , but I have not seen or heard why?

Looking on the forum and lists of scopes to purchase under $600 and most are 2 channel with Rigol and GW having scopes with 4 channels.
So if 4 channels is the thing to get and the Rigol is the scope to beat, why is there no other company besides GW that I have seen making scopes with 4 channels.
Keysight's new scope has two channels and is the price range of the Rigol, why would they even bother to make it, if 4 channels is the thing to get.

This is what is going through my mind when do you need 4 channels as a hobbies how much of your scope using time will you need 4 channels?

Just trying to make sense of this, if some hobbies who have scopes and use 4 channels could give some examples would be great, and people who have 2 channeled scopes wishing they had 4 channels and give those examples.

I hope this make sense.
Have a day

[ataradov]

For example, when debugging SPI interface, or even typical I2C + IRQ line.

Or, just recently I was debugging screen interface, and I needed to see relation between HSync, VSync, and source clock at the same time.

A lot of that can be solved by a logic analyzer, of course, but sometimes you also need to ensure signal integrity.

[RissViss]

Thanks for the info.

How much of your scope usage is 3 or more channels. And how much of that would a signal analyzer and a 2 channel do?

Just trying to gather as much info as I can.

Have a day

[ataradov]

It is hard to tell. Most of my personal usage is 1 channel. This "mode" used with hand-held probe to check on individual test points. This ensures basic signal integrity.

Then things may diverge. Depending on a situation, you can either use logic analyzer, or go with the scope. Logic analyzer is mostly used for protocols, since they are mostly slow.  Scope is used when you are dealing with high speed signals (> 10 MHz), and logic analyzers are just not fast enough to capture accurate timing picture.

I would not go for 2 channels on a primary scope.

[georges80]

They make 2 channel scopes since they are cheaper and plenty good enough for schools/colleges/unis etc.

Often I use 1 channel just to check a signal, but beyond 1 it is typically 3 channels -> 1 on the trigger input (e.g. PWM input), 1 on the voltage output and 1 to a current probe. And I DO want to see the signal rise/fall time of each of those 3 signals and their time relationship.

Obviously for SPI etc, 3 or 4 channels is also often used.

All depends on your needs and typical use case. Easier to have 4 channels and mostly ever use 1 or 2 channels than vice versa

And this question has come up often enough before that search should likely find similar threads...

cheers,
george.

[hammy]

RS232 uses four signal wires. rx, tx, dtr, cts and gnd. Hook them all up and watch the screen to determine each of them.

[Smokey]

You can use any 4channel scope for only looking at 2 signals.  Try that in reverse.

I'm actually surprised they don't make more 6 or 8 channel scopes these days considering that it doesn't take an old school Tektronix mainframe level black-magic circuit to collect the data anymore.

[guenthert]

I never used all four at the same time, but three I did use.

Unless you spend good money on (or build yourself) a differential probe amplifier, you'll have to do the subtract-one-channel-from-another trick for signals which aren't ground based.  RS485 comes to mind, but also current measurements (using a shunt resistor, if you don't have a current probe).  If you want to measure the voltage at a node at the same time (for power measurements), you need then three (generally speaking, sometime you get away with two).

My first scope had four inputs, so I'm spoiled.  If I am going to replace it (as opposed to expand my pool of test gear), I'll be looking for another one with at least three inputs.

[RissViss]

Thanks everyone for the info.

Very happy with the responses.

Have a day

[David Hess]

The only times I have absolutely needed 4 channels was when I needed to combine channels to make 2 differential channels and for synchronous protocols like SPI with data in, data out, clock, and chip select and even then, I could have done what was needed with 2 channels and a trigger.  Some people need 3 or more channels which in practice means 4 channels to make 3 phase power measurements.

For a majority of uses, 2 channels plus a trigger is fine.  2 channel oscilloscopes also have a trigger channel and having 4 channels was an outgrowth of being able to display the trigger signal as a third trace; before 4 channels was common, 2 channels + 2 auxiliary channels was common.

[DaJMasta]

I've your monitoring a signal path, maybe at a couple points, and then also monitoring a secondary voltage derived from it (a peak detector output, for example), or power rails for noise.

If you've got a controlled environment and a repetitive signal, you can just use a two channel and move things around, but you don't always have that luxury, and having them all hooked up definitely saves time over having to swap them around to get the full picture.


If you're working with very simple circuits or when you're absolutely sure of your power supply, two is probably plenty... but if you're doing more elaborate stuff, debugging with power noise as a consideration, or chasing down an intermittent fault, having more channels is very valuable.  Not even talking about for digital purposes, it can be really nice to probe a signal path between several elements and see how each is modifying the signal in real time and time-correlated.

[alsetalokin4017]

Maybe sometimes you just need to impress the person who pays the bills!     

[james_s]

I think I used 3 channels once at a previous job to capture the power up sequence of several power rails, but I've never had more than a 2 channel scope at home. I have a logic analyzer with 136 channels so that has me covered if I need to capture anything digital. I also have one of those little "Bus Pirate" doohickies that is handy for sniffing SPI and I2C and such among other things.

[tautech]

In answer to the OP, rarely is there need for a 4 ch scope if you had a scope with the additional inputs a MSO unit can provide. Some call these Logic Analyser channels and in some way they are but not as good as a dedicated LA only unit. These in many ways are things of yesteryear but some find a need for them even today. Before MCU's,  Logic circuitry was common and the best way to see what was going on was with many channels and see/check the inter-relationship between the gates and timings etc.
Dedicated scopes on the other hand have a much wider range of applications particularly as their inputs can withstand much higher voltages than a LA. The best of both worlds is a MSO, it might be a 2 or 4 ch scope with 8 or more low voltage (LA) digital inputs.
Power electronics and particularly 3 phase control circuitry need higher voltage capable inputs  and in many cases 4 channels won't be enough to monitor all that's going on.

Some 2 and 4 ch scopes have an Ext Trig input which can double as a extra channel albeit at lower voltages than the main input channels.

why is there no other company besides GW that I have seen making scopes with 4 channels.

Because as yet you haven't looked hard enough. 
http://siglentamerica.com/pdxx.aspx?id=1488&T=2&tid=1

[Howardlong]

I guess I'm an outlier, but it's pretty common for me to use three or four channels. Four a while, before I had a decent 4ch DSO, I used two pretty good 2+16 MSOs with the trigger out from the first plumbed into the trigger in of the second. Before that I used a 2465B and a TDS2024B, both 4ch.

It's standard troubleshooting analysis, which is an iterative process. You're lucky if you find the answer with your first probing. Typically it'll take several attempts before you find the problem. Multiple channels means you can still examine several other probe points as you apply stimulus or otherwise recreate the problem.

With two channels you can usually achieve pretty much everything a four ch can, but it will sometimes be a tedious and manual process.

[Richard Head]

I've used all four channels for monitoring the gate drives of a full bridge converter with a pair of differential probes for the high side devices. It's very convenient  but if I had to choose between a high bandwidth two channel scope a low bandwidth 4 channel I'd take the two channel high bandwidth every time. 

[RissViss]

Yes the Siglent is a 4 channel, but it is not in the same price range as the Rigol or GW sub $550 range scope.
As far as I can tell there is only two scopes 4 channel below $550 USD.

I was just wonder, why no other company tries to compete with Rigol for the spot of best bang for the buck. This was just an observation and not a criticism of scope manufactures. It is good that Rigol has provided a good scope for the people that can not spend the money on higher priced scopes.

Have a day

[nctnico]

why is there no other company besides GW that I have seen making scopes with 4 channels.

Because as yet you haven't looked hard enough. 
http://siglentamerica.com/pdxx.aspx?id=1488&T=2&tid=1

But the SDS2000X series is 1) more expensive compared to GW Instek and 2) has way less features than GW Instek. Also according to the most recent posts on this forum the SDS2000X firmware still has a lot of bugs and other shortcomings.

[BravoV]

why no other company tries to compete with Rigol for the spot of best bang for the buck.

Its a big business decision , risky too, and Rigol took that big bet earlier compared to other competitors, and I guess they won looking at they're selling like hot cakes.

Other companies have to think seriously before following this step, as Rigol already has the advantages as early adopter. Rigol could easily slash the price to kill or hurt the competitors.

[sibeen]

I'm definitely an outlier. Four channels is not enough which is why I use a Yokogawa DL708 scopecorder which contains 8 channels of goodness. It is very slow, with even the high speed channels only operating at 100 kS/s with the modules I use with it, but that matters not as most of the signals I'm looking at operate at 50 Hz.

I work in the power industry and looking at 3 voltage waveforms, 3 current waveforms and a circuit breaker or two switching signals is something I do on a regular basis. Quite often I pine for even more channels and will eventually upgrade the now very old 708 for a newer 16 channel model. I just have to convince SWMBO that approximately $25k is still a good spend.

[james_s]

Wouldn't it be easier, cheaper and provide more choices to use a stack of 4 channel scopes rather than something as esoteric as a 16 channel scope?

[MrW0lf]

I'm actually surprised they don't make more 6 or 8 channel scopes these days considering that it doesn't take an old school Tektronix mainframe level black-magic circuit to collect the data anymore.

Hm... 8channel, 12bit, 20MHz, 256MB
https://www.picotech.com/oscilloscope/4824/8-channel-oscilloscope
Also they seem to launch 4ch differential soon (4x2=8 signal wires, floating), other specs similar, but already 14bit.
But of course these are not for general hobbyist use at around 2000€, you need to be into something specific and cool to need this sort of thing.

[MrW0lf]

Wouldn't it be easier, cheaper and provide more choices to use a stack of 4 channel scopes rather than something as esoteric as a 16 channel scope?

Well you can "stack" any scope if using 1ch for some kind of sync signal, some have special inputs for that. I had need for 4ch signal gen. Bought 2x2ch and synced with 10MHz in/out BNC & event trigger BNC. A bit cumbersome to use sometimes to get the job done but it gets done. It does not make sense for 99%+ of users so most manufacturers do not bother because to really sync it all in single screen / interface with all triggering stuff working as single scope would be pretty nasty engineering task.

[capt bullshot]

My first scope ever has been a 5MHz single channel analog CRO.
This was way much better than having no scope.

Next bigger thing to work with was an 100MHz analog two channel dual timebase boat anchor with plugins per channel and timebase
Quite impressive compared to what I had before

Next one was a real neat, lightweight and easy to use dual channel 100MHz / 1Gs DSO with b/w LCD screen
Quite impressive compared to what I had before, never used that boat anchor anymore, gave away my 5MHz analog scope somewhen

Next one was the same real neat, lightweight and easy to use100MHz / 1Gs DSO with b/w LCD screen, but 4 channel
Not so impressive to have 4 channels on a b/w LCD, otherwise handy since I was used to the UI

Seen and used some others DSOs in the meantime, color, up to 1GHz / 10Gs, typically 4 channels, up to 8 channels.
I use quite often one channel only, but there's a lot of opportunities when 4 channels come in real handy.

Now I have in my home lab:
the good 'ol "easy to use dual channel 100MHz / 1Gs DSO with b/w LCD screen"
the 200MHz 4 ch, colour LCD variant of the above
a boat anchor 1GHz, 4ch DSO
an old 100MHz portable boat anchor 2 ch analog

Most used: the good 'ol "easy to use dual channel 100MHz / 1Gs DSO with b/w LCD screen"
And I want back the "100MHz analog two channel dual timebase boat anchor with plugins per channel and timebase"
Most wanted: the new giveaway scope from Keysight - wouldn't buy it, but for free it would be an great toy, would show it to my boss and colleagues here at work to persuade them to buy some new scopes here (of the higher series)

[rf-loop]

But the SDS2000X series is 1) more expensive compared to GW Instek and 2) has way less features than GW Instek.

Where is 2 x 140Mpts memory in 4 channel GW
GoodWill can not at all do 500000 segment/s
GoodWill do not at all have all time background real time running waveform history buffer up to 80000 last wfm (same speed as current wfm/s speed).
Also unlike GW, SDS2x04X can do 2GSa/s for two channels simultaneously?
Afaik GW do not have real time output for full speed mask test.
Also 2000X have  16 channel  MSO hardware 16ch, 140Mpts/ch  (activation need license)
Made like tank compared to GW. 
Siglent do not want follow this "as cheap as possible" trend like Rigol and Keyshit.
Road to A brand is still very long but distange is infinite if not even slowly walk and try.  This can not do if go "price on front". 
There is lot of what need do better and better in hardware side (and in this area they have developed big steps and if they can continue this developing, only sky is limit. ).  But when look software side, at this time it is bottle neck and need urgently develop better inside factory culture - but, after next 5 - 10 year we know more.

For topic:

It depends what and how I define it for MY use:
93% 1 ch
5% 2 channel
2% 3 or more channels

Every user have digfferent needs.


But also, I have two hand. Of course if I have more hands I can perhaps then use more hands.
Do we urgently need channels or just for fun to play. 
Some times also need human brain to do things with simple methods, like example using razor blade and candle you can inspect even 1/10000 mm class anomalies in telescope head mirror diy  manufacturing. 
First is problem, after then thinking and then designing how to test and how to do probing etc and in very final after all thinking  start move the muscles and play with test equipments and solve problem.  Some do opposite order. They want and "need" all features and bells and whistles , because do not know what need and why.