I latched on to this thread in the hope of enlightenment. I want to buy my first scope and have been struggling to make an actual decision due to lack of experience and knowledge. The thread's been interesting, but I fell short of enlightenment...
I'm not very experienced with electronics, not used a scope of any sort in 40+ years. I started out building kits, migrated to building from PCBs from people on DiyA and PCB's I'd ordered from gerbers published here too. I have, somewhat unexpectedly, found myself about to order boards I've laid out based on schematics I've found on the WEB.
So I need to be able to test what I build to see if it's working to spec and diagnose faults when it's not. I'm generally doing audio, though with some low voltage Arduino digital thrown in, possibly also some valve audio. Limiting myself to layout and board stuffing for the foreseeable.
Suggestions for a spec or actual scopes I can consider would be really helpful in narrowing down the search. I'd thought 2 channel, 1GHz and possibly something more than 8 bit resolution, but is that so?
I'm not very experienced with electronics, not used a scope of any sort in 40+ years. I started out building kits, migrated to building from PCBs from people on DiyA and PCB's I'd ordered from gerbers published here too. I have, somewhat unexpectedly, found myself about to order boards I've laid out based on schematics I've found on the WEB.
So I need to be able to test what I build to see if it's working to spec and diagnose faults when it's not. I'm generally doing audio, though with some low voltage Arduino digital thrown in, possibly also some valve audio. Limiting myself to layout and board stuffing for the foreseeable.
- I'd prefer a digital scope (size and convenience) and, ideally, one that can transfer data to a PC or even be controlled from a PC.
- If it's a digital scope I'd want to be confident the software driving it was supported and updated
- Looked at S/H and decided postage for scopes outside the UK was prohibitive, but I'm not opposed to buying S/H if I know what I'm looking for.
- I want a scope that'll do what I need now, with a little future proofing so I haven't outgrown it after a year or two.
- Probes? I've even less idea about type and variety.
Suggestions for a spec or actual scopes I can consider would be really helpful in narrowing down the search. I'd thought 2 channel, 1GHz and possibly something more than 8 bit resolution, but is that so?
Do you mean 1GSPS sample rate? 1GHz front-end bandwidth and 1GHz probes are a _lot_ more expensive - I'd suggest thinking 100 to 200MHz front end and probes, 1GSPS, 8 bit acquisition for something affordable. Although specs are getting better all the time.
To use a 1GHz front-end effectively you need to use special low-inductance probe tips and active probes. This is microwave frequencies.
Some 'scopes do resolution enhancement by dither/averaging to get more effective bits at lower bandwidths, for instance my Siglent SDS1104X-E does upto 11 bits effective this way.
Yes. I did mean GSPS but didn't notice 'til it was too late...
I'd been looking at the SDS1104X-E after earlier comments. It's over the budget I'd allowed for myself, and I wondered if I would be wildly over spending for a first scope. OTOH, I'm unlikely to outgrow it or be upset when I hit software problems. It's not so far over budget that I'd feel uncomfortable with the spend.
I got put off some of the cheaper two channel Siglents because they appear to have stopped doing firmware updates for them. Some have had nothing since 2018. Which, for hardware utterly dependent on software means either it's end-of-life or the software is so good, they can't possibly improve it. I've never seen software that didn't need something fixing...
I really can't think of a situation where I'd need four channels though or, probably, half the features the scope provides. Hantek appear to have some nice scopes... And so I continue to go round in circles
I'd been looking at the SDS1104X-E after earlier comments. It's over the budget I'd allowed for myself, and I wondered if I would be wildly over spending for a first scope. OTOH, I'm unlikely to outgrow it or be upset when I hit software problems. It's not so far over budget that I'd feel uncomfortable with the spend.
I got put off some of the cheaper two channel Siglents because they appear to have stopped doing firmware updates for them. Some have had nothing since 2018. Which, for hardware utterly dependent on software means either it's end-of-life or the software is so good, they can't possibly improve it. I've never seen software that didn't need something fixing...
I really can't think of a situation where I'd need four channels though or, probably, half the features the scope provides. Hantek appear to have some nice scopes... And so I continue to go round in circles
Software update is not a soo good thing.
At the expensive scopes from my job, the software updates, at some directly at the supplier during services (HDD and power supply replace) create only problems.
The scope became slower and with a delayed response, some features become difficult to use. And I do not speak about cheap oscilloscope but expensive, some veri expensive over an average new car.
At the expensive scopes from my job, the software updates, at some directly at the supplier during services (HDD and power supply replace) create only problems.
The scope became slower and with a delayed response, some features become difficult to use. And I do not speak about cheap oscilloscope but expensive, some veri expensive over an average new car.
I would say crap programming.
At work I use an oscilloscope that is already about 10 years old but comparable to the new equivalent versions that came to replace the old model, it was much faster especially if we activate measurements.
Because it was old, some maintenance repairs had to be done, the capacitors in the power supply had to be replaced and the HDD replaced with an SSD.
It was sent to the manufacturer for repair. It's back with the latest firmware version that runs slower and some features are harder to use (clear SW bugs).
When the newer versions were bought, I refused a new one and I preferred to work with the old one because it was faster. Now it moves just as slowly.
At work I use an oscilloscope that is already about 10 years old but comparable to the new equivalent versions that came to replace the old model, it was much faster especially if we activate measurements.
Because it was old, some maintenance repairs had to be done, the capacitors in the power supply had to be replaced and the HDD replaced with an SSD.
It was sent to the manufacturer for repair. It's back with the latest firmware version that runs slower and some features are harder to use (clear SW bugs).
When the newer versions were bought, I refused a new one and I preferred to work with the old one because it was faster. Now it moves just as slowly.
Not entirely crap programming.
Often the software's developed from a single codebase and moves forward to take advantage of the microprocessor in use in newer models.
Inevitably, there's more power and memory available and the developers are often forced to either cripple newer to keep older alive, or find workarounds to keep the older versions working but at slower speeds.
Usually, it's the latter that's chosen.
The alternatives are to fork the source or to embed conditional code that does different things depending on where it's running. It's a LOT more work, changing, testing and releasing multiple versions for every release. Few development groups are large enough or well funded enough to be able to do that sort of thing.
That said, there are also a lot of careless programmers out there too, and management just want the stuff out the door ASAP regardless of QC issues. It can be a very toxic environment to work in if you *do* care about what you're doing...
Often the software's developed from a single codebase and moves forward to take advantage of the microprocessor in use in newer models.
Inevitably, there's more power and memory available and the developers are often forced to either cripple newer to keep older alive, or find workarounds to keep the older versions working but at slower speeds.
Usually, it's the latter that's chosen.
The alternatives are to fork the source or to embed conditional code that does different things depending on where it's running. It's a LOT more work, changing, testing and releasing multiple versions for every release. Few development groups are large enough or well funded enough to be able to do that sort of thing.
That said, there are also a lot of careless programmers out there too, and management just want the stuff out the door ASAP regardless of QC issues. It can be a very toxic environment to work in if you *do* care about what you're doing...
I'd say go for the 1104X-E. 1GHz is likely way overkill, and more than 8 bit is not needed, especially for a first scope.
Just last week I splurged and treated myself with an 1104X-E to augment my small fleet of analog and mixed signal CRO's, which include a couple of 400MHz Tek 2465B's as the fastest. It's a joy to use the 1104. I deem it's UI performance good enough.
With a ('hacked') 200MHz frontend it's plenty good for my work and four channels is a joy when doing digital or correlating multiple signals. 'Hacked' is a bit of a misnomer as it only requires a few and very simple things to do. Firmware then upgraded to the latest, which is a few months old.
I might even buy the AWG and 16-channel logic probe for it, although I'm pretty much covered on both fronts already, since I only need to buy the HW now and not licenses too.
I wouldn't go for the Rigol 1052e but Rigol 1054Z is an option. The 1104X-E is the better choice though IMO.
Just last week I splurged and treated myself with an 1104X-E to augment my small fleet of analog and mixed signal CRO's, which include a couple of 400MHz Tek 2465B's as the fastest. It's a joy to use the 1104. I deem it's UI performance good enough.
With a ('hacked') 200MHz frontend it's plenty good for my work and four channels is a joy when doing digital or correlating multiple signals. 'Hacked' is a bit of a misnomer as it only requires a few and very simple things to do. Firmware then upgraded to the latest, which is a few months old.
I might even buy the AWG and 16-channel logic probe for it, although I'm pretty much covered on both fronts already, since I only need to buy the HW now and not licenses too.
I wouldn't go for the Rigol 1052e but Rigol 1054Z is an option. The 1104X-E is the better choice though IMO.
Speaking of the 1104X-E, can someone explains to me what functions the hack unlocks?
I know that it increases the bandwidth to 200MHz but other than that, is there anything else that offers like software options, etc...
Since I only do analogue audio I'm not sure whether it's worth the effort to hack it to 200MHz. i'm not experienced at all when it comes to programming so I hesitate a bit, and I'm afraid I might cause some problems to the scope...
I know that it increases the bandwidth to 200MHz but other than that, is there anything else that offers like software options, etc...
Since I only do analogue audio I'm not sure whether it's worth the effort to hack it to 200MHz. i'm not experienced at all when it comes to programming so I hesitate a bit, and I'm afraid I might cause some problems to the scope...
So apart from the 200 MHz, basically unlocks the license for the optional modules that you can buy:
1. Support for the separate waveform generator (you can configure it from scope menu and do Bode plots).
2. Menu support for the 16 channel logic analyzer module
3. Support for USB Wifi adapter.
You would have to buy the modules to do anything with the two first ones. If you already have a separate function generator you can do Bode plots the normal way without this option. The last one is where you actually save some, because you can find the only compatible Wifi adapter cheaper (only works with TP-Link TL-WN725N as far as I know).
The only thing I don't like with the Siglent SDS 1104X-E (I bought it last year), is the fan. It is rather loud. I might put a more silent fan in it at some point, or just lower the voltage for the current fan. I have no experience of other modern oscilloscopes, so I don't know if their fan sound level is similar. Other than that it feels like an excellent scope for the hobbyist and some of the best you can still buy for this price.
I like my Fluke 123 for analog audio stuff. At just 20MHz, probably isnt good for digital debug...
However, if I wanted to display the inductor current in some pi filter with both ends of a sense resistor 500V above ground - no problem, using the stock probes it came with. It's like a DMM in that regard. That's why I keep it.
The probes suck for connecting to small proto board stuff, as they were made for connecting to industrial strength apparatus. There's no alternative I'm aware of - proprietary connector (versus BNC) and all.
However, if I wanted to display the inductor current in some pi filter with both ends of a sense resistor 500V above ground - no problem, using the stock probes it came with. It's like a DMM in that regard. That's why I keep it.
The probes suck for connecting to small proto board stuff, as they were made for connecting to industrial strength apparatus. There's no alternative I'm aware of - proprietary connector (versus BNC) and all.
One may as well move the top end up to 200MHz, it'll be a flatter response at 100Mhz
No real "effort" involved
It doesn't cause "some problems", it just tells the front end amp chips not to restrict the bandwidth....
In the 100MHz version during power up the firmware actually tells the front end IC to restrict bandwidth, to effectively cripple the full performance of the front end.
https://www.ti.com/lit/ds/symlink/l...73621&ref_url=https%3A%2F%2Fwww.google.com%2F
With the hack you're just telling the firmware to stop doing that.....
nothing more devious.It's a common technique in modern stuff, to build one full performance front end (like this) for all models and then use software to restrict performance in the lesser model variants.
I think it's going to be a 1104X-E for me too.
Interestingly though, they're not the most grossly overpriced things to be found. Crimper dies take the biscuit for that. Farnell have plenty listed at £2-400. Just the die, no crimper tool included... It's like CNC has never been invented.
I see Farnell are happy to sell you some for > £2k
Interestingly though, they're not the most grossly overpriced things to be found. Crimper dies take the biscuit for that. Farnell have plenty listed at £2-400. Just the die, no crimper tool included... It's like CNC has never been invented.
I think it's going to be a 1104X-E for me too.
I see Farnell are happy to sell you some for > £2k
Interestingly though, they're not the most grossly overpriced things to be found. Crimper dies take the biscuit for that. Farnell have plenty listed at £2-400. Just the die, no crimper tool included... It's like CNC has never been invented.
There’s a 25MHz testec 1KV probe £400. However for the interim I’ve designed in ground scaled test points with a voltage divider as part of the amp (+320 to -320 and the 10x probe is 600V max). The test points will be at most 300V from ground.