Decent modern Signal/waveform Generator here in early 2024

[DaneLaw]

Looking for a relatively cheap modern signal generator, that is easy to adjust settings on. (important) and a step up from the cheap sub-100buck budget Chinese ones.

The last handful of years, been using an FY6900 DDS 60Mhz, it's been okay for the price but one annoyance is the cumbersome interface.  http://tinyurl.com/yckw8jxv
do cherish the ability to run it from the scopes USB-socket or a power bank and the mobility that comes with that.
* decent big screen, perhaps num-dial, perhaps touch, hackable?, dBm,

Budget is a relative lose 300 euros and what the alternatives are in that pricebracket (incl. VAT) its a more-or-less figure, can be more..

Wonder what direction people would go, with that budget in mind, and if they were in the market for a general function/waveform generator in early 2024?

[slugrustle]

Perhaps this isn't too helpful, as I don't know the market, but I'm happy with the Siglent SDG1032X that I use.  My biggest complaint is that the setup for an inverted pulse waveform (idle high, pulse low) is confusing.  I haven't used the arbitrary function on it yet, mostly just pulses and sine waves.  I wish it had double pulse capability; it doesn't.

I used a Rigol DG812 at work maybe sometime in 2020, and the interface was really buggy.  Otherwise, it seemed to work.  Hopefully they have much improved the firmware by now.

My requirements are not very high in terms of performance.  The main thing I look for is full 20Vpk-pk (±10Vpk) output on both channels.  Some models only give you that range on the first channel and something narrower on the second channel.

[Mechatrommer]

missing numeric pads is why i choose Uni-T UTG962 over FY6900. i know one day people will realize this if not mentioning it in public... UTG962 is small form factor and accepts 5V (DC jack) the drawback is trying to play arbitrary signal with it is a bit cumbersome. but i dont use arbitrary much. ymmv.
https://www.eevblog.com/forum/testgear/uni-t-utg932utg962-200msas-function-arbitrary-waveform-generator-220394

[bdunham7]

I wish it had double pulse capability; it doesn't.

Draw your own and save it as an ARB waveform!

[DaneLaw]

The Siglent SDG1000 and their entry units SDG1032 seem to have been extremely popular across the board, and the unit many people have purchased when looking for a decent sig-gen in that price bracket between 350 to 500 euros, also seems upgradeable' as far as I can tell, so no wonder its been the obvious choice for many, but it also looks like its closing in at 8 years since release which is a chunk in techworld regi, and likely not far from a new SG-version..

* SDG1032X - 30 MHz - 150 MSa/s - 14bit - 16 kpts - 2 CH - 362EURO (Incl. 25%VATDK)
* DG811 - 10Mhz - 125 MSa/s - 16bit - 2 Mpts - 1 ch - 300 EURO  (Incl. 25%VATDK) (this Rigol line also looks upgradeable, so the 1ch con could be countered)

The Owon UTG 30/60 line, is not on the plate, as it lacks some of the features, and its price is elevated quite a bit in the last few years (incl. VAT its around 200 and 250 depending on freq-model, which is around and in some cases more than double the 2019/2020 level.

[mawyatt]

A little above your budget but the SDG2042X is really a good AWG, easily "Enhanced" with help of clever folks on here to 120MHz.

If you have a Siglent DSO, then the SDG2042X works directly for Bode Function use, even allowing Frequency (and Phase, Amplitude) Offset sweeps as shown here with the built-in Channel Coupling Frequency Deviation feature:

https://www.eevblog.com/forum/testgear/siglent-sds2000x-plus-coming/4375/

Best,

[Martin72]

The SDG1032X is in deed easy upgradeable to 1062X.
That was my previous device, I currently have the SDG2042X, or 2122X...
At work we have generators from Keysight, Rigol, GwInstek and Siglent.
Rigol and Siglent because they have 2 channels, something I wouldn't want to do without.
If you can't/won't spend a lot of money, you can take the 1032X and upgrade to the 1062X.
Or a Rigol 2-channel from the same price range, I couldn't find any major advantages or disadvantages between the two.
I think the operation of all of them could be improved, which would be a sign of modernity for me if rigol and siglent were to change this.
And I still dream of a dedicated CMOS level output...

[tautech]

And I still dream of a dedicated CMOS level output...

16 channels enough ? 
SDG7000A.

[Martin72]

+15V Level?
OK, then I´ll take one....Wait a minute...9400€ entry-price....No Sir, thanks.

[Mechatrommer]

Wait a minute...9400€ entry-price...

may the force be with you..

[2N3055]

One thing to note is that Rigols are hackable but in a way that after "upgrade" they are not fully adjusted in whole new BW.
Rigol is not doing that. After "upgrade", "liberated" BW is not properly leveled. There are topics here on that...
Funny enough, older Rigol DG1000Z was fully calibrated and adjusted so that one worked properly even after "upgrade".

[wasedadoc]

In case you want to measure very low resistances ....

What is the relevance of that to the OP's question?

[blackdog]

Hi,

There are multiple function generators here.
As are power supplies and multimeters. :-)

The Siglent SDG1032X is the cheapest function generator I have here in the pictures and it is the most used.
Why, look at the picture showing how much abberation is present on the pulse display.
This is of course only one of the many functions of such a generator, but what is important for me, for the measurements I often do.

The Siglent SDG2042 is a 16Bit generator and that sometimes has its advantages and finally the Rigol DG4162 which has the important advantage of a 160MHz bandwidth.

The test condition here is 900mV TT at 50 Ohm.
This is because the Siglent SDG1032X has a special output amplifier if you stay below 1V TT output voltage.

The coax cables are good quality and I use a 20dB BNC attenuator on the scoop inputs to keep the corresponding generator from seeing reflections from the 50 Ohm scope input.

First picture
Here you can clearly see that the pulse of the Siglent SDG1032X has the highest quality.
SDG1032x = 3.3nSec
SDG2042x = 8nSec
DG4162     = 4.2nSec


.
This is the contraption to show how this measurement was done.
Everything is locked to a GPS reference and I had to make quite a few phase adjustments to compensate for generator delays and cable length variations. :-)


.
This picture is to show you how to get the best possible pulse on your scope, usually use at least a 10dB attenuator of good bandwidth and turn on the 50 Ohm input of your scope.
Also use quality good coax cables and connectors! Don't kid yourself.


.
So if a really nice pulse response is important to you then I could recommend the Siglent SDG1032.

That's the reason I bought it, even though I already had plenty of function generators.
Here in the Netherlands this Siglent SDG1032X costs about 335 Euro and I than recommend this generator to many people.
A lot of hobbyists and Pro users are not really aware of the possibilities that the modern function generators give you.

Example
Last week tested some AM detectors for demodulation distortion, a week before that tested some Darlingtons and Sziklai pairs for bandwidth directly fed from the SDG1032X.
If you need a DC source, you can also use the function generator in a number of applications and there are many more.

I wish I could have bought something like a Siglent SDG1032X 30 years ago.
Back when I still had a lot of HAM applications for my hobby.

I hope this helps a little in the considerations.
And then I haven't discussed a lot of features of modern feature generators.

Kind regards,
Bram

[nctnico]

missing numeric pads is why i choose Uni-T UTG962 over FY6900. i know one day people will realize this if not mentioning it in public... UTG962 is small form factor and accepts 5V (DC jack) the drawback is trying to play arbitrary signal with it is a bit cumbersome. but i dont use arbitrary much. ymmv.

IMHO the Uni-t models are a good buy. The more expensive Siglent SDG1000 doesn't bring much extra to the table in this market segment so value for money is low.

And I still dream of a dedicated CMOS level output...

Forget about that. The problem is that a waveform generator doesn't have the features you really need in a digital pattern generator. Been there, done that. If you want a digital pattern generator, then buy a digital pattern generator with good software.

[blackdog]

Hi

The more expensive Siglent SDG1000 doesn't bring much extra to the table in this market segment so value for money is low.

That of course depends on what your application is, for your applications the Uni-T suffices thats OK, others and I obviously have different preferences, that's how life is. 

Kind regards,
Bram

[nctnico]

Hi

The more expensive Siglent SDG1000 doesn't bring much extra to the table in this market segment so value for money is low.

That of course depends on what your application is, for your applications the Uni-T suffices thats OK, others and I obviously have different preferences, that's how life is. 

That is the kicker, there is next to nothing the SDG1000 is actually better at compared to the UTG932 / UTG962. And the UTG932 / UTG962 have better specs compared to the SDG1000 where it comes to jitter for example.

[KungFuJosh]

I used to have the UTG932. It was really nice, and really small. I liked it.

[Smokey]

Do any of these units have programmable/adjustable rise and fall times?  I had an application that I needed to dial in a specific rise and fall time on the edges and my present function gen can't do that.
 I'm looking for that as a feature on the next thing I get.
I get that I could probably write an arb routine to set that, but I just want to be able to tweak rise/fall times on the fly on the square wave. 

[nctnico]

Both SDG1000 and UTG900 series have this feature and are pretty much on par.

[tautech]

Do any of these units have programmable/adjustable rise and fall times?  I had an application that I needed to dial in a specific rise and fall time on the edges and my present function gen can't do that.
 I'm looking for that as a feature on the next thing I get.
I get that I could probably write an arb routine to set that, but I just want to be able to tweak rise/fall times on the fly on the square wave.

Certainly SDG1000X supports adjustable rise and fall times in Pulse mode.
You can see this in the User manual P129
https://int.siglent.com/u_file/download/22_06_23/SDG1000X_UserManual_UM0201X-E01E.pdf

[Smokey]

Cool.  Thanks.

[Martin72]

Both SDG1000 and UTG900 series have this feature and are pretty much on par.

Which of the two do you use yourself, or even both?
I am also a friend of as cheap as possible and also think that some manufacturers are simply overpriced.
But if something costs almost half as much but is supposed to be just as good, as you said, then I have "natural doubts".

[MathWizard]

I have the SDG2042X, and I've had no problems with it. 1st ever real sig. gen, and it's easy to use.

Sometimes I would like it if when changing frequency on the 1M digit, it would go down to 900kHz, not 0Hz, when you turn the knob. But sometimes that's ok.

The 2042X is great for the money. I have my eye on the SDG6022X, that will suit me perfect, and the next range, the RF gen's 7000series, are way way more expensive, so I'll be delighted if I can hack it to 500MHz.

[RoGeorge]

Rigol DG4162 which has the important advantage of a 160MHz bandwidth.

Thanks for the printscreen with all the 3 outputs.  Looks like there is a small reflection coming back, the step after the 1st division at each edge, most visible on the Rigol trace.

For that Rigol DG4000 series, there are 3-4 models with same hardware and software limited to 60...200MHz.  Just so you know, any DG4000 model can be unlocked to 200MHz (sinusoidal) and turned into a DG4202.

14bits DACs, 2 output channels, it's nice to have signal and sync out BNCs for each output, external trigger, and has a 200MHz counter/frequency meter, too, with a BNC on the back.  Can generate up to +/-10Vpp in a 50ohms load (at lower frequencies, at 200MHz only 1Vpp), and has a DC mode, too, which makes it a dual channel +/-20V adjustable voltage source with a 50ohms internal resistance, so it can power small current loads as if it were a power supply. 

Has 10MHz in/out, too, to drive the internal clock synchronized with other instruments.  Worked OK so far, though it costs 2-3 times the 300 euro proposed budget.

[2N3055]

Hi

The more expensive Siglent SDG1000 doesn't bring much extra to the table in this market segment so value for money is low.

That of course depends on what your application is, for your applications the Uni-T suffices thats OK, others and I obviously have different preferences, that's how life is. 

That is the kicker, there is next to nothing the SDG1000 is actually better at compared to the UTG932 / UTG962. And the UTG932 / UTG962 have better specs compared to the SDG1000 where it comes to jitter for example.

While I think that UTG962 is good value for money, and certainly much better than what you would expect for the price, let's not get carried away.
SDG1000X is a piece of entry level "proper"  lab equipment. Square wave edges on UTG962 are 16 ns, and 4ns for SDG1000X.
On UTG962 jitter is actually better, but only at >5MHz. At below 5MHz it is 2ppm+200ps VS 0.005 ppm+300ps on SDG1000X.
UTG962 has no PWM modulation, ?
Timebase accuracy is worse on UTG962. It is a small plastic box with external wall plug type PSU.
Which can be useful if you need to carry it around and need something simple for small desk...
It would make very nice little AWG for a go bag with Micsig or Pico, together with one of those MiniWare PSU.
I agree that for most hobby users and even for pro use as a kind of "around the shop" little handy AWG, UTG962 is worth it's money.
Most people will never need more.

But the SDG1000X is more expensive for a reason, as are Rigol entry level AWG.  Is it worth the money difference, is up to every one to decide for themselves.