Function generator <300$

[alex89]

Hello guys, I'm a big fan of the blog but this is my first post on the forum. I didn't see any section to introduce myself so I hope I'm not doing wrong. I know this a recurrent question and I looked through all the related post but still not sure on what to buy. My budget is not huge, around 300$, and I'm looking for a generic use function generator. I work almost always on audio stuff so a 5Mhz fg is enough for me. I'm actually considering these units:

- HP 3325A
- Siglent SDG805
- Siglent SDG1005

I know the 3325A is rather vintage but still very usable and with good spec too. It is not uncommon to find it on ebay for less than 300$ and it actually is my first choice. Also the shematics availability is a plus. All I know about the Siglent SDG1000s is they are not that great on rise time performance, there is some jittering (maybe fixable with a better reference signal ?) and their firmware is not the best. The arbitrary waveform function is pretty interesting though and it can drive two different output. There is indeed not a lot of information about the new SDG800 series but it seems they should be better in regard jitter and rising time performances. Don't know if there are any software improvement/limitation compared to the 1000 series. Any suggestion ? Am I not considering all the possibilities ? Any consideration is welcome :-)

[dom0]

I wouldn't focus on the very-well known function gens on the used market, since they are popular they are more expensive.

Look for not so well known models.  For example, HP 3312a, HP 3324A, Philips/Fluke generators (e.g. pm5138, pm5139, pm5136), there are also always some Wavetek generators around, but I found it hard to get solid specs on them and skipped them. That's a very long list.

Patience pays, I bought my fg (50 MHz Philips) for c. 150 $ (excluding shipping etc.), but I had to wait a few months for a good offer... OTOH it might be hard to get spare parts for "unpopular" generators. But then again, spare parts are usually rather expensive... especially for HP, Tek, ... devices.

When doing audio stuff it might be worth to look for an additional sine oscillator or some fg with lower distortion numbers than general-purpose fun-gens, since those usually have rather high distortion (most synthesized fgs from the 80s-90s are in the 0.1-0.5 % ballpark). Building a good, solid, low distortion sine oscillator is btw. a nice project to do. I recommend it.

[wn1fju]

Be careful with the HP 3325A as only the sine and square go to an appreciable frequency.  The triangles and ramps only go up to something like 11 kHz.

If synthesizer-like stability/accuracy is not of the utmost importance, might I suggest the HP 3314A if you like vintage HP equipment.  It's a very capable unit (has arbitrary waveform), reasonably small and only uses a couple of proprietary HP-hybrid ICs.  If you are adventurous, look for a discount on one on eBay that "doesn't power up."  The power supplies are standard and easily fixed.

[dom0]

Be careful with the HP 3325A as only the sine and square go to an appreciable frequency.  The triangles and ramps only go up to something like 11 kHz.

This is a common limitation of all early synthesized generators I am aware of. Generally, always fetch and read the specs for gear if you are about to spend an appreciable amount of money.

There is also the 8112A/8116A series. But they're not overly good generators IMHO. They're based on some very tricky analog oscillators, controlled by DACs, and have rather poor stability and precision. In this regard they are not better than analog generators.

There is also the 8165A, a very early synthesizer. I don't know anything more about it.

[pickle9000]

I have an SDG1020 and it's really works well. Easy setup no complaints.

I'm pretty sure that's the same family as the 805.

[tautech]

Carefull examination of the Siglent datasheets will show considerable differences in risetime and drive(p-p) capability between the SDG 800 and SDG1000 series.
The 1000 series are a more capable unit. Pulse risetime 7 ns.

[MartyD]

Carefull examination of the Siglent datasheets will show considerable differences in risetime and drive(p-p) capability between the SDG 800 and SDG1000 series.
The 1000 series are a more capable unit. Pulse risetime 7 ns.

Which I can confirm, I've measured between 6,8 and 7,2nS across various waveforms and the frequency range of my SDG1025. THD also isn't too shabby, great little Generator for the money.

Marty

[alex89]

Thanks for the heads up guys ! Got the SDG1020 and pretty happy with it 

[rf-loop]

Carefull examination of the Siglent datasheets will show considerable differences in risetime and drive(p-p) capability between the SDG 800 and SDG1000 series.
The 1000 series are a more capable unit. Pulse risetime 7 ns.

Which I can confirm, I've measured between 6,8 and 7,2nS across various waveforms and the frequency range of my SDG1025. THD also isn't too shabby, great little Generator for the money.

Marty

If look 20Hz - 20kHz THD then Siglent SDG800 is better than SDG1000. Much better.

Time ago I made tiny test:

SDG805 sinewave 20Hz to 20kHz (Keithley 2015THD works only with these limits)

Signal with coax cable to Keithley 2015THD.
Kethley input banana/BNC adapter - 50ohm feed thru terminator.
Level set for 1Vrms.
Result:
20Hz THD 0.01%
200Hz THD 0.008%
1kHz THD 0.007%
5kHz THD 0.006%
20kHz THD 0.006%

Fastest risetime by SDG805. (example Arb function, and here in picture "ramp" function. (triangle using low duty)
With square/pulse around 20-24ns.

[commongrounder]

I'm quoting my post from over on the Siglent SDG 1000 and SDG 800 thread to bring the distortion measurements between the two together in one place:

(quote) I (am) impressed with the low distortion from the SDG800 series.  The SDG1000 series doesn't do as well.  I measured my SDG1005 using my Tektronix AA501 distortion analyzer (residual distortion .0008%, 100khz bandwidth,  RMS reading).

Output #1:
1khz frequency @ 1 volt RMS:  .0879%  THD+N
1khz frequency @ 6.5volt RMS, (max. output, 600 ohm ref.):  .0569% THD+N

Output #2:
1khz frequency @1 volt RMS: .0758% THD+N
1khz frequency @1.958 volt RMS, (max. output, 600 ohm ref.): .0381% THD+N

Interesting that the number two output has lower distortion.  The drive circuit, although lower in output, appears cleaner.   I wish I could get my hands on an SDG800 unit to compare/corroborate RF-Loops findings, using my setup.(unquote)

Looking like around an order of magnitude lower for the SDG800.  The only caveat is that these two tests were done using different measurement setups.  It is unlikely I will have an opportunity to test a SDG800, so maybe we can hope that RF loop can test a SDG1000 sometime with his Keithley.

[rf-loop]

I will try inside some days do tests so that there is SDG1010, SDG805 both available and do as equal settings/tests as possible.

2012 tests I have only noted about SDG1025 that between 20Hz to 20kHz  THD is between 0.03 - <0.1% depending frequency and level with same Kethley 2015THD (but can not remember exactly what was level and load impedance etc.)

Also this image is from 2012 tests (SDG1025)

[rf-loop]

@commongrounder

I can do measurement set for SDG5082, SDG1010, SDG805  so least results are comparable between these equipments (and individuals)


THD can measure with many kind of settings so I will ask if it is ok/useful with this kind of settings.
(I'm not at all audio specialist)

I measure using Kethley 2015THD with  setup:
Keithley AC range jump to 10V range if I use +4dBu (1.23Vrms, standard pro audio)  level.
So I will use 1Vrms (0dBV) adjusted so that all generators all outputs measured with same Kethley with 1kHz is as close 1Vrms as possible. Is this level ok? or is it better select some other level?

Terminator 50ohm (banana / BNC adapter - 50ohm feed thru terminator - coaxial to SDG output)
Order of harmonics: 5    <--- Is this ok? Or do I need select some other amount of harmonics.
Filter: 20 readings average, mode repeat
Freq: Auto
Filters: None
AC range 1V

[dom0]

1 Vrms is on the low side of the useful range for audio THD measurements, since the noise floor of the entire setup is (optimistically) usually somewhere around 3-5 µVrms (~-105 dB. Or, if you want the damned percentage numbers, about 0.0005 % THD+N).

Around 0.008 % is very good (for a general purpose function gen), yes, but not too surprising: it means that Siglent managed to get very close to 14 bit ENOB.
Of course, thinking of analog function gens with their THD usually in the range of -50 to -60 dB due to the diode sine shaping networks of the time this is a great result.

[commongrounder]

Thank you, RF-Loop, for your willingness to perform this comparison test.   Looking at the Keithley manual, I think your settings are generally good.  1Vrms is low for THD+N measurements, but it is also a common real world level, and I think appropriate for these purposes, especially since the 2015 can work at the top of a lower range (better accuracy).  I would initially set the "type" to THD+noise, mainly for selfish purposes, to compare to my AA501 analyzer.  In this mode, the "order of harmonics" setting is disabled, and the meter measures up to 64 times the fundamental or, in this case with 1khz fundamental,  the upper detector range of 50khz.  No filters is probably okay.  Hum components might be an issue, but you would have to analyze that separately to figure out if that will bias the readings enough to need to filter the results.  It might be interesting to do a THD vs. THD+N comparison to look for dramatic differences.

The Keithley 2015THD is an interesting meter.  I think I may need to get my hands on one!

I look forward to seeing your results!

edit:  I just saw the residual noise floor of the 2015 in THD+N mode, which is -65 db or .056%.  That will certainly skew/swamp the distortion readings of the generators, so plain THD mode with 64 harmonics should give use the best comparison.  Thanks!

[rf-loop]

Some test results here
https://www.eevblog.com/forum/testgear/the-sdg1000-and-sdg800-thread/msg644687/#msg644687

[DJ]

Some test results here
https://www.eevblog.com/forum/testgear/the-sdg1000-and-sdg800-thread/msg644687/#msg644687

Thank you for the work doing the testing.

The SDG5082 did not fare well against the competition.

Has Siglent claimed better performance or offered software/firmware upgrades?

Am guessing SDG5122 and 5162 will be similar.

[rf-loop]

Some test results here
https://www.eevblog.com/forum/testgear/the-sdg1000-and-sdg800-thread/msg644687/#msg644687

Thank you for the work doing the testing.

The SDG5082 did not fare well against the competition.

Has Siglent claimed better performance or offered software/firmware upgrades?

Am guessing SDG5122 and 5162 will be similar.

It is still well inside given specifications.
As I have told before if need is better audio sinewave quality and nothing else, order is: SDG800, SDG1000 and then SDG5000.