The Siglent SDG1000 and SDG800 thread

[Miroslav]

I wouldn't open it, why risk Siglent honouring the warranty and you benefiting with a perfectly good working unit.

BTW did you pay much for it?

I'll wait

Can't tell exactly as I bought together with some other equipment as a package.

[tautech]

Bootstrap 46. Perhaps ypu have not powered up it so many times.
After factory you start it first time it is 1. (After Siglent original new package I have noted boot up counter is alwaqys treseted. (I do not know how they reset it after factory tests and factory calibration)

When I sell to customers it is (typically) between 10-25 due to my own >24-48h  "burn in" run with many thermal cycles and after then tests.

46 looks like it have used some amount after factory.

OTOH units I supply, I only usually power up only once to confirm to myself they are operating correctly knowing that some factory "burn in" has occured.

For anyone with a "new toy" probably 10- 20 power cycles would be expected initially, but from a lab it might get used more frequently than a hobbyist would and if their are a number of technicians the could have easily mad up that number of start in a few months.

I know for much of my gear most of the start #'s are from answering questions on EEVblog not real work.

[rf-loop]

Bootstrap 46. Perhaps ypu have not powered up it so many times.
After factory you start it first time it is 1. (After Siglent original new package I have noted boot up counter is alwaqys treseted. (I do not know how they reset it after factory tests and factory calibration)

When I sell to customers it is (typically) between 10-25 due to my own >24-48h  "burn in" run with many thermal cycles and after then tests.

46 looks like it have used some amount after factory.

OTOH units I supply, I only usually power up only once to confirm to myself they are operating correctly knowing that some factory "burn in" has occured.

OTOH continue.
Yes I believe factory do something like 24 hour burn in. But I have not any proofment they do it for every single unit and that they also do check/factory cal after it. 
Before I have worked with some higher level manufacturer and there they did 48 hour real "burn in" and more "fun" also this burn in was in made in very big oven using specified maximum temperature + some marginal over, where conveyor lines and elevaors  moving and trasporting whole aging time in queye like "FIFO". Whole burn in (aging) test was fully automatised and system drop out units what did not pass or fails in burn in. (I made this aging system automation for this company somewhere beginning of 90's ) We know this electronics failure "bathtub curve".

24 hour aging is not enough and mofre bad if this is just one power up cycle. It need turn on and off several times for teperature cycles (it made also mechanical stress example for big BGA chips.)

(Here I can say that with Owon I did allways 72hour burn in (aging) using also mechanical shocks and external heating, until I  get enough experience and drop it to 48hour)

Reason why I do is that I do not want that any single customer get lemon. If equipmen have survived my test, probability to failure in first months and years is much less than without this.  Also, this I can do because I do not sell for for get money for living. I have enough.

[Wmacky]

Just got a SDG1025 in, and it seems really nicely made for the price. No disappointments at all so far!    Can't wait to tear into it this weekend!  RF-Loop  -  I see the SDG 800 had very good THD numbers. Did you try the same with the SDG1000 units?

also, I read that the SDG800 does better with square waves at lower freq due to a different approach. Are their other negative hits for using the SDG800's square wave approach, or pluses for using the SDG1000 comparator approach?

[commongrounder]

I was also 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 SDG 800 unit to compare/corroborate RF-Loops findings, using my setup.
Here is a screen shot of the residual distortion output of the analyzer for the output #1 at full output.

[Wmacky]

That is rather interesting, as I was led to believe that the SDG1000 models were the better units. Perhaps the 800's are the ones to get?  Do they have any disadvantages other than being single  channel? I've read another report showing figures much the same as yours. It would be nice to see both units tested together as you suggest.

[commongrounder]

Well, I suppose "better" depends on your perspective and needs.  The SDG1000 series has been around longer, but has two independent (or couple-able, phase shiftable) outputs. The SDG800 series uses newer technology borrowed from the SDG5000 series generators (Easypulse), but with a single output.  I personally don't expect to use a function generator for distortion measurements, so the higher distortion is acceptable to me (my old analog B&K FG had a THD rating of .5%, so things are much improved, in any case).  Level flatness is as good as my Tektronix FG and SG generators, so no complaint there.  I guess if you want the low distortion, and output flexibility, with a Siglent, you have to go to the SDG5000 series.

[Wmacky]

Yeah, your right. The 1000 numbers aren't that bad, just a little disappointing after seeing how great the cheaper 800 did......

[Wmacky]

Hmmmm.



I was kinda expecting a cleaner square wave?   This was at 4Vpp and 1MHZ.

[Wmacky]

Geez! Nevermind.   

A proper 50ohm cable rather than 75, and setting the DSO to a 50ohm termination seems to have neatened up the square wave rather nicely.

[rf-loop]

Here some test results from THD measurements.

[Wmacky]

Thank you this is great info. Now we need the Rigol 1022 to get the whole picture! 

[rman_nl]

I can create a dc signal on my sdg1010. Can I use this as a power supply for my small and low voltage (0 -5V) electronic (breadboard) experiments?
Is it short circuit protected?

[nctnico]

The output always has a 50 Ohm series resistance so you can't really use the generator as a DC power supply. The accuracy is also limited because 50 Ohms is rather high.

[rf-loop]

Is it short circuit protected?

Yes, you can drive what ever SDG800/SDG1000/SDG5000 signal from its output  to what ever load impedance, including zero impedance (short circuit).

[Miroslav]

Yes, you can drive what ever SDG800/SDG1000/SDG5000 signal from its output  to what ever load impedance, including zero impedance (short circuit).

The output always has a 50 Ohm series resistance so you can't really use the generator as a DC power supply. The accuracy is also limited because 50 Ohms is rather high.

rf-loop,
For the sake of correctness, I need to "amend" your comment with nctnico's - in an ideal case, the "power supply" (usually meant as a constant voltage source) should have zero ohm output resistance (impedance). With 50 ohm that SDG provides, any fluctuation in output current will cause fluctuation in the supply voltage.

If the circuity using this supply with 50 ohm output resistance draws (for example) between 10mA and 60mA from the power supply, the supply voltage (SDG output) will have fluctuations of (at least) 2.5V !!!

[commongrounder]

Here some test results from THD measurements.

Thank you, RF loop for the great comparison info.  Two things stuck out for me. One is the relatively poor (not bad, just worse than the SDG800) performance of the SDG5000 series.  I would have expected better, since I assumed it had the generator design (Easywave) the SDG800 series was based on.  Obviously some other circuit differences coming into play.  The other thing is the distortion characteristics of the SDG1000 series seems to hint that most of it is in the low harmonics.  Adding more harmonics to the Keithley measurement FFT did not seem to cause a dramatic increase in the THD reading.  I might try some more tests with filtering and FFT analysis to see what the distortion components are.

[rf-loop]

The other thing is the distortion characteristics of the SDG1000 series seems to hint that most of it is in the low harmonics.  Adding more harmonics to the Keithley measurement FFT did not seem to cause a dramatic increase in the THD reading.  I might try some more tests with filtering and FFT analysis to see what the distortion components are.

Perhaps you have seen these old tests about SDG1000 series.
These are not just for audio raange but there is also some spectrum images what may tell something about audio freq range.
At this time my HP8568B is not just now available for tests (my workshop is under work/renew/move  and it is not ready and not time for it just now). Also it is not very good for audio range but better than nothing.

http://siglent.freeforums.org/tests-updated-siglent-sdg1000-function-arbitrary-wfm-gen-t5.html

[commongrounder]

Ah, yes!  I do remember reading through your posting there when I was researching the Siglent AWGs, pre-purchase.

This 1 kHz plot here:

http://i181.photobucket.com/albums/x2/aghp55/plot0036_1k.gif

(I hope you don't mind my "quoting" your screen shot),  very clearly shows the relatively strong first even and odd harmonics.  Pretty quiet up-frequency from there.  Interesting!  Even so, the first even harmonic is around -62db, so very respectable

[ivan747]

I can create a dc signal on my sdg1010. Can I use this as a power supply for my small and low voltage (0 -5V) electronic (breadboard) experiments?
Is it short circuit protected?

Essentially, the output of the function generator has a 49 ohm resistor in series. I *think* short-circuiting the output to ground couldn't damagae the output amplifier nor the resistor, but I need somebody to confirm this.

You can use it as a bias voltage for an up-amp or transistor, or something small like that, but I wouldn't power an entire project with it. Even batteries are better than that.

[rf-loop]

Essentially, the output of the function generator has a 49 ohm resistor in series. I *think* short-circuiting the output to ground couldn't damagae the output amplifier nor the resistor, but I need somebody to confirm this.

No need think and guess. It have been known all time after Siglent have launched this SDG.

1.
It reads here:

Waveform Output: Impedance 50?(typical)
Protection short-circuit protection.

Same information can find also in SDG1000 service manual specifications.



2.
It reads also here: 

Yes, you can drive what ever SDG800/SDG1000/SDG5000 signal from its output  to what ever load impedance, including zero impedance (short circuit).

What kind of confirmation you need?

[ivan747]

I just couldn't remember the name of the output amplifier chips,so that I could look up the datasheet, but yours is enough confirmation.

[rf-loop]

I just couldn't remember the name of the output amplifier chips,so that I could look up the datasheet, but yours is enough confirmation.

Output amplifiers:

CH1
Ti THS3095

CH2
Intersil EL5166ISZ

Both outputs have 49.9ohm series resisor after amplifier.
Simplified
CH1  Amp out -- (2x49R9 parallel) -- (2x49R9 parallel) -- BNC

CH2  Amp out -- 49R9 -- BNC

[ivan747]

That's it! Thank you very much.

[mmrow]

Hello,
Could anyone verify for me the following issue?
Setting duty to value different than 50% does not affect the shape of the square wave.
... or it is just my lack of knowledge ... 
rgs,
M

SDG1005, firmware 37R2.