General > General Technical Chat
Review: Hantek DDS 3X25. Anyone own one?
<< < (8/108) > >>
saturation:
Hi dajones:

I'd glad you brought that up.

You are correct, the sync output is adequate to about 1MHz, then its heavily distorted, and the signal is increasingly useless; I can analyze stable waveforms with various combo of the Rigol's trigger options to about 3 MHz but its easier to just trigger on the channel signal, and the output is very stable to 75 MHz on sine at -12dB and ~ 25 MHz most everything else, at -3dB output.

For others interested, I did not test the pattern generator and digital output.  The input trigger works as expected.  Hantek allows you to adjust the phase angle of the output 0-360 degrees and there is a DC voltage offset.  Note, the output is LVCMOS, and swings only 3.5Vpp.  What I've done in the past is use a non-inverting opamp at the output stage with sufficient gain-bandwidth product for the needed frequency, you can optionally use the phase angle to adjust for any shifts caused by the amp.

I weighed all the pros and cons, and for me the pros win out.  But your need may differ, particularly if you need the sync signal output, its not good.  I'll add scope grabs when I've time.  

I gave it good marks because I don't think you can find a variable sine wave source for 75 MHz for $150, or even Seilig's $100, much less the typical functions of an analog, not DDS, function generator for that price.  The output is very stable, but the jitter is a problem in certain designs requiring strict duty cycles.

Most $150-200 FG output ~ 3MHz, albeit they may have clean signals through the bands, analog FG suffer from frequency variations, between 0.1-1.0% / hr even when already at operating temps.  Of top analog FG brands, many are more stable, but far most costly, and no where in ppm stability.  Likewise, the harmonic content is very variable too.

http://www.eevblog.com/2009/04/24/eevblog-5-instek-function-generator-review/

The DDS variants change the field for FG, but the starting price for >3 MHz is at least 2-3x the cost of the Hantek.

If you have a good FG to recommend with stable clean outputs, link them here.  One recent post was a surprise, the Velleman.  But again, its under 3 MHz:

https://www.eevblog.com/forum/index.php?topic=1869.msg31157#msg31157

Its one reason I posted my raw screens, so you can decide for yourselves its value, its the equivalent of a revised spec sheet.

I'll post my scopes images of the issues you raised later on.




--- Quote from: dajones on January 20, 2011, 09:30:49 pm ---
--- Quote ---The Hantek DDS 3x25 is a solid function generator with arbitrary waveform capability, far better in performance than most analog function generators, in its price range or with similar specifications.  It one of the lowest cost 25 MHz capable DDS function generator on the market, and likely the only one with AWG capability.
--- End quote ---

Man, I don't understand how you can say that!

The thing glitches like crazy for square waves above 2.5mhz ( except a few stable spots like 10mhz ).

The sync out line has no timing relationship to the output signal except at those rare "stable" spots.

The sync out line glitches every 20us ( 50khz ) and will glitch any output waveform other than sine waves.

I could NEVER recommend this to anyone, except maybe as a source of sine waves, and as long as you
didn't need to use the sync out line at all.

If you did your stability test at one of the "stable" spots ( like 10mhz ) you might want to try it again at 10.1mhz.




--- End quote ---
Mechatrommer:

--- Quote from: Zyvek on January 10, 2011, 04:58:01 am ---Just noticed this tonight (no affiliation) on clearance for $99USD:
http://www.saelig.com/CL/CL001.htm
Not sure if this means they are discontinuing them...

--- End quote ---
its discontinued, not sold anymore. maybe i was thinking too long, gotta find another source :( or any info on the newer product?

this thread is perused:
1) sine wave: 1.5Vrms?@3MHz, 0.2Vrms?@75MHz? is it enough/good as an external clock for an mcu? or for mcu work?
2) arbitrary good to 100kHz only? or can be more? like 1-3MHz is possible?
3) what is "sync output" for?

my fingers are tinggling, should i hit the $150 3x25?
saturation:
Hi Safri,

Its at least sold out by Seilig, you still have many eBay sellers and other vendors in Asia, there are many out there I see on google, find one close to you in Malaysia.

I wouldn't be surprised if the item is discontinued by Hantek, it might be why its so cheap; it was designed with bugs in it that isn't worth repairing or recalling; that's why I think its a steal, the parts of it that work well provide what I reviewed here.

To answer your questions:

1. 3.5Vpp is ~ 2.5Vac rms.  -12 dB for 2.5Vac rms = 0.63Vac rms.  I'm not so sure -12dB is the Hantek roll off or my Rigol's.  I'll recheck and post a correction if I can confirm its source.

The clock output is stable to 3MHz.  As you go faster, that jitter will eventually cause inconsistent clocking, but when that happens depends on your circuit.

It is possibly good enough for non-LV CMOS circuits, because the maximum output of 3.5Vdc can be made by adjusting the voltage offset and is above the switching level of TTL and CMOS.

2. Arbitrary depends on how bizarre your waveform, I can get many stable waveforms over 3 MHz, and easily under 100kHz, but its best to monitor the output on your scope as you ramp the frequency up. That's because waveforms beyond just sine requires more frequency response from the hardware.

3.  

Sync output sends a logic level output that's the same frequency as the output signal.  

You can use it as an alternate trigger signal for a scope, if none of the scope's options work.
You can synchronize multiple devices to output simultaneously.  For example, you can cascade multiple signal sources together than have trigger Input and Sync output, taking the sync output of one device to the trigger input of another, and daisy chain multiple devices, to simulate multiple simultaneous synchronized inputs to your design.  This is not as straightforward as it sounds, because as you daisy chain these devices, there is always some trigger input delay so you need to manually adjust the phase shift on each device as the sync signal propagates down the line of signal sources.

Because the sync out of the 3x25 is unreliable, you can't daisy chain 3x25s, but you can trigger it. Also, if you need the sync signal for any other reason, it doesn't work past 1MHz.




--- Quote from: shafri on January 25, 2011, 02:00:42 am ---
--- Quote from: Zyvek on January 10, 2011, 04:58:01 am ---Just noticed this tonight (no affiliation) on clearance for $99USD:
http://www.saelig.com/CL/CL001.htm
Not sure if this means they are discontinuing them...

--- End quote ---
its discontinued, not sold anymore. maybe i was thinking too long, gotta find another source :( or any info on the newer product?

this thread is perused:
1) sine wave: 1.5Vrms?@3MHz, 0.2Vrms?@75MHz? is it enough/good as an external clock for an mcu? or for mcu work?
2) arbitrary good to 100kHz only? or can be more? like 1-3MHz is possible?
3) what is "sync output" for?

my fingers are tinggling, should i hit the $150 3x25?


--- End quote ---
Mechatrommer:
yes its still available in ebay. and the bugs, dajones tried to highlight that clearly. but as you said, and i think, a 75MHz sine is a hard thing to find at the price, the DDS-3005 cost more than twice and no indication it can do 75MHz sine. as for other signal than sine (square, arb), i will treat them as extra features of the generator, so slower frequency should be acceptable for me, in fact imo 3MHz is pretty good already. but the 3x25's -12db concerns me, as your pictures indicate, its around 384mVpp @ 75MHz sine. so i'm rounding it to roughly 0.2Vrms, i was quoting Vdc (Vac/2 to my understanding, sorry if i'm wrong), but whatever! your pictures worth 1K words.

my only concern if its possible feeding the signal into the mcu clock and get the mcu MIPS'ping, you know what i mean? running. as jitter is concerned, perharps that related to how accurate the timing is? so far i dont think i will need a very accurate timing/clocking project, if i do maybe i should get more stable oscillator for the project. i just want to know what happen if the mcu is running at its highest speed (say if a PIC can run at max 40MHz, then i can tune the hantek to 40MHz sine and feed it to the PIC), thats all that i can imagine so far. or is it do i need an extra amplifying circuitry before fedding to mcu? i dont know, maybe i'm too newbie on this and have to make some study. you mentioned about "voltage offset" which is not very clear to me, but dont bother, i'll go figure that out. so the answer to my concern should be just as simple as "yes" and "no".

other than that, i would like to experiment with LCR meter as you've suggested in another thread, and other possibility within the reach of the generator. pretty much thats all. i'm not that complicated beast ;) . OR just maybe this whole thing is just a delusion to me?

ps:
comparison between 3005 and 3x25 (html)
3005 manual (pdf)
3x25 manual (pdf)
saturation:
Yes, it is.  You are correct, its down ~ 384mV at 75MHz, but its still clean, harmonic free as I can measure  ;D

You can easily amplify it for more gain using a op amp rated for that speed, made as a non-inverting amplifier, the chips are ~ $5.  For such high frequencies use dead bug style wide ground plane construction.  You can then control the output as you did before but with much higher output.

http://www.analog.com/en/amplifiers-and-comparators/operational-amplifiers-op-amps/ad8011/products/product.html

Yes, the Hantek 3005 cannot do 75 MHz, very few function generators can.  At 20 MHz, they run in the $400, of course, unlike the 3x25, all outputs work  :D  Even second hand eBay RF sources cost more and not as stable.

I do not know if driving an MCU clock with the 3x25 will create issues because of the jitter, jitter becomes a problem the higher the clock speed is, so it depends how high you ramp up the clock speed.

If I need a good clean output, I use the Instek 1003, its the reason I bought it, and then the 3x35 is mostly as a signal source.

I do intend to use the sine wave as my highest output clock source, as you suggest.  A simple way to convert sine to square is simply clip the sine wave peaks:



Or use a Schmitt trigger, and adjust the threshold trigger voltage to vary the duty cycle .  This design just reuses the op amp you already have, so get several if you intend to do something like this [this example uses the 741 op amp, but op amps in general use the same design layout].




--- Quote from: shafri on January 25, 2011, 06:55:58 pm ---yes its still available in ebay. and the bugs, dajones tried to highlight that clearly. but as you said, and i think, a 75MHz sine is a hard thing to find at the price, the DDS-3005 cost more than twice and no indication it can do 75MHz sine. as for other signal than sine (square, arb), i will treat them as extra features of the generator, so slower frequency should be acceptable for me, in fact imo 3MHz is pretty good already. but the 3x25's -12db concerns me, as your pictures indicate, its around 384mVpp @ 75MHz sine. so i'm rounding it to roughly 0.2Vrms, i was quoting Vdc (Vac/2 to my understanding, sorry if i'm wrong), but whatever! your pictures worth 1K words.

my only concern if its possible feeding the signal into the mcu clock and get the mcu MIPS'ping, you know what i mean? running. as jitter is concerned, perharps that related to how accurate the timing is? so far i dont think i will need a very accurate timing/clocking project, if i do maybe i should get more stable oscillator for the project. i just want to know what happen if the mcu is running at its highest speed (say if a PIC can run at max 40MHz, then i can tune the hantek to 40MHz sine and feed it to the PIC), thats all that i can imagine so far. or is it do i need an extra amplifying circuitry before fedding to mcu? i dont know, maybe i'm too newbie on this and have to make some study. you mentioned about "voltage offset" which is not very clear to me, but dont bother, i'll go figure that out. so the answer to my concern should be just as simple as "yes" and "no".

other than that, i would like to experiment with LCR meter as you've suggested in another thread, and other possibility within the reach of the generator. pretty much thats all. i'm not that complicated beast ;) . OR just maybe this whole thing is just a delusion to me?

ps:
comparison between 3005 and 3x25 (html)
3005 manual (pdf)
3x25 manual (pdf)


--- End quote ---
Navigation
Message Index
Next page
Previous page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod