Author Topic: SDG2042X Timebase Calibration  (Read 751 times)

0 Members and 1 Guest are viewing this topic.

Offline mawyatt

  • Frequent Contributor
  • **
  • Posts: 405
  • Country: us
SDG2042X Timebase Calibration
« on: October 04, 2020, 06:19:27 pm »
I just calibrated the SDG2042X AWG 10MHz reference to the SSA3021X Plus 10MHz reference and found a simple way to quickly get very close initial calibration.

Instead of following the counter in the Timebase Calibrate section exactly, connect the external 10MHz reference to the counter input as required but also to a DSO input (High Z) C1. Now set the AWG to 10.000000MHz and connect to the other DSO input C2, trigger off either input and set DSO sweep time to 20ns/div.

Carefully adjust the DAC code in the Timebase Calibrate section until the two waveform are somewhat synched. Try and set the DAC so the random walk & drift between the waveforms has equal but opposite excursions, thus averaging out over a long time period.

I just did this and running the SDG2042X counter is showing a mean frequency error of ~0.002ppm while counting the SSA3021X 10MHz reference over the past ~ hour. Seems the SDG2042X Reference Oscillator is pretty good, of course much longer monitoring period is required :)

Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike
 
The following users thanked this post: Johnny B Good

Online Johnny B Good

  • Frequent Contributor
  • **
  • Posts: 446
  • Country: gb
Re: SDG2042X Timebase Calibration
« Reply #1 on: October 28, 2020, 05:14:35 pm »
 I came to the same solution when I was checking out the clock calibrate function a few days ago. The only difference being that I used my much modified FY6600-60M to provide the 10MHz reference (its internal 10MHz OCXO which drives a clock multiplier to create the 50MHz FPGA clock was frequency injection locked to my GPSDO reference).

 I agree with you on this method offering a more precise and speedy way to calibrate its internal clock reference as closely as is possible when said internal clock is merely a VCXO rather than as others have assumed, a VCTCXO. To my mind, there seems to be far too much warm up drift and instability for it to be a Temperature Compensated VCXO. Am I the only one who thinks this the case?

===========================================================================================
[EDIT 2020-10-28]

 The datasheet +/-2ppm 0 to 40 deg C specification suggests just a VCXO rather than a VCTCXO and there's no mention of a VCTCXO being installed which confirms my impression that it's just a VCXO rather than the VCTCXO that everyone else seems to have assumed.
===========================================================================================


 I used a cheap 50MHz 0.1ppm TCXO board for my first clock oscillator upgrade before moving onto an OCXO because the best calibration and stability I could wring out of that TCXO (with calibration trimmer) had been a mere 30ppb which the SDG2042X's internal 10MHz clock oscillator comes nowhere near to achieving.

 Of course, this difference may simply be down to my foresight in keeping it well away from the 60 deg hot spot the original ten cent smd XO had been placed in. I'd located it immediately above a small cooling fan I'd fitted into the base of the FY6600's case to not only act as an airflow deflector but also to make sure it never ran any hotter than 2 or 3 deg above room temperature.

 The 0.1ppm figure applied only to the 0 to 25 deg range, degrading to 0.2ppm in the 25 to 50 deg range so I wanted to keep it out of that higher temperature range despite the fact that most other cheap TCXOs claim an even worse degradation to 0.4 or 0.5 ppm at higher temperatures.

 Of course, I may just be observing the less than stellar performance typical of an smd VCTCXO when it's subjected to the higher and more variable temperatures when it's mounted directly onto a main circuit board carrying the heat load from the components which which comprise the bulk of a typical AWG. >:D [Belay that thought - it's only a VCXO.]

 Perhaps I've simply raised my expectations beyond what could reasonably be expected when the circuit layout has been compromised to keep manufacturing cost to a minimum. [Although that still remains true, in this case it's only a VCXO]  Luckily in this case, this isn't a deal breaker on account of the presence of an external clock reference option socket.

 Talking of which (external/internal reference clock control), it seems to me that it lacks a third option in its simple choice of "import external clock" and "export internal clock", namely, "use internal clock without exporting it to pollute a perfectly good GPSDO reference feed". Since, despite the sub standard internal clock reference, it can still be useful in the absence of a proper external reference for synchronising other test kit, having a separate option to enable or disable the export function in the original binary choice would be the most optimal way to control this rather than have it as a third option to toggle through. If you do have a proper external clock distribution, disable it, otherwise enable it.

 If the use of a high impedance input on such external clock reference input sockets is common practice (presumably to allow daisy-chaining the feed to other kit with a 50 ohm dummy load terminator at the far end of the line if the final bit of kit doesn't already have a term/thru switch as my FY6600 has), the current options mean you need to disconnect it from the line before switching back to the internal clock if you wish to avoid disruption to the rest of the kit relying upon an actual reference. Unfortunately, as things stand, that means you get an error message to inform you of the loss of the external clock reference and a disabling of signal output.

 However, that's just one small deficit out of a much larger deficit in the whole front panel UI. If Siglent ever get round to employing a proper UI developer to create a user friendly UI version, he'll find no shortage of problems to address. >:(

John
« Last Edit: October 28, 2020, 08:10:19 pm by Johnny B Good »
 

Offline Performa01

  • Frequent Contributor
  • **
  • Posts: 917
  • Country: at
Re: SDG2042X Timebase Calibration
« Reply #2 on: October 29, 2020, 01:15:31 pm »
===========================================================================================
[EDIT 2020-10-28]

 The datasheet +/-2ppm 0 to 40 deg C specification suggests just a VCXO rather than a VCTCXO and there's no mention of a VCTCXO being installed which confirms my impression that it's just a VCXO rather than the VCTCXO that everyone else seems to have assumed.
===========================================================================================

So everyone else is wrong then?
Maybe everyone else have had bad luck when they tried to get a simple VCXO with better than +/-20 ppm stability...
 

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 3068
  • Country: hr
Re: SDG2042X Timebase Calibration
« Reply #3 on: October 29, 2020, 01:56:13 pm »
===========================================================================================
[EDIT 2020-10-28]

 The datasheet +/-2ppm 0 to 40 deg C specification suggests just a VCXO rather than a VCTCXO and there's no mention of a VCTCXO being installed which confirms my impression that it's just a VCXO rather than the VCTCXO that everyone else seems to have assumed.
===========================================================================================

So everyone else is wrong then?
Maybe everyone else have had bad luck when they tried to get a simple VCXO with better than +/-20 ppm stability...

I agree.

I would like to see source of non temp compensated VCXO that will have tempco of 100ppb per °K. And cheap too.

Of course it is VCTCXO.... Maybe not best or most expensive one, but VCTCXO nevertheless.

And AWG is not meant to be used as frequency reference. If absolute frequency is needed, you should get proper frequency reference and ref in devices when needed.

 
The following users thanked this post: rf-loop, Performa01

Online Johnny B Good

  • Frequent Contributor
  • **
  • Posts: 446
  • Country: gb
Re: SDG2042X Timebase Calibration
« Reply #4 on: November 02, 2020, 04:34:05 am »
===========================================================================================
[EDIT 2020-10-28]

 The datasheet +/-2ppm 0 to 40 deg C specification suggests just a VCXO rather than a VCTCXO and there's no mention of a VCTCXO being installed which confirms my impression that it's just a VCXO rather than the VCTCXO that everyone else seems to have assumed.
===========================================================================================

So everyone else is wrong then?
Maybe everyone else have had bad luck when they tried to get a simple VCXO with better than +/-20 ppm stability...

I agree.

I would like to see source of non temp compensated VCXO that will have tempco of 100ppb per °K. And cheap too.

Of course it is VCTCXO.... Maybe not best or most expensive one, but VCTCXO nevertheless.

And AWG is not meant to be used as frequency reference. If absolute frequency is needed, you should get proper frequency reference and ref in devices when needed.

 Those responses from yourself and Performa01 are rather curious to say the least. For starters, where in the specification is there any mention of a VCTCXO? I've checked out both datasheet versions (the 2015 one and the current 2020 one) and can not see mention of a VCTCXO anywhere. I see mention by the marketing copy writers of a +/-1ppm figure for the initial accuracy (+/-5K of 297K) and a +/-2ppm figure over the range 273 to 313K followed by the spurious 1st year ageing of +/-1ppm and 10 year ageing of +/-3.5ppm figures - spurious on account of the DAC calibration facility to recalibrate such ageing effect completely out of the picture. Those copy writers sure were clutching at every straw they could find to pad out the specs with as many "facts and figures" as they could lay their hands on (as all competent marketing bullshitters are obliged to do to justify their salaries).

 Considering that the specifications have been carefully crafted to make them appear more impressive than they really are (RMS jitter values anyone?), it seems very unlikely that they would have overlooked an opportunity to boast of a VCTCXO, (not just once but twice!) when you consider the 700 quid price tag on a return to factory TCXO upgrade on the more upmarket Keysight AWGs (incidentally, over nine times what I'd paid for my FY6600-60M almost two years ago ::)).

 Other than positively identifying the XO chip actually soldered to the main board as being a VCTCXO type, I cannot see how anyone could conclude that the XO chip is anything more than a decent quality VCXO type. However, I did run some extensive comparison tests between my cheap 50MHz 0.1ppm XO clock module board and the output of the Siglent's internal clock reference which, rather unconscionably, is output onto the 10MHz reference socket by default when not running from an external reference clock.

 I set up my SDG2104X to trigger from the GPSDO reference on CH1 with a timebase setting of 10ns/div and used CH3 to display the 50MHz TCXO and CH4 to display the SDG2042X's 10MHz reference. After running several tests involving a small sheet of paper to temporarily block the fan intake to raise the internal temperature a few degrees in the SDG and a small usb powered desk fan to provide a modest cooling boost and using the same fan to cool the 50MHz board and one of those "Dust Grabbing" cloths as an insulating cover to similarly vary the 50MHz TCXO temperature, I powered the SDG and the 50MHz TCXO down for about half an hour or so to ready them for a start from stone cold warm up test.

 With everything ready including a stopwatch in hand, I fired up both signal sources and monitored the 50MHz TCXO's behaviour whilst waiting for the SDG to finish its 22 seconds boot up sequence and finally restore the DAC calibration voltage to its VCXO. The 50MHz starts off around 50ppb low for the first split second,  swiftly increasing frequency to within 4ppb after ten seconds or so to finally drift bang on frequency some 20 minutes later when, purely coincidentally, the Siglent's internal reference had also landed bang on frequency as it continued its upward drift.

 The Siglent had jumped to a frequency some 250ppb low once the DAC calibration voltage had been restored and then slowly drifted exactly onto frequency at the same moment the TXCO module had as mentioned above. from then on, the Siglent carried on slowly drifting upward in frequency, eventually to settle some 76ppb high an hour or two later, with the TCXO settling at just 3.7ppb low.

 These results simply confirmed my initial impression that the XO in the Siglent is simply just a VCXO that could be recalibrated through the UI to place it exactly on frequency and so allow it to vary through the whole of its specified +/-2ppm temperature range without violating the SDG2042X's frequency stability specification.

 I think this is just another demonstration of the power of marketing in manipulating the perceptions of potential buyers so that they infer some desirable feature that simply doesn't exist. In this case, the unusually tight frequency tolerance spec of the VCXO within +/-5 deg of the optimum 23 deg C operating environment temperature range and the slightly looser +/-2 deg spec over the much wider 0 to 40 deg C range suggesting to those more susceptible to such psychological warfare tactics (which has, annoyingly, included myself in times past  :palm:) that the VCXO in question could only be of the VCTCXO type.

 You only need a few mentions of the phrase "VCTCXO" in an SDG2000X topic thread and, before you know it, it has become "Common Knowledge" that Siglent are using such an XO chip in their SDG2000X product line. I'f I'm wrong in this matter, I'll humbly apologise but I rather doubt in all honesty, that I'll be needing to say sorry to anyone. The warm up frequency drift is completely contrary to what you'd see with a TCXO of any type.

 Incidentally, whilst searching the list of features for any mention of a VCTCXO, I noticed the rather irksome claim of it having a high precision frequency counter function built in (don't they all? ::)). At just 8 digits resolution, that might have rung true 2 or 3 decades ago but even the counter in my 'toy' Feeltech AWG can produce an 8 digit readout on its 1 second gate time setting which can be pushed to a 10 digit resolution if you have the patience to wait out the full 100 seconds gate time.

 Since I was somewhat curious about what it would make of the Siglent's "10MHz" clock, I waited it out to see a reading of 10,000,000.68Hz eventually materialise which had me re-timing ten passes of the drifting clock signal I'd previously measured at 13.2 seconds to calculate the 0.75757Hz offset I'd based my +76ppb figure upon. I was pleased to see a figure of 0.6896551Hz which closely matched the readout which alternates the last two digits between 68 and 69 (there's a 1 digit uncertainty in the final digit of this counter whatever the chosen gate time - in that regard, it's a little "Old Skool" in its behaviour).

 This is yet another example of marketing bullshit since, presumably to save immediate embarrassment at making such a claim, they neglected to mention just how many (or few) digits of resolution this "high precision" counter actually has at its disposal. As a prospective buyer (especially so for the hobbyist or small business buyer) it's important to not let any enthusiasm for the brand blindside you into failing to look your chosen Gift Horse very closely in the mouth and examine any unsubstantiated claims, implied or otherwise, very carefully before committing to a purchase you could later regret ever making.

 I've attached three screenshots to show the waveforms. The first two are with a 1 second persist to give some idea of the relative phase shift velocity of the 50MHz TCXO and the Siglent's 10MHz clock. The third is a reference 'snapshot' without any persistence.

John
 

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 3068
  • Country: hr
Re: SDG2042X Timebase Calibration
« Reply #5 on: November 02, 2020, 08:21:28 am »
@Johny B Good,

I'm not trying to fight you. We merely pointed out that AT-cut crystal has 20 ppm temp stability, and I frankly don't know if there is one that has better than 10 ppm temp stability to be bought anywhere.
That's physics.  Any crystal oscillator that has better than that (including 2 ppm temp stability for full temp range, no matter how pedestrian that spec might look to you) has to have some sort of temperature compensation (analog or digital) to achieve even that spec. And that makes it TCXO. Fact that it can pull frequency with outside control voltage makes it VC, so that makes it VCTCXO, right there.
Making oscillator without temperature compensation that would achieve 2 ppm stability over full temp range would need a crystal that has better than +-10ppm stability (unheated) and those don't exist...
Even if you could run some characterisation process and pick some unicorn crystals from standard AT-cut that would be that good, those would be very expensive.

If you want to be datasheet nazi, and scrutinize all the stuff into detail, that's fine by me. But by Occam's razor, whether they published part number or not, what do you think, what is more possible: that they invented new type of crystal or process to make crystals that nobody else can, and made all that so cheap that is was worth it, or they went and bought first cheapest VCTCXO (that are sub USD priced) and threw that one in...

Regards,
Sinisa
« Last Edit: November 02, 2020, 08:29:13 am by 2N3055 »
 
The following users thanked this post: Johnny B Good

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3380
  • Country: cn
  • Born with DLL21 in hand
Re: SDG2042X Timebase Calibration
« Reply #6 on: November 02, 2020, 08:39:09 am »
Of course it is TCXO without any doubt.

But then overall about xtals.  And this is only part of whole oscillator.

Xtal itself temp drift... It depends...

If we make AT cut XTAL what is designed for owen it is very different than if it is designed for "room" temp or temp inside some instrument. It can optimize for quite small temp drift ...



For more accurate average frequency there is connector for external freq reference what is much more important than 0.01 or 5ppm accuracy in internal reference. Why. Because in most work where need high accuracy also all instruments need be sync as usually is case in all serious lab. Usually there is one house standard shared to all instruments.

But back to XTALs. If look this green 0' ...  (of course this kind of cut can not use in owen at all. cut need select for owen temperature so that in this point freq change is minimal, example 6' - 8')



« Last Edit: November 02, 2020, 09:58:04 am by rf-loop »
If practice and theory is not equal it tells that used application of theory is wrong or the theory itself is wrong.
-
Harmony OS
 
The following users thanked this post: Performa01, Johnny B Good

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 3068
  • Country: hr
Re: SDG2042X Timebase Calibration
« Reply #7 on: November 02, 2020, 10:02:00 am »
It depends...

If we make AT cut XTAL what is designed for owen it is very different than if it is designed for "room" temp or temp inside some instrument. It can optimize for quite small temp drift ...



For more accurate average frequency there is connector for external freq reference what is much more important than 0.01 or 5ppm accuracy in internal reference. Why. Because in most work where need high accuracy also all instruments need be sync as usually is case in all serious lab. Usually there is one house standard shared to all instruments.

But back to XTALs. If look this green 0' ...  (of course this kind of cut can not use in own at all. cut need select for owen temperature so that in this point freq change is minimal)





Hi!

I was talking exclusively about NON HEATED non OCXO crystals and oscillators.  In those, achieving +- 2ppm over full temperature range is pretty hard if you don't apply at least some first order compensation in electronics... Which, technically, makes it a TCXO.. ^-^

But if you simply slap a crystal on a Colpitts  oscillator, 2ppm temp stability over full temp range will be hard to reach..

Even if you cleverly match capacitor tempco to slightly compensate for crystal tempco, that's technically TCXO...

As soon as you have varactor in the circuit to have voltage control, you make a compensation network with some NTC and resistors, and voila that's VCTCXO....

I think this is more of a terminology misunderstanding, than anything else..

Regards,

Sinisa
 
 
The following users thanked this post: Johnny B Good

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3380
  • Country: cn
  • Born with DLL21 in hand
Re: SDG2042X Timebase Calibration
« Reply #8 on: November 02, 2020, 11:19:36 am »
It depends...

If we make AT cut XTAL what is designed for owen it is very different than if it is designed for "room" temp or temp inside some instrument. It can optimize for quite small temp drift ...



For more accurate average frequency there is connector for external freq reference what is much more important than 0.01 or 5ppm accuracy in internal reference. Why. Because in most work where need high accuracy also all instruments need be sync as usually is case in all serious lab. Usually there is one house standard shared to all instruments.

But back to XTALs. If look this green 0' ...  (of course this kind of cut can not use in own at all. cut need select for owen temperature so that in this point freq change is minimal)





Hi!

I was talking exclusively about NON HEATED non OCXO crystals and oscillators.  In those, achieving +- 2ppm over full temperature range is pretty hard if you don't apply at least some first order compensation in electronics... Which, technically, makes it a TCXO.. ^-^

But if you simply slap a crystal on a Colpitts  oscillator, 2ppm temp stability over full temp range will be hard to reach..

Even if you cleverly match capacitor tempco to slightly compensate for crystal tempco, that's technically TCXO...

As soon as you have varactor in the circuit to have voltage control, you make a compensation network with some NTC and resistors, and voila that's VCTCXO....

I think this is more of a terminology misunderstanding, than anything else..

Regards,

Sinisa

Fully agree.

I think it happen that when you write your comment I have also edited and added my comment partially for more accurate but also avoid misunderstood - example that of course SDG have TCXO without any single doubt.
I think I fully understand what are oscillators, how to compensate these and what are different XTALS and oscillators. Tens of years ago I have also ordered manufacturer to make XTAL with my own specifications when I design and make some instrument for one special purpose.

Of course also you talk about "whole temperature range" in some cases it can be -55 - +80 celsius and some cases +18 - +26 celsius... it depends...
Quite easy XTAL can stay inside +- 2ppm in 60 celsius wide temp range. If XTAL have made and selected for this. But as previously told XTAL is not oscillator. 
Some times if we find some XTAL, resistor, active component and some capacitors carefully also including these tempco even with these can make quite stabile oscillator and yes this is TCXO. Sometimes need add some more components for find enough nice compensation for acceptable result. And all these are TCXO's of course.
In some cases we can measure temperature and logic can select right compensation.

Good example is inside RTC DS3231N where is quite simple 32kHz XTAL but it can keep  ±2ppm from 0°C to +40°C and  ±3.5ppm from -40°C to +85°C.  (M version do not have TCXO oscillator at all. It have MEMS oscillator.)

Sidenote. Chinese sellers do not understand or care what is difference between DS3231  S/N  versions and M versions and then perhaps some them sell out of specs chips or bad copycats.
Inside there is TXCO what have array of capacitors and temperature measurement (and bit more things) and depending temp and programmed curve, if need,  it switch capacitor aka step adjustabe capacitor, itf I remember in 64 second interval.


Simplest ones are these analogically compensated ADTCXO's. (R, C components tempco, NTC etc solutions)
Other types are DTCXO, MCXO and DCXO.


« Last Edit: November 02, 2020, 11:26:20 am by rf-loop »
If practice and theory is not equal it tells that used application of theory is wrong or the theory itself is wrong.
-
Harmony OS
 
The following users thanked this post: Performa01, Johnny B Good

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 3068
  • Country: hr
Re: SDG2042X Timebase Calibration
« Reply #9 on: November 02, 2020, 12:22:54 pm »
@rf-loop

Well said. You might get a crystal that will be tight enough but rest of the circuit has to temp compensated..
But ordering crystal with such tight specs will be expensive. VCTCXO are cheap and apart from cheapest crap most have amazing specs for the price.

As for timekeeping, i had my share of work on precise clocks for motorsports (rallying and hillclimb is popular here in my hometown) previously, some years ago..
Spend lot of time devising clever compensation techniques to achieve some pretty tight timing requirements comparable to Tag Heuer Sport timers at the time... Today, you would simply slap a nice DCTCXO, tune it up, and put GPS module for control and calibration...
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3380
  • Country: cn
  • Born with DLL21 in hand
Re: SDG2042X Timebase Calibration
« Reply #10 on: November 02, 2020, 02:00:09 pm »
@rf-loop

Well said. You might get a crystal that will be tight enough but rest of the circuit has to temp compensated..
But ordering crystal with such tight specs will be expensive. VCTCXO are cheap and apart from cheapest crap most have amazing specs for the price.

As for timekeeping, i had my share of work on precise clocks for motorsports (rallying and hillclimb is popular here in my hometown) previously, some years ago..
Spend lot of time devising clever compensation techniques to achieve some pretty tight timing requirements comparable to Tag Heuer Sport timers at the time... Today, you would simply slap a nice DCTCXO, tune it up, and put GPS module for control and calibration...

Yes.

Btw, bit challenging to keep xtal tightly in freq due to G and vehicle angle changing and including even pressure change...  as we know from time nuts, all is changing, everything and all is affecting, everything... all drifting, all are off... no one know what is perfect time or freq. But I have stopped this all long time ago because it is waste of time. My time is proportional to my age... more age...more fast time goes so where I need precise time or precise frequency... what do not even exist, just as there is not real DC, also it exist only in theory books.

World with exactly one clock or what ever time reference have perfect time with zero error.

If practice and theory is not equal it tells that used application of theory is wrong or the theory itself is wrong.
-
Harmony OS
 
The following users thanked this post: 2N3055

Offline mawyatt

  • Frequent Contributor
  • **
  • Posts: 405
  • Country: us
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike
 

Online Johnny B Good

  • Frequent Contributor
  • **
  • Posts: 446
  • Country: gb
Re: SDG2042X Timebase Calibration
« Reply #12 on: November 02, 2020, 05:17:14 pm »
@rf-loop and @2n3055,

 Thank you for your very considered responses. Our differences of opinion over what exactly that VCXO/VCTCXO actually is within the SDG2000X model range seems to be over the terminology used.

 I think the VCXO used by Siglent in this case would be better described as a "Low TempCo" VCXO rather than a VCTCXO with a best temperature stability of 0.1ppm (limited by the retrace effect), typically applied over the 0 to 25 deg C range versus a Low TempCo VCXO optimised to stay within the wider but still impressively good 1 to 2ppm limits achieved by a wise choice of crystal cut and components to minimise the overall XO temperature drift. There's no specific temperature compensating circuit involved, just a combination of components that passively minimise the overall tempco effect.

 I'm sure that if the copy writers had decided they could legally describe such a low tempco VCXO as a VCTCXO, they would have done so (they're not in the habit of missing such golden opportunities to legally use such higher value descriptions in their feature lists or specifications whenever possible).

 The fact that they'd pointedly avoided the description "VCTCXO", relying only upon "Letting the figures speak for themselves", suggests that my interpretation is the more reasonable one in this case. I'm not trying to "score points" here, just trying to set the record straight since a VCTCXO would typically offer an order of magnitude improvement in temperature stability over the low tempco version of VCXO Siglent have elected to use here.

 It might seem that I'm "nit picking" over the terminology but since Siglent clearly understand the distinction between a low tempco VCXO and a VCTCXO by avoiding any possibility of being accused of making false claims by sticking only to the actual facts and figures of the frequency stability parameters, I felt it was a point worth making over the unjustified use of the description "VCTCXO" when not even Siglent themselves were prepared to use such a description.

 I have no issues with Siglent's careful precision in their description of the frequency stability of the internal clock oscillator. Indeed, they're to be lauded for their honesty, whilst at the same time, admired for the effectiveness of their copywriter's art in persuading prospective buyers into reading more into the description than actually exists, i.e. the presence of a VCTCXO when only a high quality low tempco VCXO is involved. :)

 In this case, Siglent have quite reasonably included an external clock reference socket to render the distinction between a low tempco VCXO and a VCTCXO rather academic. Although the "Export the internal clock" option when operating from the internal clock can be a handy way to synchronise other test kit in the absence of an actual 'Lab reference', it's just a shame that this export feature couldn't be toggled on and off in a separate menu option to avoid polluting an otherwise perfectly fine daisy-chained lab reference distribution setup.

 Far better to have provided this internal clock export function over a separate dedicated BNC socket but that adds to the BoM costs, hence this second best compromise of using a single BNC socket for both functions. It's just rather unfortunate that the user's options choice is so poorly implemented (as is the case for so many other aspects of the front panel UI which remains so unconscionably far from being a mature interface considering that they've had over four years in which to correct all the shortcomings which make it such a pain to use - notably the frequency parameter setting UI).

 You might question why I even bothered to buy a Siglent AWG in the first place but in spite of this, this still remained the best affordable option and the issue is essentially just one major firmware revamp away from being fully resolved. The only worrying aspect being that Siglent have been sitting on their hands in fixing the blatant deficiencies in the UI for the past four years or so.

 I'm an optimist and since the fix is do-able as a firmware update, I still have hopes that this will ultimately be resolved, especially in view of their record with other of their T&M model ranges regarding timely responses to bug fix requests and user wish lists. I'm just a little puzzled at the AWG division's apparent "Poor Man's" status in regard of UI developer talent.

 Apropos of bug fix requests, I've not been able to find any topic threads relating specifically to the SDG2000X series as exists for the SDS2000X DSO related topic thread so I'm wondering whether anyone here knows of such a thread or am I simply going to have to start one of my own? :-\

John
 
The following users thanked this post: 2N3055

Offline mawyatt

  • Frequent Contributor
  • **
  • Posts: 405
  • Country: us
Re: SDG2042X Timebase Calibration
« Reply #13 on: November 02, 2020, 06:06:13 pm »
According to Vectron's (well respected OEM, used their oscillators in many applications) App note, see Figure 1, a +-2ppm XO per Siglent's data sheet for the SDG2042X is in the TCXO category.

https://www.vectron.com/products/literature_library/tutorial_on_tcxos.pdf
[attach=1]

From Siglent data sheet.
[attach=2]

That's good enough for my understanding of the boundary between a XO, TCXO and OCXO. And since the Siglent data sheet shows a 0~40C temp range, which implies (assumes linear change) a TC of +-2ppm/40C or +-0.05ppm/C which places it well within the TCXO category shows this actually a quite good TXCO indeed ;)

I can say that the stability relative to the 10MHz reference in the SSA3021X Plus which is the best reference I have available is outstanding :-+
[attach=3]

Best,
« Last Edit: November 02, 2020, 06:10:36 pm by mawyatt »
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike
 
The following users thanked this post: rf-loop, Performa01, 2N3055, Johnny B Good

Online Johnny B Good

  • Frequent Contributor
  • **
  • Posts: 446
  • Country: gb
Re: SDG2042X Timebase Calibration
« Reply #14 on: November 03, 2020, 01:27:38 am »
According to Vectron's (well respected OEM, used their oscillators in many applications) App note, see Figure 1, a +-2ppm XO per Siglent's data sheet for the SDG2042X is in the TCXO category.

https://www.vectron.com/products/literature_library/tutorial_on_tcxos.pdf
[attach=1]

From Siglent data sheet.
[attach=2]

That's good enough for my understanding of the boundary between a XO, TCXO and OCXO. And since the Siglent data sheet shows a 0~40C temp range, which implies (assumes linear change) a TC of +-2ppm/40C or +-0.05ppm/C which places it well within the TCXO category shows this actually a quite good TXCO indeed ;)

I can say that the stability relative to the 10MHz reference in the SSA3021X Plus which is the best reference I have available is outstanding :-+
[attach=3]

Best,

 Thanks for providing a link to the Vectron app note. Page 3 provides a more detailed look at the temperature compensation curves typically applied in TCXOs. As you can see, you cannot safely assume a linear change of frequency over its total temperature range (at least in the case of a TCXO - it might be a less 'lumpy' curve for a low tempco VCXO as used in the Siglent SDG2000X series but still not the linear change you've assumed).

 In the case of the 50MHz TCXO I'd used as a reference to compare the Siglent's 10MHz internal clock reference to, I've always assumed its given 0.1ppm stability over the 0 to 25 deg C range to be a maximum deviation from its calibrated frequency (+/-0.1ppm).

 In any case, it had never occurred to me to calculate an equivalent frequency shift with temperature in terms of ppb/K since I knew there was little to be gained in knowing this figure. It was what it was and all I could do was to do my best to avoid operating outside of this specified 0 to 25 deg C range as much as I could possibly arrange and also try to arrange that it never suffered any more change of temperature than those mediated by changes in room temperature, hence my choice of mounting location within the FY6600-60M, right above the small cooling fan to act as both a deflector and benefit from the incoming room temperature air flow to hold it within two degrees above ambient (typically within a degree or two of the 25 deg upper limit of its 0.1ppm temperature range specification.

 If I had bothered to calculate a per deg frequency change figure to get a sense of the stability involved, I would have arrived at a 4ppb/K for a simplified linear curve and perhaps multiplied it by 2.5 to approximate the sinuous curve that's typically applied to realise such a tight temperature stability over a 25 K range.

 The biggest issue I'd had with that clock module in the FY6600 was that of the backlash in its trimcap, making it virtually impossible to calibrate it any closer than 30ppb of frequency other than by dumb luck. In the end, when I'd gotten hold of a bunch of 10MHz OCXOs for cheap, I decided to replace it with one of these OCXOs and a 3n502 clock multiplier to save any further frustration.

 That OCXO upgrade had given me a whole new insight into the value of precise frequency measurement which has since allowed to observe phenomena that had previously been masked by the previous lax standards of "Good Enough" typical of AWGs ranging from Feeltech cheapies right through to expensive Keysight and not so pricey Siglent AWGs.

 Funnily enough, now that I have finally built myself a modern day version of the James Miller GPSDO design to give me a stable GPS based frequency reference, re-calibrating that 50MHz TCXO for this latest testing session, though still a pain went a lot smoother and, in fact, some 4 hours after powering up yesterday's test setup late this afternoon, it's now settled to within -200ppt of 50MHz (10mHz low) and I'm just about to cool it down with that USB desk fan... within a few seconds it had increased by 4ppb in frequency before going back down to within 300ppt short of where it had been until I shut the fan off when it slowly resumed its drop in frequency before reversing back to around 1ppb high courtesy of the retrace issue which rather spoils the TC game plan which brings us back to that chart on page 1.

 I'm surprised that Vectron should even think there's any overlap between high end TCXOs and low end OCXOs, especially when they show a gap between VCXOs and low end TCXOs. To my mind, there's no justification for having a gap between TCXOs and OCXOs any less than an order of magnitude of temperature stability rating.

 I mean, why would anyone go to all the additional power requirement costs of a sub standard OCXO when a less power hungry TCXO of higher frequency stability with temperature could do the job without all that expense and additional heat loading? It makes no sense that such an overlap should even exist.

 Perhaps I'm overlooking the possibility of cheaper AT cut crystal based OCXOs operating at a lower oven temperature (~60 deg C?) with lower peak power demand being a cheaper alternative to the highest specced TCXO designs as a possible reason for this overlap?

 A 0.1ppm over the 0 to 25 deg C range is pretty close to the limit before retrace starts making a complete mockery of any such improved temperature compensation attempts. If you're going to make all that extra investment required to support an OCXO, you'll certainly be looking to get better than just one magnitude of improvement with more likely a two magnitudes or better improvement to justify such an expense.

 If you've no choice but to bite the bullet on account not even the best available TCXOs can meet your stability requirement,  I should imagine it will be a case of, "In for a penny, in for a pound" and "You might as well be hanged for a sheep as for a lamb." and all that.

 In other words, you'd be more inclined to go for a 1ppb rated OCXO than for a barely adequate 10ppb rated one unless your company's chief bean counter feels he can get away with doing ten dollar's worth of damage to the company's reputation for the sake of a 50 cent saving on the BoM costs without risk of being given his marching orders.

 Am I right in thinking that Siglent datasheet image relates to the SSA3021X Plus? It looks almost, but not quite, identical to the frequency characteristics section in the SDG2000X's datasheet. Unfortunately I can't easily check images whilst in the 'reply' window but I seem to recall a 20 year ageing figure versus the 10 year ageing figure in the SDG2000X's frequency specifications.

 If that's currently your best frequency standard then be prepared to "Have your socks blown off" by a thousandfold improvement should you ever invest in a decent 10MHz GPSDO lab standard to plug into the Ext Ref input socket. :)

John
 

Offline mawyatt

  • Frequent Contributor
  • **
  • Posts: 405
  • Country: us
Re: SDG2042X Timebase Calibration
« Reply #15 on: November 03, 2020, 02:23:13 am »
The images are 1st from Vectron, 2nd SDG2042X and 3rd SSA3021X Plus.

Some time ago I worked over 20 years for the company mentioned in the Chip-Scale Atomic Clock post, so been around precision time and navigation systems & standards a long time  ;)

I don't need anything better than what I have regarding an accurate and stable timebase now, but agree the GPS enhanced standard is very interesting and enticing ::)

Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike
 
The following users thanked this post: Johnny B Good


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf