GW Instek AFG-2225 vs Rigol DG4102

[mahwe]

http://www.batronix.com/versand/funktionsgeneratoren/Rigol-DG4102.html
http://www.datatec.de/GWinstek-Arb-Generator-AFG-2225.htm
http://www.gwinstek.com/en/product/productdetail.aspx?pid=5&mid=305&id=1286

Hi i need a new arbitrary waveform generator.
I would prefer an Agilent - 33512B but their is the price i dont like.
Is their anybody who tested oder use the AFG-2225.
I know Dave test with the rigol .
   thanking you in anticipation
Marc

[jpb]

I'm in the same boat - looking to buy an arbitrary function generator, trying to decide between a new Rigol or similar or an old/used Agilent or Tabor or an ex-demo TTi TG5011.

I did look at the GW Instek models but the 300 series are very expensive and most of the 200 series are rather basic with low sampling rates.

The 225 is probably the best as it has 125MS/s sampling. BUT

it only has 4k points in arb and it is only 10bits.
Also flatness is 10% at 25MHz and distortion figures and accuracy aren't brilliant.

To be fair though, Rigol give their figures at 0dBm (0.6Vpp) whilst the GWInstek figures might be over the full range up to 10Vpp - the Rigol figures could be a lot worse over this range,
I just don't know.

This is the main problem I have, it is just so difficult to compare specs as most manufacturers give their specs under different conditions and for different ranges of frequency and at
different signal amplitudes - established suppliers tend to cover a reasonable range whilst the newer players (like Rigol) just give figures at the best point (only 0dBm when they are
not using the output amplifier at all).

The new Siglent 5000 series looks quite interesting so it might be worth hanging on to see what that turns out to be like in practice.

[grego]

Instek is going to be sending me a demo unit to poke at so if you can wait a few weeks I can put together a video for you.

Please note that the 2225 is significantly less than the 4102 -- so a little "you get what you pay for" needs to be kept in mind.  They aren't apple-to-apple.

[mahwe]

i think i can wait
thank you very much

[ACvolts]

Instek is going to be sending me a demo unit to poke at so if you can wait a few weeks I can put together a video for you.

Please note that the 2225 is significantly less than the 4102 -- so a little "you get what you pay for" needs to be kept in mind.  They aren't apple-to-apple.

Did you get the unit yet?  My problem with the AFG-2225 is the unknowns with the Arbitrary PC Software!
I really would like to have a DC waveform loadable from the PC software and allow me to increase/decrease
the Frequency.  But I'm still not sure.  This is what I got from Instek Support:
--Begin Quote--
Our AFG-2225, the DC in arbitrary mode is really not included, but you can
use the drawing line function of the arbitrary function.  You can set the
point and axis-Y value of start and stop. For example, setting the (1,512)
and (100,512) it becomes a DC wave.Hope this helps.
--End Quote--

So I am not sure if I can load a DC pc soft. waveform in the ARB section of the AFG-2225.
But the DG4000 series from Rigol does and a whole lot more for $300 bucks!
All I can say is:
Brother can you spare a Demo Unit for $100.00 bucks? 

ACvolts

[JuanPC]

http://www.batronix.com/versand/funktionsgeneratoren/Rigol-DG4102.html
http://www.datatec.de/GWinstek-Arb-Generator-AFG-2225.htm
http://www.gwinstek.com/en/product/productdetail.aspx?pid=5&mid=305&id=1286

Hi i need a new arbitrary waveform generator.
I would prefer an Agilent - 33512B but their is the price i dont like.
Is their anybody who tested oder use the AFG-2225.
I know Dave test with the rigol .
   thanking you in anticipation
Marc

Rigol DG4162 is $1300usd. http://www.tequipment.net/RigolDG4162.html
Rigol DG4102 is $1Kusd. http://www.tequipment.net/RigolDG4102.html
Rigol DG4062 is $800usd. http://www.tequipment.net/RigolDG4062.html

Instek afg-2225 is $549usd. without the INSTEK15OFF coupon.
http://www.testequipmentdepot.com/instek/signalgenerators/afg-2225.htm

Agilent 33509B is $1653usd.
http://www.testequity.com/products/4586/

ARB upgrade for 1-ch models $300usd.
335ARB1U Add arb to 1-channel models.

335BW1U 30 MHz Upgrade for 1-ch models $200usd.

34190A 2U Rackmount Kit half-rack $63usd.
34191A 2U Dual Rack Mount Flange Kit $60usd.

Opt. MEM (335MEM1U) Additional 16M Memory $250usd.
Opt. OCX Ultra-high Stability Timebase $701usd. factory return needed,.


if Price is more important go for Instek.
if Accuracy is more Important go for Agilent.
if Memory is more Important go for Agilent.
if Jitter is more Important go for Agilent.
if testing low voltage circuits, needs 16-Bit, go for Agilent.
Rigol has 14-Bit, Instek has 10-Bit.
if need more than 30Mhz go for Rigol. Agilent is limited to 30Mhz, True Bandwidth. to see 25Mhz SQR without errors, a 500Mhz Scope is needed, 1GHz preferably.
True Bandwidth is determined by RiseTimes, a faster rise time means true Higher Bandwidth,
for example Agilent has maximum 30Mhz, but a Rise Time of 8.4ns,
but Rigol DG4162 160Mhz also has a <8ns RiseTime, usually a true 150Mhz scope has <2.3ns RiseTimes, at 100MHz has <3.5ns, 350MHz 1ns, 500MHz 700ps, or 0.7ns, that means to see a 40ps Jitter is impossible with a 500MHz scope.
Rigol DG4162 is <8ns, DG4102: <10ns & DG4062: <12ns, 
Instek is <25ns,

The only interesting difference is the Sample Rate, Agilent is 250MSa/s, Rigol is 500MSa/s, Instek is 120MSa/s.
would be nice to see with a 1Ghz scope, how does it look a 30Mhz SQR coming from the Agilent 33500B vs. the Rigol DG4162.

Allmost all DDS clocks have 700ps jitter rms, Agilent TrueForm "Modified DDS" has less than <40ps p-p.
Higher Frequency needs less Jitter & Fast RiseTimes to work properly.

[ACvolts]

I bought a Instek AFG-2225 and am reviewing it in this thread:

https://www.eevblog.com/forum/testgear/afg-2225-function-generator-testing/msg270226/#msg270226

There is no question that the Rigol DG4000's AFG's are better than the
Instek AFG-2225 or lower models.  The Instek AFG-3051/AFG-3081 are
over $1000 dollars but you could compare them with the DG4000's more
in functionality.  The DG4062 is a 60 Mhz Func. Gen. and outclasses the
AFG-2225 completely in my view since I own one.

AFG-2225 - 25Mhz AFG and affordable in it's class ($500 dollar price)
DG-4062 - 60Mhz AFG better functions with competitive price ($800)

I would consider look very carefully at specs because you may not need
the Rigols if all your using is low end electronics.

[Carrington]

http://www.batronix.com/versand/funktionsgeneratoren/Rigol-DG4102.html
http://www.datatec.de/GWinstek-Arb-Generator-AFG-2225.htm
http://www.gwinstek.com/en/product/productdetail.aspx?pid=5&mid=305&id=1286

Hi i need a new arbitrary waveform generator.
I would prefer an Agilent - 33512B but their is the price i dont like.
Is their anybody who tested oder use the AFG-2225.
I know Dave test with the rigol .
   thanking you in anticipation
Marc

Rigol DG4162 is $1300usd. http://www.tequipment.net/RigolDG4162.html
Rigol DG4102 is $1Kusd. http://www.tequipment.net/RigolDG4102.html
Rigol DG4062 is $800usd. http://www.tequipment.net/RigolDG4062.html

Instek afg-2225 is $549usd. without the INSTEK15OFF coupon.
http://www.testequipmentdepot.com/instek/signalgenerators/afg-2225.htm

Agilent 33509B is $1653usd.
http://www.testequity.com/products/4586/

ARB upgrade for 1-ch models $300usd.
335ARB1U Add arb to 1-channel models.

335BW1U 30 MHz Upgrade for 1-ch models $200usd.

34190A 2U Rackmount Kit half-rack $63usd.
34191A 2U Dual Rack Mount Flange Kit $60usd.

Opt. MEM (335MEM1U) Additional 16M Memory $250usd.
Opt. OCX Ultra-high Stability Timebase $701usd. factory return needed,.


if Price is more important go for Instek.
if Accuracy is more Important go for Agilent.
if Memory is more Important go for Agilent.
if Jitter is more Important go for Agilent.
if testing low voltage circuits, needs 16-Bit, go for Agilent.
Rigol has 14-Bit, Instek has 10-Bit.
if need more than 30Mhz go for Rigol. Agilent is limited to 30Mhz, True Bandwidth. to see 25Mhz SQR without errors, a 500Mhz Scope is needed, 1GHz preferably.
True Bandwidth is determined by RiseTimes, a faster rise time means true Higher Bandwidth,
for example Agilent has maximum 30Mhz, but a Rise Time of 8.4ns,
but Rigol DG4162 160Mhz also has a <8ns RiseTime, usually a true 150Mhz scope has <2.3ns RiseTimes, at 100MHz has <3.5ns, 350MHz 1ns, 500MHz 700ps, or 0.7ns, that means to see a 40ps Jitter is impossible with a 500MHz scope.
Rigol DG4162 is <8ns, DG4102: <10ns & DG4062: <12ns, 
Instek is <25ns,

The only interesting difference is the Sample Rate, Agilent is 250MSa/s, Rigol is 500MSa/s, Instek is 120MSa/s.
would be nice to see with a 1Ghz scope, how does it look a 30Mhz SQR coming from the Agilent 33500B vs. the Rigol DG4162.

Allmost all DDS clocks have 700ps jitter rms, Agilent TrueForm "Modified DDS" has less than <40ps p-p.
Higher Frequency needs less Jitter & Fast RiseTimes to work properly.

Wow, more clearly, is impossible!

[grego]

I would also urge folks to look at the Siglent SDG5082. You can get it for under $700usd shipped from China on eBay and it has specs to beat the Rigol.  See rf-loops thread about it. I have one as well and it's a very nice device.

[Electro Fan]

http://www.batronix.com/versand/funktionsgeneratoren/Rigol-DG4102.html
http://www.datatec.de/GWinstek-Arb-Generator-AFG-2225.htm
http://www.gwinstek.com/en/product/productdetail.aspx?pid=5&mid=305&id=1286

Hi i need a new arbitrary waveform generator.
I would prefer an Agilent - 33512B but their is the price i dont like.
Is their anybody who tested oder use the AFG-2225.
I know Dave test with the rigol .
   thanking you in anticipation
Marc

Rigol DG4162 is $1300usd. http://www.tequipment.net/RigolDG4162.html
Rigol DG4102 is $1Kusd. http://www.tequipment.net/RigolDG4102.html
Rigol DG4062 is $800usd. http://www.tequipment.net/RigolDG4062.html

Instek afg-2225 is $549usd. without the INSTEK15OFF coupon.
http://www.testequipmentdepot.com/instek/signalgenerators/afg-2225.htm

Agilent 33509B is $1653usd.
http://www.testequity.com/products/4586/

ARB upgrade for 1-ch models $300usd.
335ARB1U Add arb to 1-channel models.

335BW1U 30 MHz Upgrade for 1-ch models $200usd.

34190A 2U Rackmount Kit half-rack $63usd.
34191A 2U Dual Rack Mount Flange Kit $60usd.

Opt. MEM (335MEM1U) Additional 16M Memory $250usd.
Opt. OCX Ultra-high Stability Timebase $701usd. factory return needed,.


if Price is more important go for Instek.
if Accuracy is more Important go for Agilent.
if Memory is more Important go for Agilent.
if Jitter is more Important go for Agilent.
if testing low voltage circuits, needs 16-Bit, go for Agilent.
Rigol has 14-Bit, Instek has 10-Bit.
if need more than 30Mhz go for Rigol. Agilent is limited to 30Mhz, True Bandwidth. to see 25Mhz SQR without errors, a 500Mhz Scope is needed, 1GHz preferably.
True Bandwidth is determined by RiseTimes, a faster rise time means true Higher Bandwidth,
for example Agilent has maximum 30Mhz, but a Rise Time of 8.4ns,
but Rigol DG4162 160Mhz also has a <8ns RiseTime, usually a true 150Mhz scope has <2.3ns RiseTimes, at 100MHz has <3.5ns, 350MHz 1ns, 500MHz 700ps, or 0.7ns, that means to see a 40ps Jitter is impossible with a 500MHz scope.
Rigol DG4162 is <8ns, DG4102: <10ns & DG4062: <12ns, 
Instek is <25ns,

The only interesting difference is the Sample Rate, Agilent is 250MSa/s, Rigol is 500MSa/s, Instek is 120MSa/s.
would be nice to see with a 1Ghz scope, how does it look a 30Mhz SQR coming from the Agilent 33500B vs. the Rigol DG4162.

Allmost all DDS clocks have 700ps jitter rms, Agilent TrueForm "Modified DDS" has less than <40ps p-p.
Higher Frequency needs less Jitter & Fast RiseTimes to work properly.

Juan, great post - very nice compare and contrast.  EF

(Here is another video on the 33500B Series:


Looks like a deluxe product with very cool features - especially if you have the budget for 2 channels and the Arb.

[jpb]

True Bandwidth is determined by RiseTimes, a faster rise time means true Higher Bandwidth,
for example Agilent has maximum 30Mhz, but a Rise Time of 8.4ns,
but Rigol DG4162 160Mhz also has a <8ns RiseTime, usually a true 150Mhz scope has <2.3ns RiseTimes, at 100MHz has <3.5ns, 350MHz 1ns, 500MHz 700ps, or 0.7ns, that means to see a 40ps Jitter is impossible with a 500MHz scope.

Whether or not you can see 40psecs of jitter on a 500MHz scope I would have thought was dependent on the trigger jitter and timebase accuracy rather than the scope rise time. For example, my scope is only 350MHz BW with 1nsec rise time but in equivalent time sampling mode it goes up to 100GS/s which implies that it can place time points within 10 psecs. For a 500MHz scope the fastest timebase is 500ps/div and each division is 50 pixels so each pixel is 10psecs. 40psecs of jitter would mean that the line would be thickened to around 4 pixels which should be visible. It wouldn't be a very accurate measurement of course!

[JuanPC]

True Bandwidth is determined by RiseTimes, a faster rise time means true Higher Bandwidth,
for example Agilent has maximum 30Mhz, but a Rise Time of 8.4ns,
but Rigol DG4162 160Mhz also has a <8ns RiseTime, usually a true 150Mhz scope has <2.3ns RiseTimes, at 100MHz has <3.5ns, 350MHz 1ns, 500MHz 700ps, or 0.7ns, that means to see a 40ps Jitter is impossible with a 500MHz scope.

Whether or not you can see 40psecs of jitter on a 500MHz scope I would have thought was dependent on the trigger jitter and timebase accuracy rather than the scope rise time. For example, my scope is only 350MHz BW with 1nsec rise time but in equivalent time sampling mode it goes up to 100GS/s which implies that it can place time points within 10 psecs. For a 500MHz scope the fastest timebase is 500ps/div and each division is 50 pixels so each pixel is 10psecs. 40psecs of jitter would mean that the line would be thickened to around 4 pixels which should be visible. It wouldn't be a very accurate measurement of course!

It's imposible to explain all in 3 lines, but Yes, in DSO the Sampling also does affect, 5GSa/s RealTime has 200ps per division, if designed properly.
40ps It's also imposible to see with a 5GSa/s RT scope.
Equivalent Sampling mode is like Image interpolation....
If i have a 320x240 pixel image and upsample to 4k, is not the same as if taken at 4k, or more and downsample.
Equivalent Sampling is good if the Real Time is good.
But also wfm/s affect.
100GS/s EquivalentSampling is useless if RealTime is 500MS/s or less, for jitter measurements.

I had a 1GS/s RT 100Mhz scope with 25GS/s ES and to see 700ps from my soundcard s/pdif was imposiblle, infinite persistance and all.
The funny thing is, if loudspeakers are good enough most people can hear 700ps very easy when compared to another more stable clock,.