Author Topic: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator  (Read 190627 times)

0 Members and 1 Guest are viewing this topic.

Offline qq354813374

  • Contributor
  • !
  • Posts: 15
  • Country: england
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #50 on: July 20, 2017, 04:09:09 am »
Sine wave is not 14bit, waveform data only 2048 points, the resolution is only 11bit, the waveform is very rough
 

Offline pantelei4

  • Regular Contributor
  • *
  • Posts: 93
  • Country: ua
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #51 on: July 20, 2017, 04:14:51 am »
 

Online Messtechniker

  • Frequent Contributor
  • **
  • Posts: 774
  • Country: de
  • Old analog audio hand - No voodoo.
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #52 on: July 20, 2017, 08:07:08 am »
Does the JDS6600 carry a CE mark?

I'm asking, because without it German customs may
have me return the unit or scrap it at my cost which is not funny.

Yes, I know the CE story. It basically means China Export :-DD
Agilent 34465A, Siglent SDG 2042X, Hameg HMO1022, R&S HMC 8043, Peaktech 2025A, Voltcraft VC 940, M-Audio Audiophile 192, R&S Psophometer UPGR, 3 Transistor Testers, DL4JAL Transistor Curve Tracer, UT622E LCR meter
 

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 14072
  • Country: de
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #53 on: July 20, 2017, 03:08:15 pm »
Having 11 Bit resolution is already quite good from an simple R2R chain at the FPGA. For me it looks surprisingly good.
Medium grade DDS chips like the AD9850 only use an 10 Bit ADC, though they use a longer table (e.g. 4096 points). With just 2048 points for a period, there is limited use of more than 10 Bit resolution though.

The staircase like waveform with quite some ringing suggests that there is no good reconstruction filter. So one can expect quite some high frequency images. The fast output amplifier also has some disadvantages here.

Some jitter in the square wave is normal for directly generating the square from the FPGA.
 
The following users thanked this post: RD Tech

Offline qq354813374

  • Contributor
  • !
  • Posts: 15
  • Country: england
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #54 on: July 21, 2017, 12:40:49 am »
R2R resistor network to achieve 11bit resolution is indeed very good, I do not know if you have not observed 14bit signal generator, the waveform is much better than this good, the use of R2R resistor network there is a drawback, that is, noise Problem, do not know if you have not found a small signal noise is very large, because the digital signal is grounded and the analog signal ground will not be separated, so the noise in the digital signal will interfere with the analog signal. I bought the JD6600 signal generator when the seller introduced a resolution of 14bit, but the actual waveform is only 11bit, I think I was cheated, and this business with the integrity and honesty have a great relationship. |O |O |O |O
 

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 14072
  • Country: de
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #55 on: July 21, 2017, 07:47:02 am »
Form the picture it looks like the DAC is not a pure R2R chain, but more like using thermometer style for the upper bits. So maybe someone confuses the 14 I/O pins used for control with a 14 bit resolution. At least the DAC seems to work reasonably well as 11 Bit this way.

The DAC type does not have direct influence on the noise with small signals. The waveform should be generated at constant amplitude and only the following multiplier and attenuators reduce the amplitude if needed.
 
The following users thanked this post: RD Tech

Offline qq354813374

  • Contributor
  • !
  • Posts: 15
  • Country: england
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #56 on: July 21, 2017, 08:05:08 am »
R2R resistor network to do out of the signal generator and DAC out of the signal generator to produce the waveform is completely different, if the same, the production of DAC manufacturers do not need to exist,
Directly with a few resistance can produce a DAC, digital signals and analog signals if not isolated, the noise can not be eliminated, there is, FPGA I / O port drive capacity is limited, not every I / O The current generated by the mouth will be exactly the same, which is the reason for the instability of the waveform, 11bit and 14bit waveform generated by a great difference, I do not know if you have not carefully observed, 14 I / O port can not say that the resolution of 14bit , He also has a large number of data stored with the waveform has a great relationship.
 

Offline qq354813374

  • Contributor
  • !
  • Posts: 15
  • Country: england
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #57 on: July 21, 2017, 08:16:07 am »
Form the picture it looks like the DAC is not a pure R2R chain, but more like using thermometer style for the upper bits. So maybe someone confuses the 14 I/O pins used for control with a 14 bit resolution. At least the DAC seems to work reasonably well as 11 Bit this way.

The DAC type does not have direct influence on the noise with small signals. The waveform should be generated at constant amplitude and only the following multiplier and attenuators reduce the amplitude if needed.

You can look at the delicate degree of 14bit waveform
 

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 14072
  • Country: de
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #58 on: July 21, 2017, 08:59:05 am »
A special DAC chip is generally better than using just a resistor chain at the FPGA output. Most DACs do not offer that much of isolation between the digital side and analog output. So noise is in both cases a question of a good layout. This is still a low cost generator - so the cheap resistor ladder DAC is a kind of compromise.

The failure is to label it 14 Bit resolution if they actually only get 11 Bits.

The Resistor chain looks like 8 Bit R2R and 14 or 15 resistors thermometer style. So at best they could get something like 12 Bit resolution. Still I am surprised how good it seemed to work. If not careful, even at 8 Bits R2R one can get significant errors. One sees that with other low cost generators that call for 8 or 12 Bit resolution and only deliver 6-7.  With an actual 11 Bit resolution it is already a big step forward.

Using true 14 or maybe just 12 Bit DACs is a cost factor - though they might save on the amplitude control this way.
 
The following users thanked this post: RD Tech

Offline pantelei4

  • Regular Contributor
  • *
  • Posts: 93
  • Country: ua
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #59 on: July 21, 2017, 01:57:35 pm »
You can look at the delicate degree of 14bit waveform
Do you have a FY6600?
It can amplitude and frequency modulation and has a sync output?
 

Offline qq354813374

  • Contributor
  • !
  • Posts: 15
  • Country: england
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #60 on: July 22, 2017, 02:17:59 am »
You can look at the delicate degree of 14bit waveform
Do you have a FY6600?
It can amplitude and frequency modulation and has a sync output?

I have one, the parameters look better than the JDS6600 much better, I found a user manual and parameter description, you can look at
https://drive.google.com/file/d/0B3N_NW8vylrmWFZQWkRnWDI5cms/view?usp=sharing
 

Offline pantelei4

  • Regular Contributor
  • *
  • Posts: 93
  • Country: ua
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #61 on: July 22, 2017, 12:21:32 pm »
I have one
When the frequency is tuned, the phase of the synchronized 2 channel does not change for a short time?
This is important if at 1 and 2 output an antiphase pulse signal with a deadtime, used to control the inverter.
 

Offline qq354813374

  • Contributor
  • !
  • Posts: 15
  • Country: england
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #62 on: July 22, 2017, 12:36:23 pm »
I have one
When the frequency is tuned, the phase of the synchronized 2 channel does not change for a short time?
This is important if at 1 and 2 output an antiphase pulse signal with a deadtime, used to control the inverter.

The phase of the adjustment frequency CH2 changes because it needs to synchronize the CH1 signal, but the adjustment amplitude and the bias level do not change
 

Offline pantelei4

  • Regular Contributor
  • *
  • Posts: 93
  • Country: ua
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #63 on: July 22, 2017, 12:49:19 pm »
The phase of the adjustment frequency CH2 changes because it needs to synchronize the CH1 signal
MHS5200A phase does not change, at least the deadtime does not intersect.
The only generator that made me for this reason.
« Last Edit: July 22, 2017, 12:51:34 pm by pantelei4 »
 

Offline qq354813374

  • Contributor
  • !
  • Posts: 15
  • Country: england
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #64 on: July 22, 2017, 01:28:09 pm »
The phase of the adjustment frequency CH2 changes because it needs to synchronize the CH1 signal
MHS5200A phase does not change, at least the deadtime does not intersect.
The only generator that made me for this reason.
I have just tested my instrument to adjust the frequency of the CH1 channel when the CH2 phase does not change, maybe I have an error in his understanding
 

Offline gby

  • Regular Contributor
  • *
  • Posts: 238
  • Country: us
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #65 on: July 22, 2017, 01:42:56 pm »
The Feeltech FY6600 has so many differences to the JDS6600:
   "Real" 14 bit DAC vs. discrete resistor driven by FPGA pins
    8192 point arbitrary table at 250 Mhz vs. 2048 points at 200 MHz
    Analog input for modulation source of frequency, amplitude, duty cycle vs. none (unfortunately limited to 1000 Hz)
    Digital modulation mode: ASK, FSK, PSK from external input vs. none

With these differences and more I suggest that the Feeltech FY6600 generator deserves its own thread for discussion.  It would be great to see some pictures of the insides and see some performance testing like I asked Vk2seb to do on the JDS6600.  See https://www.eevblog.com/forum/testgear/anybody-know-anything-about-this-signal-generator/msg1240865/#msg1240865

Hopefully an early FY6600 owner can get this started.
 

Offline pantelei4

  • Regular Contributor
  • *
  • Posts: 93
  • Country: ua
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #66 on: July 22, 2017, 02:30:02 pm »
I have just tested my instrument to adjust the frequency of the CH1 channel when the CH2 phase does not change, maybe I have an error in his understanding
A successful and unsuccessful frequency tuning, the pulses of channels 1 and 2 should not overlap.

 

Offline Diabolo

  • Regular Contributor
  • *
  • Posts: 124
  • Country: fr
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #67 on: July 23, 2017, 12:00:57 am »
Hello,

At the request of @gby

Here is the new Feeltech DDS generator 14 bits-250 msa-8192 points without R2R equipped with an Altera Cyclone IV clocked with SPXO 50mhz .
The model I bought is FY2300H-25 Mhz.


regards,
Diabolo
https://www.eevblog.com/forum/testgear/new-generateur-dds-avec-altera-cyclone4/
« Last Edit: July 23, 2017, 12:18:10 am by Diabolo »
 

Offline colorado.rob

  • Frequent Contributor
  • **
  • Posts: 419
  • Country: us
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #68 on: July 23, 2017, 01:54:20 am »
Here is the new Feeltech DDS generator 14 bits-250 msa-8192 points without R2R equipped with an Altera Cyclone IV clocked with SPXO 50mhz .
The model I bought is FY2300H-25 Mhz.

What are the missing ICs for?
 

Offline Diabolo

  • Regular Contributor
  • *
  • Posts: 124
  • Country: fr
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #69 on: July 23, 2017, 03:20:19 am »
Hello,

For the absent CIs I do not know.
They may be for a more powerful version, or with little interest for this version.
The generator works very well and the signals are good.
Without advertising, I am very satisfied with this device.

Regards,
Diabolo
 

Offline qq354813374

  • Contributor
  • !
  • Posts: 15
  • Country: england
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #70 on: July 24, 2017, 02:04:05 am »
Hello,

At the request of @gby

Here is the new Feeltech DDS generator 14 bits-250 msa-8192 points without R2R equipped with an Altera Cyclone IV clocked with SPXO 50mhz .
The model I bought is FY2300H-25 Mhz.


regards,
Diabolo
https://www.eevblog.com/forum/testgear/new-generateur-dds-avec-altera-cyclone4/



Well-known brands of signal generators are all using DAC, which is R2R resistor network can not go beyond
 

Offline technogeeky

  • Frequent Contributor
  • **
  • Posts: 555
  • Country: us
  • Older New "New Player" Player Playa'
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #71 on: July 24, 2017, 05:23:45 am »
Form the picture it looks like the DAC is not a pure R2R chain, but more like using thermometer style for the upper bits. So maybe someone confuses the 14 I/O pins used for control with a 14 bit resolution. At least the DAC seems to work reasonably well as 11 Bit this way.

The DAC type does not have direct influence on the noise with small signals. The waveform should be generated at constant amplitude and only the following multiplier and attenuators reduce the amplitude if needed.

You can look at the delicate degree of 14bit waveform

I'm sorry to be rude, but I am a bit concerned about your posts in this thread. I think you are trying to imply from the context and content of your posts that you are a mere consumer who has tried the Rui-Deng model and a FeelTech model function generator. Your profile says you're from England.

However:
  • you list "feeltech" in your profile
  • your username is typical of the email addresses offered primarily to Chinese users by Tencent
  • a google search for that username reveals other posts in Chinese in various technical forums
  • your English, while adequate, leads me to believe you speak English as a second language
  • FeelTech is headquartered in China, and I'm not able to find any offices in the UK

I should note that all of this is just from information you offered (your username, listing feeltech in your profile) or otherwise publicly available information. And I don't know of any reason or rule that one's profile must be correct on this forum, so I doubt you've done anything wrong. I suppose I could believe that you are an Chinese expat living in the UK.

In any case, if you do work for Feeltech (and perhaps if you designed the product you are discussing) I  think you should be up front and honest about that fact. After all, who else would be better qualified to analyze the properties of a modern cheap function generator than a person who just designed a modern cheap function generator?

Thanks.
 
The following users thanked this post: CustomEngineerer

Offline pantelei4

  • Regular Contributor
  • *
  • Posts: 93
  • Country: ua
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #72 on: July 24, 2017, 06:37:54 am »
The JDS6600 on the main signals does not have one clock jitter, if the duty cycle is 50%, but it is 5ns if the duty cycle is not 50%
« Last Edit: July 24, 2017, 06:40:26 am by pantelei4 »
 

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 14072
  • Country: de
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #73 on: July 24, 2017, 08:53:29 am »
The jitter should not depend on the H/L ratio. It should be just a question of the 5 ns time raster. So it depends on the frequency too. This might even be the worst aspect of it. At frequencies that don't have a period that is an integer multiple of 5 ns will show the jitter. So more like low jitter only at a few special good frequencies.

Depending on how frequency calibration is done the good frequencies can be slightly different, depending on the exact quartz frequency. In principle they could use a different frequency scale - allowing only good ones (e.g. divider) and maybe warn about poor ones.
 
The following users thanked this post: RD Tech

Offline RD Tech

  • Regular Contributor
  • *
  • !
  • Posts: 89
  • Country: cn
Re: RD JDS6600 25MHz 2-Channel DDS AW Function Signal Generator
« Reply #74 on: July 24, 2017, 09:31:44 am »
JDS6600 Phase does not sync
Yes :horse:
I have admite and explain , we have improve this , this don't affect normall using ...
about this little problem, we are updating this.
« Last Edit: July 24, 2017, 09:54:08 am by RD Tech »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf