General > General Technical Chat
Review: Hantek DDS 3X25. Anyone own one?
Bored@Work:
--- Quote from: Aurora on May 09, 2013, 09:50:22 am ---It is a great pity that the software lets Hantek products down. The hardware looks to have been well thought through, and produced to a good standard. They need to find a new software engineer.
--- End quote ---
They first need to find new management.
Management that first of all acknowledges they have a relevant problem. Then comes up with a feasible plan to fix the problem. That plan should, among others, include hiring competent software developer. Then obtaining funding for the plan. Then executing the plan, correcting and adjusting it while they go.
Bored@Work:
You have no understanding how companies work.
marabut:
--- Quote from: marmad on May 08, 2013, 10:43:24 pm ---What is so strange is that they failed to write PC software which correctly implements some of the features of the original DDS-3X25 firmware; for example, 'burst' mode for doing single-cycle waveforms and short bursts. This is something I use all the time (via my own software) - but what did they do? Instead of re-writing the software to implement these things correctly, they just eliminate or bury features. [...]
--- End quote ---
Hi all!
I plan to buy 1025G soon, so I'm trying to get as much information as possible about (potential and real) flaws.
marmad
Could you make clear what you've had in mind, please?
What exactly features of 3x25 firmware aren't implemented in 1025G? This is important for me as 1025G is ~$65 cheaper than 3x25 at local distributor's store (yes, I know it's strange :o).
However if some of features was eliminated in 1025G, it can be better to get older but well known product. I used to think about 1025G as repackaged 3x25.
Is it true at functional level or not?
torch:
The external hardware differences are reasonably clear.
1. the 3x25 has a 5v external power connector, or can be powered by USB. The 1025g is USB only.
2. the 3x25 has 5 BNC connections: OUTPUT, SYNC OUT, COUNT IN, CONTROL, TRIG IN. The 1025g has only 2 BNC connections: OUTPUT, SYNC OUT.
3. The 3x25 has 6 digital inputs, I 0 through I5. The 1025g has only 4 digital inputs, I 0 through I3 (of the 20 pin connector). The advertising (and specifications page in the manual) claims the 1025g has 6 bit input, but the manual clearly shows it's only 4 bit.
4. The 3x25 also has 12 digital outputs, O 0 through O11. The 1025g has 12 digital outputs, O 0 through OB (of the 20 pin connector).
5. The 3x25 has two ground pins on the 20 pin connector, the 1025g has 1 ground pin.
6. If you have been doing the math thus far, that makes a total of 20 pins accounted for on the 3x25, but only 17 on the 1025g. The other 3 pins seem to correspond to the missing BNC connectors: CTRL, COUNT IN, AND TRIG.
7. The 3x25 has a second 20 pin connector labelled "MULTIPROCESSOR LINK" used to synchronize multiple 3x25 units for multiple signal outputs using the homebrew ribbon cable shown in the manual. There is a section of the manual titled "Combine Devices" outlining how to do this. The 1025g does not have this connector, and the manual does not have a "Combine Devices" section.
The 1025g advertising claims "Standard USBXITM interface, easily inserts into USBXI housing to make up a combination instrument." However, this mysterious "USBXI housing" seems to be tradmarked vapourware. It's not listed anywhere in their product pages that I could find. Hantek's site search pulls up the 1025g and 3 pc-based scopes that claim to use it, but not the housing itself. Google produces similar results. As far as I can tell, the "USBXI housing" does not exist. I can only guess that Hantek meant to produce some sort of hub that would synchronize these products through their USB connectors but for some reason couldn't get it to work properly and gave up.
Just to be complete about this: Hantek doesn't seem to sell the "Optional Multiprocessor Link Cable" for the 3x25 either. However, they do at least show a couple of photos of it in the manual, so a standard IDC 20 pin cable can be modified quite easily:
Cut wires 1, 2, 3, 4 at each connector and leave them open.
Cut wires 17, 18 and 19 between the master and slave 0 connectors, connect all 3 together at the master connector and leave them open at the slave 0 connector.
Cut wire 16 between the master and slave 0 connectors. Connect wire 16 on the master connector to wire 15 on the slave 0 connector. Leave wire 15 open at the master and wire 16 open at the slave 0 connection.
Cut wires 16, 17, 18, 19 between slave 0 and slave 1. Connect 17 and 19 together at slave 1 connector. Leave all others open.
Cut wires 16, 17, 18, 19 between slave 1 and slave 2. Connect 18 and 19 together at slave 2 connector. Leave all others open.
NOTE: I only have one DDS 3x25, so I have never actually built and tested the above, I'm just going by the pictures! I'm sure Hantek would never make a mistake or omission in any of their documentation :palm:, but use at your own risk anyway.
marabut:
torch,
Thank you for detailed description of differences between the two units. Some of missing features aren't "true" faults (from my point of view).
For example lack of external power connector doesn't seem to be a problem as the device will be connected to PC through galvanic isolation circuit (ADUM4160) and will draw power from stabilized linear PS (this should have positive effect on noise level too). Having some signal inputs at pins (instead BNC connectors) - especially CONTROL - can be even a plus.
Anyway, I plan to make a"frontend" with programmed low pass output filter, buffers for digital signals and analog buffer/amplitude amplifier similar to built by Mechatrommer - less BNC's means easier connection in this case. Things bothering me are: uneven length of tracks leading from FPGA to DAC (is it safe to leave them without damping resistors at such frequencies?), limited number of digital i/o (as I plan to use them to control the "frontend" circuit) and potentially limited firmware features (as e.g. triggered one cycle waveforms). Thanks to Aurora's posted pictures one can see potential drawbacks, however I haven't seen complaints so far - so maybe it's just OK.
BTW - maybe somebody will reveal here some of secrets of Hantek's USB protocol? I know it is possible to write custom software (as Mechatrommer and marmad did) but I wonder if implementation e.g. SPI or I2C on digital I/O is possible that way. Is it fast enough to control the pins and transfer data both ways? It would be nice to control output LPF, attenuator etc. from one piece of software and using only USB as interface.
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version