New Hantek DSO2X1X models?

[Cántabro]

Hello, greetings, with the latest updates, has the trigger problem been solved?, which did not stop the image, it was said that the trigger worked by software, not by hardware. Does it work with the latest updates correctly? It is something that stops me from purchasing

[morgan_flint]

...sorry for my bad english...

Maybe, your problem is the same like mine DSO2D10 after a few month: suddenly shortcut in one of rotary encoder case (GND) to the inner encoder PCB.
It's blocking the data lines.
Mine repair was simple: teardown the encoder case, a little tape to isolate.
Search this video in YouTube: "Hantek DSO5202P: Knobs/Encoders Not Working Repair"
Maybe helping you...

Hello and thank you! Sorry for the late response! I didn't get or maybe I didn't see the email announcing your post. I'll see the viden and try if it's the same problem.

In the meantime, I've ordered the HC4051 to replace U2, waiting for it to arrive. With a bit of luck, the problem will be solved by the time it arrives!

[serafis]

In the manual, P51 Auto Scale, it says "Trigger Coupling is adjusted to DC, Noise Reject or HF Reject". Implies it has this capability - is there a way of setting Trigger Coupling manually (e.g. to Noise Reject)? I can't find it under the Trigger menu or Auto Set menu!

[morgan_flint]

...sorry for my bad english...

Maybe, your problem is the same like mine DSO2D10 after a few month: suddenly shortcut in one of rotary encoder case (GND) to the inner encoder PCB.
It's blocking the data lines.
Mine repair was simple: teardown the encoder case, a little tape to isolate.
Search this video in YouTube: "Hantek DSO5202P: Knobs/Encoders Not Working Repair"
Maybe helping you...

Hello and thank you! Sorry for the late response! I didn't get or maybe I didn't see the email announcing your post. I'll see the video and try if it's the same problem.

In the meantime, I've ordered the HC4051 to replace U2, waiting for it to arrive. With a bit of luck, the problem will be solved by the time it arrives!

EDIT: Indeed, that was the problem!!!!

One of the encoders had one of the sides derived to the case, which in turn is connected to the ground. As the video that @Joemega suggested to view explains, this is equivalent to having Scan_Data_4 or Scan_Data_6 (the latter in my case) connected to ground via the corresponding diode, so no change for any of the encoders is detected when Scan_Line_X selects them (see first attachment).

So I did the same as @Joemega: opened the offending encoder by straightening the 4 tabs that hold the plastic part to the case and insulating the plastic part from the metallic part with some tape.

As you can see in the 2nd and 3rd photos, the failure was caused by a bad encoder's design (I don't think it can be fully blamed on Hantek, but mostly on their supplier...). The plastic part has conductive circular tracks embedded and these tracks have four little spikes (two at each side) protruding from the plastic matrix exactly where the aforementioned tabs lay when the encoder is closed. Depending on how the tabs are pressed during manufacturing, maybe there will be contact or not, but the distance would be very short, so just with normal use or with a light blow on the buttons this contact may happen, as was in my case.

Now the scope is back to normal and the panel self test goes OK. From what I've seen in the eediscuss forum, this failure is not that uncommon (not only for DSO2000 series, as the video is for another model), so maybe this can help if your oscilloscope develops this failure over time.

Many thanks to @Joemega for spotting the solution and to @DavidAlfa for his help analyzing the failure.

[serafis]

After applying all the mods I wanted and overclocking to CPU 720MHz, memory 408MHz, to my DSO2D10, I'm very happy with the performance and stability of my upgraded DSO2D15. No freezes or glitches after several hours of use. Thanks to DavidAlfa and others for the tips and tricks.

[IonutL]

Hello everybody,

After my first adventures with DSO2D15 (hardware issue still unresolved, I will need to replace the memory chip which is on the way together with a programmer and converter board), I ended with my laptop loaded with all the software necessary to connect to and program the oscilloscope.
Then I started to have strange behavior on my laptop, and errors related to USB connected devices (smartphone, DVD-writer) which are caused by the before mentioned software. So I have put together some instructions on how to "clean" your PC and restore back the USB to working condition, and I am now sharing this with you:

USB is derailed from normal functioning:
- connecting phone does no longer work
- writing more than one DVD on a writer connected to USB fails randomly (some of the written files do not verify)
- possible other errors on USB-connected devices

Steps to recover:
- uninstall Phoenix Suite, do not press the button with chinese characters (it says "Cancel" !), be patient it will uninstall in a minute or two
- no need to touch zadig
- under Device Manager/ Show hidden devices in View menu, then list USB devices with VID_1f3a_PID_ef08, there are possibly multiple devices and in multiple places; then normal uninstall all BUT leave only one of them, the last uninstall do it with checking "Delete the driver software for this device"; look for devices under:
Universal Serial Bus controllers
Universal Serial Bus devices
- uninstall Windows software listed in Settings/ Apps (here is what I had installed on my laptop):
DSO2000
IVI Shared Components 3.0.0
IVI.NET Shared Components 1.1.2 for .NET 2.0
IVI.NET Shared Components 2.0.0 for .NET 2.0
Keysight Communication Fabric 2.2
Keysight Discovery Service
Keysight Distributed Infrastructure
Keysight IO Libraries Suite 2023 Update 1 -> it will complain about not finding an uninstaller, there is no problem, just go on
Keysight VISA Cross-Platform Components - windows -> will be uninstalled by "Keysight IO Libraries Suite 2023 Update 1"
Keysight VISA Cross-Platform Components - windows x64 -> will be uninstalled by "Keysight IO Libraries Suite 2023 Update 1"
VISA Shared Components 7.2.0 (64-bit)
VISA.NET Shared Components 5.6.0 (64-bit)
VISA.NET Shared Components 7.2.0 (64-bit)

- reboot PC

Note: this will repair the MTP driver for Android smartphone, if not: under Device Manager, right click on the MTP driver with yellow exclamation mark/ Update, choose to browse computer/ pick from a list of available drivers/ click on the MTP driver, then reboot again

[nemasu]

...
I honestly can't understand why anyone would buy one of these when Siglent and Rigol (and other manufacturers) with great reputations build fantastically better instruments for about the price of a Big MAc dinner and a 6-pack more.
...

smh. I just ordered a DSO2D10, but I can still cancel it. Does anyone have a model recommendation that has similar specs? (100MHz, 1GSa/S, function gen) for just a bit more?
The only one I can find that has a wave generator is the Rigol DS1104Z-S Plus, but this one is over 5x the price. 😩

[Aldo22]

smh. I just ordered a DSO2D10, but I can still cancel it. Does anyone have a model recommendation that has similar specs? (100MHz, 1GSa/S, function gen) for just a bit more?
The only one I can find that has a wave generator is the Rigol DS1104Z-S Plus, but this one is over 5x the price. 😩

As far as I know, there is nothing from Rigol or Siglent in this price range.

The cheapest are Rigol DS1102Z-E or Siglent SDS1202X-E (or SDS1102CML+), as far as I know. Prices differ by country.

A signal generator could be bought separately. E.g. a FY3224S for the beginning does not cost much.

[nemasu]

smh. I just ordered a DSO2D10, but I can still cancel it. Does anyone have a model recommendation that has similar specs? (100MHz, 1GSa/S, function gen) for just a bit more?
The only one I can find that has a wave generator is the Rigol DS1104Z-S Plus, but this one is over 5x the price. 😩

As far as I know, there is nothing from Rigol or Siglent in this price range.

The cheapest are Rigol DS1102Z-E or Siglent SDS1202X-E (or SDS1102CML+), as far as I know. Prices differ by country.

A signal generator could be bought separately. E.g. a FY3224S for the beginning does not cost much.

Thanks for the suggestion. Looks like those are 2x+ the price unfortunately, at least in Japan. Oh well, I guess I'll stick with the Hantek and see how it goes.

Edit: Gosh darn it, looks like the DS1102Z-E went on sale, and just as soon as I get the shipping notification for the Hantek -_-

[DavidAlfa]

Also the new Rigol DHO800 series look very nice.
12 bit ADC and 1024x600 touchscreen, $350 2ch/70MHz version, $400 4ch/70MHz, blows Hantek out of the water.
I think the 802/804 can be hacked to the faster 100MHz 812/814, but I didn't search for it.
https://www.batronix.com/shop/rigol/DHO800.html

[dmulligan]

Also the new Rigol DHO800 series look very nice.
12 bit ADC and 1024x600 touchscreen, $350 2ch/70MHz version, $400 4ch/70MHz, blows Hantek out of the water.
I think the 802/804 can be hacked to the faster 100MHz 812/814, but I didn't search for it.

Early tests show hacks are possible but they aren't perfected yet though I am sure there won't be a problem once there are enough oscilloscopes in peoples' hands.

Also thank you David for voicing your opinions of the DSO2x1x oscilloscopes.  I nearly bought one but held off to see if the firmware improved.  Now I have a Rigol DHO804 on the way.  You might consider adding a new first section to your DSO2x1x FAQ titled "Should I buy one?" with links to you and others saying "don't."

[Aldo22]

Anyway, in one respect the Hantek is unbeaten: it costs one third of a Rigol with AWG.
As problematic as it may be, no one has ever doubted the bandwidth or sampling rate as far as I know.
Unfortunately, the Hantek has no competition in terms of price and is therefore difficult to compare.

[IonutL]

Dear friends, I need your help again.
I received a faulty (memory chip error) DSO2D15 and Hantek support instructed me to replace the W25N01GVZEIG. I bought some chips from AliExpress and now the plan is to: (1) desolder the old chip from the board, (2) sit/ solder a W25N01GVZEIG on a SOP8 board with pins, (3) prepare some sort of socket for this convertor I manufactured and (4) solder the socket with some wires to the motherboard to replace the original W25N01GVZEIG (David has done something like this using old MMC card and a MMC socket).
But now my problem is that I desoldered the memory chip from the board using hot air, but I managed to rip two of the pads which were used to connect the W25N01GVZEIG (attached), the corresponding pins are 7 and 8, on the left-down part of the empty place where the chip was.

Do you know any alternative locations on the motherboard where I can solder wires, i.e. where those 7&8 chip pins were going onto the board?

Many thanks for any ideas you would provide!

[DavidAlfa]

Always poke the IC with the tweezers before trying to remove it, if not moving freely then more heating is required!
I disencourage doing the socket thing, I did in a time where almost nothing was knew about the scope.
I removed all that short after learning there's nothing important inside and no special procedures are needed for recovery.
Check the schematics, also this picture.

[vistorik]

Dear friends, I need your help again.
I bought some chips from AliExpress and now the plan is to: (1) desolder the old chip from the board, (2) sit/ solder a W25N01GVZEIG on a SOP8 board with pins, (3) prepare some sort of socket for this convertor I manufactured and (4) solder the socket with some wires to the motherboard to replace the original W25N01GVZEIG (David has done something like this using old MMC card and a MMC socket).

I did something similar to what you want to do in order to be able to quickly change memory chips during various tests or in cases of checking problematic dumps from members of the Russian forum. As it turned out, this socket thing does not work without any problems with any wires. You can solder everything correctly, but the scope will not boot, or after updating or using platform-tools, the scope stops working. Maybe I was just unlucky. The wires I use now without any problems are a set of breadboard jumper wires from Aliexpress. So I would not advise you to build such socket things.

[IonutL]

Thank you both for your advice, I appreciate your experience. Unfortunately for me it's not a choice, I need to revive the scope I just bought, and soldering back the memory with hot air seems to be impossible now (with the two pads exfoliated, and my soldering skills seem to be poor...).
So if I do not try with wires and some kind of socket, the scope is as good as a paperweight.
I will let you know what happens.
Best regards to all.

[vistorik]

soldering back the memory with hot air seems to be impossible now (with the two pads exfoliated, and my soldering skills seem to be poor...).

Such damaged pads are easily repaired using two thin wires as shown in David's photo. At least you could try this. This is better than the story with socket thing.

[IonutL]

@vistorik: there is one more impediment I did not mention: the couple of W25N01GVZEIG I received from AliExpress are used ones (despite the seller said they are new!!), and the back was polished to hide the fact they were desoldered from some board. As a result, they do not adhere well to solder, I spent a few days trying to make them stick on a converter board using hot air gun and solder paste. So putting such a chip on the DSO motherboard would inevitably lead to over-heating (due to my lack of experience with hot air soldering) and more pads exfoliating, eventually desoldering other components and destroying the motherboard beyond repair.

So despite your solution being very elegant, I went on with the converter-socket-wires approach. The picture shows the hardware part of the result..

First I put a memory flashed with the image I saved with dsoflash when the scope arrived. The scope started, but the functionality was poor: buttons were sluggish, de-calibration of the input, could not save into internal memory (kind of errors I have seen previously when I started asking help in this forum).
Then I put another memory (huh, this is one advantage of having a socket - could change the memory very easy!) and flashed the scope using platform tools in FEL mode. Scope started and was working much better, then I upgraded to the latest firmware from Hantek website, calibrated, and all works well now. I tried to set the serial number and board ID using the upks created when I ran Backup builder when I received the scope, but I got error -3 (and this error appeared on any upgrade with a package from that Backup builder created set. Finally I managed to change those two using commands in Keysight IO Monitor (PRIVate:SET:SERIal and PRIVate:SET:PCB).

So I would say that it is still a success. I can use this scope for visual inspection of waveforms and will not trust this scope for exact measurements, but at least I did not loose all the money, which is something! And all this is due to you all who worked to decipher the intricacies of this scope and share this information with us.
Thank you all !

[DavidAlfa]

Error -3?
Make sure you're placing the upk files at the root of the USB drive, better avoid folders, absolutely avoid spaces in folders or filenames, always use safe drive extraction.

Also, place a small decoupling cap in the socket pins to compensate the long wires, the usual 100nF should do it.

The only thing you can really lose is the generator amplitude calibration, the scope channels will work fine after running a calibration procedure, so relax, you can trust it.
Still, I don't think the upk backups are broken, I can check them if you send them to me.

[IonutL]

It turns out that the generator calibration is fine, I checked it at 1kHz and 100kHz, for 1.5 and 6V amplitude and for all the waveforms it can provide - it seems that the amplitude and frequencies are ok (at least gross values, not talking about mVolts).

As for upks all conditions were met. I unpacked the dso3kb_CNxxxxxxxxx.upk and all the files are ok, they are even exact content with the other backup by backup builder I took with the new memory chip and after setting everything up - all files within the upk are identical.

[IonutL]

And I have uploaded on Gdrive the files (in the writable folder).
Thanks again.

[vistorik]

Then I put another memory (huh, this is one advantage of having a socket - could change the memory very easy!) and flashed the scope using platform tools in FEL mode.

After you flashed the scope using platform-tools, the calibration generator file dds_calbration.dat could be restored by the system using calibration coefficients from EEPROM PT24C02. You could check this using dso3kb_CNxxx_Calibratio.tar.bz2 and dso3kb_CNxxx_EE.tar.bz2 archives from your backup. From the dso3kb_CNxxx_EE.tar.bz2 you could get ee.bin - this is the dump of PT24C02, and from the dso3kb_CNxxx_Calibratio.tar.bz2 you could get dds_calbration.dat. Using any HEX-editor you could see calibration coefficients in these files in Hex and Double float format. To convert these formats you can use a Floating Point to Hex Converter (https://gregstoll.com/~gregstoll/floattohex/). /). Calibration coefficients in HEX format (HEX value) must be entered byte by byte in reverse order, and then you click "Convert to double" - these should be calibration coefficients in ee.bin (see the pictures as example). The calibration coefficients in the dds_calbration.dat and the ee.bin are in red frames.

[stjohn909]

I've been reading this thread for 3 days now, and I can't thank everyone enough for sharing their knowledge and experiences.  Very much appreciated. 
I just got a Hantek DSO2D15 off Amazon for $215USD, and so far, it's been great except for one thing:  I work with MIDI, so I need to decode UART at custom baud rate of 31250.  So far I haven't been able to get there because:

• The initial value is ~115200, and the multifunction knob only increments/decrements by 1.  You can imagine how long that takes to get to 31250.  Still, I tried.
• Once I got down around 49000, the number would randomly increment by ~10000, ~1000, or ~100, with no discernible logic to which one.  If Hantek tried to implement ballistics on the multifunction encoder for custom baud rate, it needs work.  Or I'm doing it wrong.
• The error always increments the value, never decrements.
• At this point, I can never get below 55000, because once I get down around there, no matter how slowly I roll the multifunction encoder, the value will always jump up by one of the aforementioned amounts.

Has anyone else seen this? 
I would assume a workaround is to set it manually via PC connection, but it seems odd that Hantek didn't implement the custom baud rate by allowing you to choose a starting point and rolling from there.  But like I said, I may be doing it wrong.  The manual is fairly sparse on this subject. 

I'm also on a Mac, so I haven't tried to set up the SCPI tools yet. The eediscuss forum shows SCPI commands available for the Trigger/UART namespace, so maybe I can set it there?

Software is 1.0.3.0.0(230831.00)
Hardware rev is 003.002.000.000.000.000.001
Firmware is 3205.

[DavidAlfa]

Rotate slower, then it'll let you adjust it. If you go too fast then yeah, it'll go in 1000, 10000 steps. (Or maybe it's a new software bug introduced in the newer version you have!)
Unpack this setup file, copy the .pho file to the root of a usb drive and load it from the Save/Recall menu (Choose external)

23.08.31 is a newer software version, not released yet!
Please run Backup Builder and send me the FULL package so I can make a new update, everything is in the FAQ (Check my signature).

Edit:
For whatever reason, I mean, because it's *** Hantek, it saved a crooked setup, the scope went extremely slow after loading it.
Saved a new setup, this worked as it should.

[stjohn909]

DavidAlfa, Full backup uploaded to your drive per the FAQ and the Hacking thread.   Your .pho file gave me the 31250 baud rate I needed, cheers!