Hantek - Tekway - DSO hack - get 200MHz bw for free

[Uwe Hermann]

For those using Linux, I did a quick Python script (using PyUSB >= 1.0) which lets you upgrade the scope from 60MHz to 200MHz (or other variants if you edit the script, see comments).

IMPORTANT: USE AT YOUR OWN RISK! YOU HAVE BEEN WARNED!

Anyway, here goes. Maybe it's helpful for some people:

http://randomprojects.org/wiki/Voltcraft_DSO-3062C#USB_.28Linux.29

Uwe.

[Spawn]

Tinhead did really good job on this topic as he did on the firmware hacks, he posted nice link on start post for the Voltcraft DSO so it saved me a lot time. Really big kudos to Tinhead.

Marmad, I don’t have kvk registration anymore and don’t have BTW number, but still this price is great, like you mentioned it is out of stock till 20 april, which worries me I am not sure which version FW we get if we buy one, I don’t think Voltcraft will make a custom FW just for this scope but you never know.

[marmad]

@Spawn - 14 days cooling off period in the Netherlands - if FW is not what you want OR they've changed the scope somehow to prevent upgrades, you can return it - only losing €6.75 for postage.

[tinhead]

Really big kudos to Tinhead

you welcome

out of stock till 20 april

don't worry about the date, anther container is already on the way from China to EU.

which worries me I am not sure which version FW we get if we buy one,
I don’t think Voltcraft will make a custom FW just for this scope but you never know.

the original one-time paid firmware was already "Voltcraft" customized (the update on Conrad website not anymore,
its 100% generic Hantek fw), however these scopes are designed to be easy customizable - not only model
name (resulting in bw) but also in manufacturer names.

The only thing Hantek could do is to design a real 60MHz bw und 500MSs DSO (with less ADCs),
They already produced (and Conrad sold) 1000pcs Voltcraft DSOs, this is still not that number
where you will think about PCB re-design. Additionaly the last real 500MSs firmware is year old (2.06.2),
the difference is really big! This means they need not only new (well, the old original DSO5000C series PCB eventually?)
PCB to be produced but also new firmware.

I know that Hantek is working right now very hard on other things than the firmware bugfixes,
they told me "on new models" which i believe them, but even is this was not truth and they already
working on real 500MSs DSO you can still send it back, so not a big deal at all.

[Spawn]

@Marmad, I still buy a lot from Conrad, if you return it with a good reason you don’t have to pay sometimes they even pick it up for free, I am not sure if they still do that but in the past they picked up couple things from me.
More details here

@Tinhead, thanks for making me feel better
 I am really considering get one, like I said I don’t really need a DSO, I got a Tek 2221A 100MHz with its huge 4kb memory   But this is so cheap and being able to capture a lot more and having screen shots would be really great.

[mightyzen]

It's a very nice piece of equipment, and I really like it. The hacking and linux is a nice plus too. And the price for these devices is excellent. Even without a coupon. For someone without a scope this is definitely the unit to get (imo).

However if you already own a nice tek scope, or an owon or even a rigol, you must ask yourself whether this scope will bring you much extras. Upgrading to 200 MHz is nice, but you will need matching probes. The supplied probes won't cut it. High quality probes from tek or agilent will cost you seriously, cheap probes from china may suprise you or disappoint you horribly. And then: do you need it?

[mightyzen]

For those using Linux, I did a quick Python script (using PyUSB >= 1.0) which lets you upgrade the scope from 60MHz to 200MHz (or other variants if you edit the script, see comments).

Nice. You always recognize a good python script: it is small. 

[Spawn]

Migtyhzen, I got 2 scopes one 20 MHz dual channel and a 100 MHz dual channel with 4kb memory, but I can’t resist the price on this one right now, I think I just need a new toy and a nice digital scope with all modern capabilities. Well I got one more day to think about it and it looks like I am going to order one.

[mightyzen]

Migtyhzen, I got 2 scopes one 20 MHz dual channel and a 100 MHz dual channel with 4kb memory, but I can’t resist the price on this one right now, I think I just need a new toy and a nice digital scope with all modern capabilities. Well I got one more day to think about it and it looks like I am going to order one.

Just follow your heart. I think you will get one  and you won't be disappointed. The question will be: what about your other scopes?

Forget about the 20MHz. But the 100 MHz dual channel with 4kb memory, if it is a tek as I understood it may be nicer than what you will be ordering from conrad. Care to share some details?

[Spawn]

.... Care to share some details?

Sure It’s a Tek 2221A I got it couple weeks ago See here To be honest I still grab my 20 MHz one for simple things, and I am attached to my gear always so the 20 MHz will stay on my bench next to others, besides my son is studying electronics so he can use it too.

I might need some help when I get the Voltcraft from you since you figured it out all

[Uwe Hermann]

I started documenting my JTAG setup for the S3C2440 (basically I abused an old IDE cable and soldered directly to the PCB), and I also wrote a small OpenOCD config (which has been included upstream already).

http://randomprojects.org/wiki/Voltcraft_DSO-3062C#JTAG_.28for_Samsung_S3C2440.29

JTAG seems to work fine, I'm currently dumping the NAND flash chip contents via JTAG (takes quite a while though).

I've also started playing with Barebox (will also test U-Boot later) and buildroot (for rootfs and cross-toolchain). Building a Linux kernel and flashing my own rootfs will be the next step.

My UART setup is also documented here:

http://randomprojects.org/wiki/Voltcraft_DSO-3062C#UART_2

For this, I used a standard FTDI FT232R-3V3 cable and some custom-soldered mini-addons to allow for the cable being removed later (if I need it for another project).


Also, if anyone is interested in more PCB photos, UART bootlogs, lsmod/cpuinfo/mtdinfo/etc., waveform comparison before/after the 200MHz hack, and such stuff: I've put some more photos and info on the randomprojects.org wiki, and more detailed photos of individual PCB parts / chips will follow.


Uwe.

[marek]

Hello @ everybody,
just signed in, after digging page by page throught this thread... Tinhead! Kudos! I bought a handheld DSO1062B (now "almost DSO1202B") mostly because of your work on these scopes.

Now, there's a bug in the handheld I have found, that I haven't seen mentioned yet. Before I explore and document it fully, I'd like to know if it hasn't been already noticed and documented...

To reproduce it:
- Activate CH1 only. Apply some HF signal so that you see the effect (I put a 33MHz clock on CH1)
- In double windows mode zoom into the signal (horizontal resolution M:200ns and W:4ns or so).
- Confirm you can see the signal clearly.
- switch CH2 on. As expected the samplerate halves and the signal resolution drops (this is not part of the bug! this is expected!)
First part of the bug:
-> although the time scale still shows 4ns/div the signal is stretched 2x. You have to switch forth and back on the timebase to get it back right.
Second part of the bug:
Press CH2 again to switch it off. Watch the top window. On the left and right margin you'll see some trash displayed.

There's also a second bug in Pass/Fail function:
- Set up everything and let "stop after fail" on.
- It will stop at fail, but trigger once again and so overwrite the one-off signal you are looking for :/

Marek

EDIT

here: http://www.biasedlogic.com/index.php/hantek-dso1062b-dso1102b-dso1202b-vs-5gss-lecroy-waverunner/ I have put a few observations about triggering on tricky signals from LeCroy's "demo generator". There's also a comparison of 33MHz "square" wave image from the Hantek and from a 600MHz 5GS/s LeCroy Xi-series (Xi 660 I think).

[tinhead]

Tinhead! Kudos!

prosze bardzo

Now, there's a bug in the handheld I have fo

indeed, i can confirm the first bug (actually only the first part, i ddin't saw any trash on my display)
on 111212 firmware version. Btw, what is loaded on your handheld?

Not sure about second bug, sometimes  it looks like what you said but sometimes it seems to be
just a bit delayed post-processing.

here: http://www.biasedlogic.com/index.php/hantek-dso1062b-dso1102b-dso1202b-vs-5gss-lecroy-waverunner/ I have put a few observations about triggering on tricky signals from LeCroy's "demo generator". There's also a comparison of 33MHz "square" wave image from the Hantek and from a 600MHz 5GS/s LeCroy Xi-series (Xi 660 I think).

nice test!

I did compared both handheld and bench versions, from what i can saw handheld
is having bit less jitter (pcb routing and new fpga design) and bit higher bw.
With active probes (and the /tst file hack to set bw filter higher) i got 250MHz -3db bw,
more than sufficient for 1GSs DSO.

"PP008 price", yeah, this is "the dark side of cheap scopes", a good set of probes costs
more than such DSO itself.

[marmad]

here: http://www.biasedlogic.com/index.php/hantek-dso1062b-dso1102b-dso1202b-vs-5gss-lecroy-waverunner/ I have put a few observations about triggering on tricky signals from LeCroy's "demo generator". There's also a comparison of 33MHz "square" wave image from the Hantek and from a 600MHz 5GS/s LeCroy Xi-series (Xi 660 I think).

Yes, Marek - very nice 'tutorial' on finding common problems in modern digital signals. You might consider posting it in a new thread for new DSO owners to reference - I think it's very handy. Nicely done!

Mark

[tinhead]

before i forget, the really nice tool from:

http://www.dreisiebner.at/dso-usb-tool/

is not 100% working on handhelds.  This is because on handhelds (and probably bench BM/BMV models)
a newer linux version with completly different driver is installed.

In principle they compatible to each other, unfortunately the newer driver is sometimes "too fast"
which breaks the communication.

The autor of this tool can not fix it without access to affected hadrware, i have to check with him,
maybe he can try to fix it when i send him such hardware (my knownledge of RealStudio is very basic only).

[tinhead]

i think i will rename this thread ... these new handhelds are in principle like the BM/BMV bench model -
same linux, same features (of course except DMM and ignore display size), same special hacks (LAN, SD, etc.)

I'm honestly a bit too lazy to maintain two thread ..

[marek]

prosze bardzo

What's your mother tongue?

indeed, i can confirm the first bug (actually only the first part, i ddin't saw any trash on my display)
on 111212 firmware version. Btw, what is loaded on your handheld?

111212.0, same as you mentioned

Not sure about second bug, sometimes  it looks like what you said but sometimes it seems to be
just a bit delayed post-processing.

I'm positive about the next acquisition running, as I did try to hunt down a one-off pulse and it was NEVER to be seen although the scope would trigger on it. I fed the signal parallel to the Hantek and a LeCroy and could see the one-off on LeCroy at the same moment the Hantek would beep...

I did compared both handheld and bench versions, from what i can saw handheld
is having bit less jitter (pcb routing and new fpga design) and bit higher bw.
With active probes (and the /tst file hack to set bw filter higher) i got 250MHz -3db bw,
more than sufficient for 1GSs DSO.

Yup, the scope is quite well balanced about the design: there's no single bottleneck, if you really want to improve it you have to get all parts updated.

"PP008 price", yeah, this is "the dark side of cheap scopes", a good set of probes costs
more than such DSO itself.

And by what margin... brrr!

By the way:

I have populated the DM9000EP along with a LF8505 magnetics (picked one from an old network card) and in the bootlog I see the driver picking it up. So far so good. I'm still missing a suitable jack, but I have ordered one that might fit, it just didn't arrive yet.

The scope menu shows network not supported, I believe this is configuration issue, do you have any details on this?
Is there any useful functionality available over the network interface BESIDES a console (which was reason enough for me to try the thing with a DM9000EP)?

Greets!
Marek

[liebref]

Hello ,

I am new here  and have an Voltcraft DSO upgraded to 200MHZ . All worked fine  so far .
Now by looking at the Hantek home-page , I found  an new Firmware 120224.0
Has anybody already experience with this version -repectivly what known bugs are fixed ?

Thanks for  feedback in advance

Liebref

[tinhead]

prosze bardzo

What's your mother tongue?

No idea, she never told the truth ... but let me tell you something, we both (you and i) could use
at least 3 (same) langauges to ask for beer

I have populated the DM9000EP along with a LF8505 magnetics (picked one from an old network card) and in the bootlog I see the driver picking it up. So far so good. I'm still missing a suitable jack, but I have ordered one that might fit, it just didn't arrive yet.

The scope menu shows network not supported, I believe this is configuration issue, do you have any details on this?

send me PM and i will explain you how to make it working.

Is there any useful functionality available over the network interface BESIDES a console (which was reason enough for me to try the thing with a DM9000EP)?

there is indeed full support in the firmware, you can use LAN to communicate like you do over USB - the full protocol is implemented,
even the original PC software support LAN connectivity.

[tinhead]

Hello ,

I am new here  and have an Voltcraft DSO upgraded to 200MHZ . All worked fine  so far .
Now by looking at the Hantek home-page , I found  an new Firmware 120224.0
Has anybody already experience with this version -repectivly what known bugs are fixed ?

Thanks for  feedback in advance

Liebref

i would not use the 120224.0, sure it does fix the "roll mode" bug, but it does have few new bugs.
Even Conrad removed this fw from their website. Soon there will be new version available, so just be patient.

[liebref]

Hello ,

Few hours ago , I checked  the online store from Conrad ; SURPRISE !  It does show  the Voltcraft  DSO 3062C  "Sold out "
As the original delivery dates  were already changed in the past from  20 April  to  20 May , I guess somebody has stopped this offer .
Guess they have figured out the real value of the offer 

[marmad]

Few hours ago , I checked  the online store from Conrad ; SURPRISE !  It does show  the Voltcraft  DSO 3062C  "Sold out "
As the original delivery dates  were already changed in the past from  20 April  to  20 May , I guess somebody has stopped this offer .

Well, it was listed on the German Conrad site as an anniversary offer - which would indicate limited-time.

[prpplague]

Hello ,

Few hours ago , I checked  the online store from Conrad ; SURPRISE !  It does show  the Voltcraft  DSO 3062C  "Sold out "
As the original delivery dates  were already changed in the past from  20 April  to  20 May , I guess somebody has stopped this offer .
Guess they have figured out the real value of the offer 

There are some other reasons why the Voltcraft DSO-3062C is no longer available, hopefully there will be an update on the status in the near future....

[marmad]

There are some other reasons why the Voltcraft DSO-3062C is no longer available

Huh?  Would you please care to explain?  This forum is a place of open exchange and dispersal of information.

[prpplague]

I started documenting my JTAG setup for the S3C2440 (basically I abused an old IDE cable and soldered directly to the PCB), and I also wrote a small OpenOCD config (which has been included upstream already).

http://randomprojects.org/wiki/Voltcraft_DSO-3062C#JTAG_.28for_Samsung_S3C2440.29

JTAG seems to work fine, I'm currently dumping the NAND flash chip contents via JTAG (takes quite a while though).

I've also started playing with Barebox (will also test U-Boot later) and buildroot (for rootfs and cross-toolchain). Building a Linux kernel and flashing my own rootfs will be the next step.

My UART setup is also documented here:

http://randomprojects.org/wiki/Voltcraft_DSO-3062C#UART_2

For this, I used a standard FTDI FT232R-3V3 cable and some custom-soldered mini-addons to allow for the cable being removed later (if I need it for another project).

Also, if anyone is interested in more PCB photos, UART bootlogs, lsmod/cpuinfo/mtdinfo/etc., waveform comparison before/after the 200MHz hack, and such stuff: I've put some more photos and info on the randomprojects.org wiki, and more detailed photos of individual PCB parts / chips will follow.

Uwe.

in conjunction with Uwe, i too have started some documentation on hacking around with the Hantek devices. we are calling it the Das_Oszi project for now:

http://elinux.org/Das_Oszi

all the core resources are listed there as well howtos such as adding an external uart to the case using a 3.5mm audio jack:

http://elinux.org/Das_Oszi_3.5MM_Uart

additionally i have git repo's for barebox and mainline kernel which are now able to boot on the device.

Dave