New Hantek DSO2X1X models?

[masta_k74]

If you do not save on details, you can make a reliable power supply. I myself developed power supplies for technological equipment and passed various certifications, it turned out very reliably and quietly, I can’t imagine a stabilized power supply for 12V 100A with a transformer. Normally, the input voltage is ～90-260V, transformer units cannot do this.

Of course, there are cases where a switching power supply is appropriate (or absolutely necessary) and there are certainly high-quality and reliable power supplies, but the one used here seems rather questionable to me. As I said, I have had many defective power supplies of the cheapest type from China on my desk and I know the problems of these devices by now.
But that doesn't apply to all Chinese power supplies. They can also do it very well.
But you can see that immediately when you open it.
In general, I have nothing against switching power supplies, so please don't misunderstand.
I occasionally use and build switching power supplies with galvanic isolation, but so far only for small input voltages, e.g. to generate higher symmetrical voltages for amplifier circuits (audio).
(My last switching power supply, a regulated push-pull converter, 24Vdc --> 2x 44Vdc works in an old HiFi - amplifier to use it in the garden arbour at the photovoltaic system. The garden has no connection to the mains).

[DavidAlfa]

So I had some ultrafast 20ns 8A rectifiers (BYC58X-600), decided to make the test.
Replaced a capacitor I didn't last time (The 5V output for the digital power).
Cutted an aluminium thin foil, wrapped the power supply with it, grounded it with the board screws.
Last, I added a small toroid near the +-8V input connector of the analog side.

Seems like it might have -very slightly- improved the noise, but can't really tell..
Previously there was a subtle hiss noise coming from the psu, now its's completetely gone.
20mV looks pretty much the same.

[Russ3000]

Diodes? I don't think so.
More like the capacitors you replaced at the same time, which are by far the most important components when filtering the spikes.
[/quote]
The eldest son is studying at the Technion in his last year, they study diodes in depth, I asked him about the operation of Schottky diodes at elevated temperatures, he asked his professor, he replied that short reverse current surges are possible when heated above 100C, he explained the theory of the issue to me for a long time , but I didn’t remember much, because the children are smarter than me.
The main source of interference is the ground loop between digital and analog power, if there is a source of electromagnetic interference inside this loop, then they will be displayed on the screen. If I have free time, I will try to wind 3 turns of wire on the transformer from above, without disassembling the transformer, I will put a diode and a capacitor, I will start the feedback of the regulator and it will turn out to be more stable 5V, I will not connect the ground, but there is no need to stabilize + -8V, there are 7805 and 7905 on the board , they perfectly stabilize the voltage.

(My last switching power supply, a regulated push-pull converter, 24Vdc --> 2x 44Vdc works in an old HiFi - amplifier to use it in the garden arbour at the photovoltaic system. The garden has no connection to the mains).

I bought a large solar battery, I am waiting for the battery charge control unit with the MPPT system to arrive, I will experiment, the most interesting idea seems to be to convert the voltage to 230V and return it to the mains so that the energy company also pays money

[DavidAlfa]

They do increase reverse current with temperature, but nowhere to become a problem.

As you see, even at 150°C (Quite extreme conditions) the reverse current will get around 10-20mA, not optimal for power efficiency but a small current after all.
To cause these issues you need fast and large current spikes.
Or simply crappy (High ESR) capacitors not able to filter the high frequencies.
If you have time, try restoring the original diodes while keeping the upgraded capacitors, you'll see.

[masta_k74]

I bought a large solar battery, I am waiting for the battery charge control unit with the MPPT system to arrive, I will experiment, the most interesting idea seems to be to convert the voltage to 230V and return it to the mains so that the energy company also pays money

Unfortunately, this is not possible at the arbour, as there is no electricity grid.
The reason for converting the amplifier to 24V was to reduce conversion losses. You would have had to convert from 24Vdc to 230Vac first, and then transform back down to 2x 44Vdc in the amplifier's transformer power supply. In this case, the losses in the transformer and in the voltage converter are already significant, since only a limited capacity of batteries is available.
This is the clear advantage of the switched-mode power supply.
Of course, there is also a 230V inverter in the arbour.
We also have a petrol generator. In winter, you have to recharge the batteries when the sun is rarely visible.

So, enough off-topic

[nokin]

Hi all,
Can you give me idea about U24/U27 opamp's speed could be related to waveform refresh rate on screen ?

Now I'm temporary replacing LMH6702 by Ad797 and the refresh rate quite slow and very annoying.

[Russ3000]

As you see, even at 150°C (Quite extreme conditions) the reverse current will get around 10-20mA, not optimal for power efficiency but a small current after all.
To cause these issues you need fast and large current spikes.

I suppose the temperature of the crystal is much more than 150 degrees, the diodes have a large thermal resistance of 48 ℃ / W, users on 4PDA have measured more than 100 degrees outside.

[DavidAlfa]

I don't think so, took one of the original diodes and tortured it.
Tests were done aplying 12V in reverse polarity.
Used a large solder tip, applying a blob of solder that almost surrounded the whole the diode, except one pin to avoid making a short.
The tip was placed and left untouched, each temperature step was applied for 1 minute.

Code: [Select]

185ºC   600uA
215ºC   3.25mA
230ºC   7.5mA
245ºC   19.5mA

At 215ºC it started to make that terrible burned electronics smell (No way this is the temperature inside the PSU).
Finally crancked the temp to 280ºC, it avalanched and died (Permanent short).

You replaced the diodes and the capacitors, then blamed the diodes, that doesn't make sense.
It's like if my engine is broke, replaced 20 pieces and blamed a random part.
The only way of getting a proper result is doing a procedural method, testing one element at a time.
I'm not discussing this topic anymore!

[Russ3000]

I'm not discussing this topic anymore!

Ok, let's leave the diodes and move on to the capacitors. before replacing, I measured the soldered capacitors, their ESR was 3 mΩ, the ones that I soldered showed 1 mΩ, not much less, I believe this did not lead to a noticeable reduction in noise.

[badbod]

Placed ferrite over the end of the long white cable coming from PSU (at scope end), surely any cable of length leading into the faraday should have this no?
Placed heatsinks on the 2 chips, deffo at least 5 or 6C above room temperature those chips were.
Added a 5v 20mm brushless fan pointing directly at the PSU (was almost 20C above room) Silent fan so no downside. Massive diode already on the PSU for an easy 5V, right there, for the taking, candy from a baby :-) IN4007 for a small drop as I do not want to cryogenically preserve the PSU. (yeh I know 1000V, but I got a whole bunch of them)

Result is impressive, 
A) Made absolutely no difference whatsoever to the scope. This is good, did not adversly affect it in anyway way that I can tell.
B) Completley worth the effort, I just feel so much better now. Though a little dissapointed I did not need to break out the microscope for it but happy that my eyes are still good enough that I did not need to break out the microscope for it, unlike reading the instructions 'leaflet' provided with just about anything these days.

To be honest, since earthing the unit properly it isn't noisy. Nothing I have done is to fix anything, normally I would expect these types of things to already be there. Anyway, might wanna overclock it when Im playing DOOM (which is, after all, why we are all here).

There is a 240uV sine @ x1, 1.02mV, 2ns, but I'm pretty sure that is just there to troll.

[DavidAlfa]

I agree, but also remember that electrolytics are usually particulary bad a high frequencies, so the measured esr cannot be applied in the same way.

A flyback transformer usually causes some ringing at the secondary, at much higher frequency than the main switcher.
See Vsec? Doesn't it look quite similar to the noise seen on some scopes?


My proof is that it went away after replacing only the capacitors and properly grounding the wall socket.
Maybe we should bet some real money

[Russ3000]

A flyback transformer usually causes some ringing at the secondary, at much higher frequency than the main switcher.

Just the EMI ringing is generated by fast diodes when turned off, I encountered this when I was designing a power supply module with PFC, I had to put silicon carbide SiC diodes instead of Schottky diodes and the ringing stopped, otherwise it was impossible to pass certification. In the oscilloscope, I replaced the diodes with higher capacitance diodes, which most likely eliminated ringing during switching. I did not expect such an effect, I changed it because it got very hot.

[DavidAlfa]

You quoted the translated output  .
Yeah, we need someone with two scopes (Not another DSO2xxx), so he can see the secondary waveforms.

[masta_k74]

@Russ3000
What do you think about the F12C20CT as a replacement for the two SS310s in parallel in the PSU? However, this is not a Schottky, but a Fast Recovery Diode.

[Russ3000]

You can try, without experience you can not immediately say.

[Systops]

Hi everyone,

I am new to the group and I was pointed to this thread because I recently purchased a Hantek DSO2D10, which is their 2 channel 100Mhz Oscilloscope, which I am returning. I have found this unit extremely buggy and while I thought it would be a great scope for what I need it for, the bugs have become way to intrusive.

For example, if you change settings too fast, then the unit will completely lock up, and will act as though it's bricked. To fix this you have to reflash the firmware. Being that I tend to move quickly I wind up hitting this bug every few minutes, have to stop what I am doing, reflash the firmware and start over.

The other bug which has me really perplexed is its inability to accurately display proper voltage when you are trying to measure under 1V or is it 1mV, honestly, I can't recall, but I learned about this from a youtube who did a comparison between this unit and the equivalent Firsi and OWON, and sadly the OWON beat it out as far as accuracy, Needless to say, I am returning it.

Interestingly, I also have a Hantek 2D72 which is a 2 channel Scope, DMM, and Signal Generator, and I have not run into any similar bugs, I would be using this as my primary shop scope, but I have found the screen to be just too small for me. Thankfully, Amazon offered me a full refund on it, and advised me not to return it, so I decided to keep it and this way I have a spare scope to use during this time where I now have no desktop scope until I purchase a new one, however, which one to replace the DSO2D10 with, is something still under discussion in another thread where everyone is being extremely helpful.

Anyway, I thought you all might find this post useful since this discussion is about the Hantek 2Dxx models. and I thought I should share my experience which honestly has not been the best. Hope this helps.

Cheers,

Systops

[nokin]

The lastest firmware + hardware (psu) from manufacturer + modding from DavidAlfa + Schematics from Pcprogrammer make these Hantek become the best scope for electrical hobbies. Mine is best performance now for $120.

Of course for x3 price Rigol or Siglent should do the job better.

[masta_k74]

As I said, I will provide (at least the dual 8V) from a linear power supply. The device is well usable for hobby purposes, but there are clearly better oscilloscopes for small money. I will definitely buy a Siglent SDS1202X-E when I have some money left. Then follows a conversion to Li-Ion battery operation for the Hantek, as a mobile device for outdoor use.

[Russ3000]

I believe that the best solution for a battery power supply circuit would be 2x3 3.7V batteries with an equalizing and charging circuit, they power the analog part of the device with a midpoint, and power the digital one from an isolated DC - DC 9-32V at + 5V 2A, for example URB2405YMD-10WR3 .
A very good device for a hobby.

There was an idea to make a differential input from 2 channels .... I found a 100 MHz bistable relay and a place on the board where to insert it. To do this, it is enough to restart the signal CHANNEL_1_IN_M and connect it to CHANNEL_2_IN_P.

[masta_k74]

I was thinking of a stepdown to dual 9V @600mA and another to 5V 2A. Powered from a 3S3P pack of 18650 cells. For this version I would already have a BMS, a charger with 2A, the 5V - converter and 3 cells lying around here. Only the converter for the dual 9V and 6 battery cells I would still have to buy.

Thus, it is also possible to use the device during charging without any problems. (Of course, you should then ground it properly, otherwise you may get the interference from the SMPS - charger (Y - capacitor against isolated GND in the charger) directly into the oscilloscope and the circuit to be measured, which may not be so optimal.

[DavidAlfa]

2S is enough, providing 8.4V(100%) ... 7.4V(0%)

[Russ3000]

For charging, I use an external power supply with a transformer and there is no need for Y capacitors between the primary and secondary side. I tried the laptop power supply, the interference is small and without grounding.
I wanted to improve the 2x9V power supply, soldered 2.2 uF ceramic capacitors, the noise on the screen increased 10 times I soldered it back. At such frequencies, it is impossible to immediately understand how it will be better.

[DavidAlfa]

Some pictures of the PSU.
Switching frequency is 50KHz, but each pulse creates 100MHz armonics
Last picture is the secondary output before rectifier.

[nokin]

Hi DavidAlfa,
How can we see the effect of overclocking ?
I overclocked @720MHz Cpu and 360MHz Ram but the scope seems to be same as before.

[masta_k74]

@ DavidAlfa

This is the problem with square wave signals. One builds an oscillating circuit super fast without wanting or noticing it. Capacitance of diodes, inductance of conductors, capacitors etc. can cause the strangest effects. This is the reason why I like to work with a linear regulator. Sine signals with 50Hz are much easier to control than a square wave signal with 50kHz and its countless harmonics.
Even with a linear power supply, the diodes can produce high frequency noise. But usually much less.