Rigol DHO804 Test and Compare Thread

[core]

Personally, I would not change a fan without knowing that the new one does not have at least the same airflow or without comparing the measured temperature with a thermal imaging camera.

These have a single heatsink and can show their internal system temperature on screen. Thermal camera isn't really needed.

Depends. If there is a buck power supply outside the single heatsink and the airflow in the case is too low, an overheat and a fault in this regulator will cause a major malfunction. The voltage at the input can reach the output. I have seen laptops in this situation.

This may not be noticeable in the temperatures monitoring menu.

That's why I wouldn't change the cooling conditions without making sure I'm not missing something.
A functional noisy device is better than a silent "dead" one 

[Martin72]

Then it's really silent.

[Fungus]

That's a bit meagre, actually too little.

Indeed, that's less than I expected. The DHO1000 dialog shows much more info.

Not really. All it says is that "ambient temp" is a degree or two less than chip temp.

What's "ambient temp" anyway? was that screenshot taken in a room at 45 Celsius?

[Fungus]

Depends. If there is a buck power supply outside the single heatsink and the airflow in the case is too low, an overheat and a fault in this regulator will cause a major malfunction. The voltage at the input can reach the output. I have seen laptops in this situation.

A 100mm fan stuck to the back is going to have far more overall airflow than that little fan stuck to the heatsink.

(which isn't directly moving any air in/out of the case at all, it just moves it across the heatsink and convection does the rest)

[ebastler]

Indeed, that's less than I expected. The DHO1000 dialog shows much more info.

Not really. All it says is that "ambient temp" is a degree or two less than chip temp.

What's "ambient temp" anyway? was that screenshot taken in a room at 45 Celsius?

Umm -- when I look at the dialog, I also see temperature measurements for the ADC and (presumably) the leftmost and rightmost front-end amplifier?

The "ambient temperature" readings are obviously not room temperature, but apparently the immediate environment of the respective chips. I have wondered about those too: Did Rigol actually populate NTCs next to each of these four chips? It's the first model using the new chipset, so maybe they wanted to play it safe?

Not my own screenshot, btw. I took it from another thread which discusses tweaking the DHO1000 fan via its PWM control.

[2N3055]

Indeed, that's less than I expected. The DHO1000 dialog shows much more info.

Not really. All it says is that "ambient temp" is a degree or two less than chip temp.

What's "ambient temp" anyway? was that screenshot taken in a room at 45 Celsius?

Umm -- when I look at the dialog, I also see temperature measurements for the ADC and (presumably) the leftmost and rightmost front-end amplifier?

The "ambient temperature" readings are obviously not room temperature, but apparently the immediate environment of the respective chips. I have wondered about those too: Did Rigol actually populate NTCs next to each of these four chips? It's the first model using the new chipset, so maybe they wanted to play it safe?

Not my own screenshot, btw. I took it from another thread which discusses tweaking the DHO1000 fan via its PWM control.

Most chips with any complexity have built in temp measurements. Otherwise, mostly diodes are used as measurement sensors on PCBs.

[ebastler]

Most chips with any complexity have built in temp measurements. Otherwise, mostly diodes are used as measurement sensors on PCBs.

Did you look at the screenshot? For each chip there's an internal temperature readout (obvioulsy from a built-in sensor) plus an "ambient" temperature. It's the latter which we were discussing.

Whether it's a diode or a temperature-dependent resistor, it looks like Rigol went to the effort of populating four of them (plus providing some readout), so they can monitor all chips attached to the central heat spreader. Looks like they planned for a nice, temperature-controlled cooling solution for the DHO1000 and 4000, since PWM control for the fans has also been prepared. Unfortunately it is not enabled in software so far; the fans run at a constant and relatively high speed.

[Randy222]

Depends. If there is a buck power supply outside the single heatsink and the airflow in the case is too low, an overheat and a fault in this regulator will cause a major malfunction. The voltage at the input can reach the output. I have seen laptops in this situation.

A 100mm fan stuck to the back is going to have far more overall airflow than that little fan stuck to the heatsink.

(which isn't directly moving any air in/out of the case at all, it just moves it across the heatsink and convection does the rest)

It does push air out the top and bottom vents on my 804, can noticebly feel air being forced out.
Since it's a 2-wire fan I don't see how the rpm's are monitored, so my goal is to slow the internal fan and use an external 90x15mm to push air in.

[Fungus]

It does push air out the top and bottom vents on my 804, can noticebly feel air being forced out.

Yeah, you're right. There's air coming out under the handle where the heatsink fins are very close to the vent.

There's nothing on the vents at the top though, which is where the "power" components mentioned are.

[2N3055]

Most chips with any complexity have built in temp measurements. Otherwise, mostly diodes are used as measurement sensors on PCBs.

Did you look at the screenshot? For each chip there's an internal temperature readout (obvioulsy from a built-in sensor) plus an "ambient" temperature. It's the latter which we were discussing.

Whether it's a diode or a temperature-dependent resistor, it looks like Rigol went to the effort of populating four of them (plus providing some readout), so they can monitor all chips attached to the central heat spreader. Looks like they planned for a nice, temperature-controlled cooling solution for the DHO1000 and 4000, since PWM control for the fans has also been prepared. Unfortunately it is not enabled in software so far; the fans run at a constant and relatively high speed.

Sorry, didn't look at screenshot 5 posts above..
What you are saying does not contradict with what I said. There are bunch of chips that have on chip and aux external temp sensors. RK3399 has built in 2 ADC channels for temp sensors (all you need is few analog multiplexers and you can have 16ch of temp no problem). Sensors are cheap and it is good that they populated them all. That is always better then too little. I don't know what is your room temp, but temps shown are not very high. Let's hope they provide temp regulation in code, it certainly looks like there is headroom for it on DHO1000...

[Martin72]

I'm a little surprised that the DHO800 only evaluates the CPU temperatures, up to now the software was basically the same, so I would have expected at least the ADC temperatures to be included.

[Randy222]

I may have missed it, what was the end-user fix to get FFT flattop working correctly?

[Martin72]

Apart from what the user Turbotom himself changed, there is still no firmware update that has fixed this bug.
Perhaps rigol doesn't know either, because an MSO5000 user has already reported that it is the same with this model.

[ebastler]

I may have missed it, what was the end-user fix to get FFT flattop working correctly?

TurboTom's fix is to replace the file which contains the window function's weight coefficients. Described here with step-by-step instructions.

[Noy]

I'm in trouble..
Should i get a DH804 (hacked) or a DHO1074 mated to my MSO5354?
I know there is the discount for the DHO1074 currently, but still then it is nearly 1/3 more on top..

And my desk is space limited.. The size of the DHO804 is really nice / fitable ..
And i think it is enough for a MSO5k addition..? I only need / would like to have it for power supply checking / using with my current clamps..
The stuff where 12bit are nice (but do i really need it ?? )... For my main work (mostly digital stuff) i like the >350MHz BW / 8GS.

[2N3055]

I'm in trouble..
Should i get a DH804 (hacked) or a DHO1074 mated to my MSO5354?
I know there is the discount for the DHO1074 currently, but still then it is nearly 1/3 more on top..

And my desk is space limited.. The size of the DHO804 is really nice / fitable ..
And i think it is enough for a MSO5k addition..? I only need / would like to have it for power supply checking / using with my current clamps..
The stuff where 12bit are nice (but do i really need it ?? )... For my main work (mostly digital stuff) i like the >350MHz BW / 8GS.

DHO1074 .

7" screen is really tiny...

[Noy]

Sure but i can use my PC Monitor / i'm using a HDMI Grabber on my 5k for documentation purpose..
And i can use the webserver.

The 8" from MSO5k are enough.. My Hameg HMO3524 which maybe has to leave (shelf space for DHO804) has less than 7".
So Display size is not a concern for me i think..

[2N3055]

Sure but i can use my PC Monitor / i'm using a HDMI Grabber on my 5k for documentation purpose..
And i can use the webserver.

The 8" from MSO5k are enough.. My Hameg HMO3524 which maybe has to leave (shelf space for DHO804) has less than 7".
So Display size is not a concern for me i think..

So you already know what you want but want us to confirm it?

DHO1000 is class above DHO800 in every respect. If you really want a scope from DHO series (and no other type/brand) DHO1000 is much more for your money, at the price it is now..

[TurboTom]

+1 And since the Black Friday offers resulted in many new owners of this scope, a new "hacking wave" is underway which may present us with quite some new, cool features soon that so far were only available to the DHO4000 owners (fingers crossed...  ).

[Noy]

Not really a "confirm"...
The question is, is it worth to spend 1/3 more money on top to get the DHO1000 if i already have a MSO5k as main scope..
The DHO800/DHO1000 would only be used for switching regulator / in combination with a CP2100A..
My question is, are 12bit are worth enough to get a 2. scope...?

I think 12bit are the main advantage.. anything else should better on MSO5k ? For HF / FFT / Bodeplot i have a Siglent SVA..


And regarding DHO1k hacking.. I think unless you get a spare ADC everything is squeezed out of the DHO1k until now..
And it looks like DHO1k is phasing out? So maybe it will be dropped from rigol? DHO800 will be a better "cash cow"?


Two points pro DHO1K are the 14/16bit mode which the DHO800 not has?
And the 10MHz ref in is nice to be used with my GPSDO but would this be a big advantage?? I don't know.. on the MSO5k it would probably more helpfull..?

[ebastler]

I only need / would like to have it for power supply checking / using with my current clamps.

If you want to potentially do more in-depth work there: Neither the DHO800/900 nor the DHO1000 series offer Rigol's Power Analysis option. For the DHO1000, its big brother (the 4000 model) does support Power Analysis, and the "advanced hacking" approaches TurboTom mentioned should be able to unlock that option on the DHO1000 too. This is however work in progress, and there is no guarantee that it will be made available as a ready-to-use hack.

[ebastler]

And regarding DHO1k hacking.. I think unless you get a spare ADC everything is squeezed out of the DHO1k until now..
And it looks like DHO1k is phasing out? So maybe it will be dropped from rigol? DHO800 will be a better "cash cow"?

The recent hacking approach (patching some "which functions are supported here?" calls in the Auklet Android app) has been able to unlock the 50 Ohm inputs and other features normally reserved for the DHO4000, without the unwanted side effects of assuming the dual-ADC configuration of the 4000.

There is some speculation that the original DHO1000 series might be phased out, based on the large discounts currently offered. But reassuringly, the DHO4000 as well as the DHO120xU (recently introduced in China) share the same firmware. So I would hope that future maintenance/upgrade releases will be made available for the DHO1000 tooo, even if it should be discontinued.

Two points pro DHO1K are the 14/16bit mode which the DHO800 not has?
And the 10MHz ref in is nice to be used with my GPSDO but would this be a big advantage?? I don't know.. on the MSO5k it would probably more helpfull..?

If you want to use the DHO as your "precision scope", the 10 MHz reference might make sense. The 14/16 bit modes (with significantly reduced bandwidth) certainly do.

There are various additional aspects where the 1000 series is ahead of the 800/900; whether they are important to you, only you can decide: Separate EXT trigger input, 200 MHz usable in 4-channel mode (at 500 MSa/s per channel), higher waveform update rate, faster boot time, automated measurements in cursor-defined regions...

For me personally, the larger form factor is an important argument in favor of the 1000. It is not just the absolute screen size, but also its higher resolution which leaves more spave for "net content". For my taste, the relatively large control elements and window frames leave too little room for trace display on the DHO800. Also, I prefer the built-in power supply, but realize that others might see the external USB-C power as an advantage.

[Noy]

The recent hacking approach (patching some "which functions are supported here?" calls in the Auklet Android app) has been able to unlock the 50 Ohm inputs and other features normally reserved for the DHO4000, without the unwanted side effects of assuming the dual-ADC configuration of the 4000.

OK that was new to me. I thought 50Ohm input wasn't working on DHO1k even with hacking..

[Martin72]

I have started to "play through" the Batronix demo board again with the rigol.
I will probably upload the results to googledrive.
I already have a few pictures in the box today.
Note to self: In the pass/fail menu do not activate the screenshot at fail, but save it yourself - otherwise the save message will be included in the picture.
DS17: Demo Acquisitionmode Normal/Average (16x)
DS18/19: Demo Acquisitionmode Normal/Peak
DS20: Demo cursors
DS21-DS24: Demo measurements, several modes
DS25: Demo persistence 1sec
DS26/DS27: Demo Pass/Fail

[Martin72]

I had to wait for dinner a while ago, so I did a little noise measurement on a channel when there was nothing on it.
Once with 50ohm termination, once blank 1Mohm, full bandwidth (200Mhz), 20Mhz limited.
Then once at 1µs/div, 1ms/div.
It really is a Rigol scope, it's hard to believe given the values if you had an MSO5354 like I did back then...
(open a pic in a new tab, then zoom in for better viewing the values)