DS1104Z-S Plus Signal Generator, Have I been Had?

[blazin_penguin]

Hi All,

So I just got a nice Rigol DS1104Z-S Plus with the signal generator.   and I'm wanting to test out PWM on a fan that requires  0-5V PWM, 1KHz - 25 KHz input.  0 - 100% PWM. 

Now These are the specs for the Signal generator amplitude


(Here's the link to the whole specs: https://beyondmeasure.rigoltech.com/acton/attachment/1579/f-0504/1/-/-/-/-/MSO1000Z_Datasheet.pdf)
I would expect that means that I can have a 5V_p-p pulse signal, and then offset is by 2.5V to get my 0-5V signal, but once I set the amplitude to 5V_p-p it tells me the "Parameter Limited" when I try to adjust the offset. 

I have double checked that I'm in High Z mode, Am I being dumb and missing an obvious setting here?  This feels like something a signal generator should be able to do (I know my Keysight one can do it...)

Thanks All!

[Fungus]

The spec says it can only do that in HighZ mode. I suspect you're not in that mode.


It's never going to work though: Fans need amps to drive them, there's no way you can expect a signal generator to drive a fan directly.

HighZ mode in particular expects a very high resistance on the output (that's what "HighZ" means)

[TomKatt]

Seems like the Hi-Z output limit of 5Vp-p would be a +/- 2.5V signal.  I note the manual states When the impedance is HighZ, the range is from (-2.5 V + the current amplitude/2) to (2.5 V - the current amplitude/2); when the impedance is set to 50 Ω, the range is from (-1.25 V + the current amplitude/2) to (1.25 V - the current amplitude/2)..  Also, in the DC Output section it states When the impedance is HighZ, the range is from -2.5 V to +2.5 V; when the impedance is set to 50 Ω, the range is from -1.25 V to +1.25 V..

How or why you'd apply a 2.5V offset to an output only capable of +/- 2.5V seems odd to me, but I don't have any experience with this gear so there must be some rationality behind it.

Anyway, I think it's possible that you might be stuck with a 2.5V PWM signal.  Hopefully I am wrong because that seems illogical.

[TomKatt]

It's never going to work though: Fans need amps to drive them, there's no way you can expect a signal generator to drive a fan directly.

HighZ mode in particular expects a very high resistance on the output (that's what "HighZ" means)

I'm thinking he's working with a fan that has a dedicated PWM input for speed regulation, like most pc fans.  The motor power is provided by a dedicated power input.

[hexreader]

I would expect that means that I can have a 5V_p-p pulse signal, and then offset is by 2.5V to get my 0-5V signal

Your expectations will not be met by Rigol DS1104Z
Output will never go above +2.5V or below -2.5V
With no load, you can have 5Vpp with no offset, but not 0 to +5V
You are out of luck
TomKatt guessed right

[TomKatt]

Not that the DS1104Z-S is a bad scope, but issues like this are why I prefer dedicated gear over all-in-one types. 

I’ve been tempted to add a sig gen to my bench (replacing the old home built analog box I use now) and I’m saving up the few extra bucks to buy a dedicated box over the Siglent scope add-on AWG…. There’s usually some kind of limitation in the built-in options, plus you can upgrade more easily when the time comes.  My DSlogic analyzer is easier for me to use than the similarly priced Siglent scope MSO addon - more features, better decoding and longer capture memory.  If I need to synchronize with an analog signal I can use the trigger out to trigger the scope or vice versa.

[bdunham7]

I have double checked that I'm in High Z mode, Am I being dumb and missing an obvious setting here?  This feels like something a signal generator should be able to do (I know my Keysight one can do it...)

High-Z mode doesn't actually affect the output of the generator, it just adjusts what it shows you.  The generator has an output impedance of 50R, so if you put a 50R load on it the voltage gets dropped by half. 

It's not an obvious solution, but if your fan circuit is not grounded at all (battery operated on your bench) then you can get your 0-5V signal by setting generator CH1 to your 5V_p-p square wave and CH2 to -2.5V offset.  Then use appropriate connectors and use the output of CH2 as your "ground" and CH1 as the PWM input.  Don't connect anything from ground or the BNC shells to the fan circuit.  Don't do this if your fan is connected to anything other than a totally isolated, battery-powered simple circuit.

[TomKatt]

OP might get lucky and the 0->2.5V pulses are enough for the fan's controller logic to see...  Sometimes if the signal amplitude gets over whatever the threshold value is logic gates can still work.  Just recently I was working with serial ports and got my ttl level com port to communicate with a VFD expecting 'real' +/- 12V (or whatever serial ports are).

Worth a shot to try.

[Aldo22]

This is why many “real” signal generators have 20Vpp. That sounds like a lot, but it is only max. +5V at 50Ω.
At higher frequencies, perhaps not even that.

[blazin_penguin]

Thanks Everyone,  I guess this was a RTFM kind of issue because it is there in the Manual (if not the Spec sheet) I guess I should have pushed harder to get my Boss to approve the Keysight. Oh well, the Scope on it is still pretty decent.  I'll just put this in my back pocket of experiences for more fuel to the "This is why you buy good quality instruments and not just the cheap ones" That 2.5V - 1/2 amplitude should really be spelled out in the datasheet.  Keysight actually spells out the formula for the offset in their datasheet. (and has a 20Vpp generator)

It's never going to work though: Fans need amps to drive them, there's no way you can expect a signal generator to drive a fan directly.

I'm thinking he's working with a fan that has a dedicated PWM input for speed regulation, like most pc fans.  The motor power is provided by a dedicated power input.

That is correct, The fan has a dedicated PWM control on it. 

OP might get lucky and the 0->2.5V pulses are enough for the fan's controller logic to see

The spec sheet for the fan says it wants min 2.3V for V_H and I did try it, unfortunately no dice.

It's not an obvious solution, but if your fan circuit is not grounded at all (battery operated on your bench) then you can get your 0-5V signal by setting generator CH1 to your 5Vp-p square wave and CH2 to -2.5V offset.  Then use appropriate connectors and use the output of CH2 as your "ground" and CH1 as the PWM input.  Don't connect anything from ground or the BNC shells to the fan circuit.  Don't do this if your fan is connected to anything other than a totally isolated, battery-powered simple circuit.

I think I'll just move on to using the PWM from my Raspberry Pi Pico instead of risking the horrible consequences if anything in that connection goes wrong :-)  Or bring my Keysight in from home.

Not that the DS1104Z-S is a bad scope, but issues like this are why I prefer dedicated gear over all-in-one types.

Yeah I normally agree, but I don't have enough use for one to get a whole other instrument,  this was just part of the "Playing around" stage of getting this fan up and working. and now I'm wishing I hadn't spend the extra $200 US on the one with the signal generator, because I'm not certain what I'm going to be able to use it for.  most of what I deal with is PWM like what I'm doing here.

[TomKatt]

most of what I deal with is PWM like what I'm doing here.

In your case, a simple Pi / Arduino based PWM generator might be sufficient for your needs.  I bet you could even find a dirt cheap PWM module board.  These kind of things are handy to have on the bench - often I will grab some cheap box over setting up a 'real' piece of gear lol.

Another lesson learned and more experienced gained - lord knows I've learned a lot by similar experiences ;-)

Best of luck!

[tautech]

Ya got 2 options, engage the forum with your requirements asking for advice OR learn how to read and fully understand datasheets and manuals.

[bdunham7]

I think I'll just move on to using the PWM from my Raspberry Pi Pico instead of risking the horrible consequences if anything in that connection goes wrong :-)  Or bring my Keysight in from home.

Another option is to get a low-cost amplifier for it.  These have the addtional advantage that they are low-impedance output (not 50R) so they can be used to run a variety of things and even as a high speed programmable power supply.  They vary in gain, bandwidth and power so look at the specs this time!  You'll want one with a gain of at least 4, preferrably 10, so the popular Juntek models with a gain of 2 won't work.  You can see this by searching for "DDS amplifier", "Juntek amplifier", "Feeltech amplifier" or "FPA2000". 

[Aldo22]

Perhaps a so-called "process calibrator" would be something for you?
They are handy, versatile and there are models available for around $50.
The picture shows an SG-003A. Of course there are better ones.

[Fungus]

Just get an Arduino...

[blazin_penguin]

In your case, a simple Pi / Arduino based PWM generator might be sufficient for your needs.

Yeah, that's what I'm doing I just always like to start by using test equipment to make sure that if it doesn't work I know if the problem is (in this case) the fan, or my programming,  but this will in no ways actually slow me down.  other then the time I spent trying to get the wave gen to output 5V.

Ya got 2 options, engage the forum with your requirements asking for advice OR learn how to read and fully understand datasheets and manuals.

I'm sorry I didn't read every page of the manual to find the specification that they appear to be intentionally misleading about in the spec sheets (because other manufacturers actually give the formula in theirs) 

Another option is to get a low-cost amplifier for it.

I may look into that in the future.

[TomKatt]

In your case, a simple Pi / Arduino based PWM generator might be sufficient for your needs.

Yeah, that's what I'm doing I just always like to start by using test equipment to make sure that if it doesn't work I know if the problem is (in this case) the fan, or my programming,  but this will in no ways actually slow me down.  other then the time I spent trying to get the wave gen to output 5V.

For $5, I'd just get something like this - https://www.amazon.com/Module-Generator-Pulse-Frequency-Modules/dp/B0CBND5LB4

How these can sell for $5 shipped to your door is beyond me...  There's not much to it, but still - controller, lcd, buttons, pcb...

[Fungus]

I'm sorry I didn't read every page of the manual to find the specification that they appear to be intentionally misleading about in the spec sheets (because other manufacturers actually give the formula in theirs) 

5 Vpp into a high impedance load.
2.5 Vpp into a 50 Ohm load

What's misleading about that?

What's the impedance of your fan's control signal? That's the only missing into. I'd have guessed it was "high impedance" but apparently not.

[TomKatt]

I'm sorry I didn't read every page of the manual to find the specification that they appear to be intentionally misleading about in the spec sheets (because other manufacturers actually give the formula in theirs) 

5 Vpp into a high impedance load.
2.5 Vpp into a 50 Ohm load

What's misleading about that?

What's the impedance of your fan's control signal? That's the only missing into. I'd have guessed it was "high impedance" but apparently not.

In fairness, if I looked quickly I might assume that 5Vpp might get me 0 -> 5V, which is what the OP needed.  The PWM input on the fan is a HIz input, so if the specs allowed a ground based reference it would have worked for him.  As it turns out, the sig gen output only allows a max of +/- 2.5V no matter the setting or environment.

[Fungus]

In fairness, if I looked quickly I might assume that 5Vpp might get me 0 -> 5V, which is what the OP needed.

I guess if you combine it with "2.5V offset" you might expect that.

It's probably all done digitally though: +/-2.5V is the output range of the DAC and everything else is just math (scale/offset).

[Fungus]

The PWM input on the fan is a HIz input, so if the specs allowed a ground based reference it would have worked for him.

Are you sure it doesn't have a pullup/pulldown or something in case that wire isn't connected?

A 10k resistor in there would probably be enough to drag down the AWG.

Next Question: What voltage range does the AWG show if you connect it to the 'scope via 10x probe and set a 2.5V offset?

[TomKatt]

A 10k resistor in there would probably be enough to drag down the AWG.

Next Question: What voltage range does the AWG show if you connect it to the 'scope via 10x probe and set a 2.5V offset?

That's a fair point - I suspect the PWM input is treated as a cmos logic pin.  Though I'd guess there might be a pull up resistor rather than pull down.

If it were me, I'd be checking the sig gen output on the scope as well.  Though the manual and earlier discussion seem to agree that the output cannot go beyond 2.5V in either polarity.

[bdunham7]

A 10k resistor in there would probably be enough to drag down the AWG.

It should have a 50R output impedance and 10k should have minimal effect on the output voltage.

I think it is just limited to a voltage source of +/-2.5V peak voltage followed by a 50R resistor.  This limits output current to 50mA which makes it a lot cheaper to implement.  DC offset isn't a separate circuit and can't add to the peak voltage.  IMO it's cheap and lame, like most auxiliary AWGs included in scopes. 

[Fungus]

IMO it's cheap and lame, like most auxiliary AWGs included in scopes.

It's designed for industry-standard 50 Ohms.

[bdunham7]

It's designed for industry-standard 50 Ohms.

?

That part I'm sure is true, it's the low power output that is cheap and lame.  The one in my Siglent scope isn't much better, thats 3V max and a 24ns rise time.  I wouldn't pay extra for it and would prefer some other additional feature, like a reference clock input, WIFI or HDMI out.