Chinese Function Generator limitations

Thread Starter

SamR

Joined Mar 19, 2019
3,807
While trying to set up a Triangle or Sawtooth wave with a 12V amplitude from Zero I discovered that my chinesium function generators won't do it! First off, they don't have Sawtooth as a wave pattern. May be possible to program it as an arbitrary wave but not easily done on the fly. Secondly, they have a 20V max amplitude, OK there. But when you try to set a 12V amplitude with a 6V offset so it will be from 0 - +12V it clamps the amplitude to 8V. Best I can get out of them 8V amplitude using a 4V offset to set the bottom of the wave on zero. The only way to get +12V to 0 is to let them run the standard +12 to -12 full wave across zero. You can do that, but the timing may be off depending on what you are doing with that signal.
IMG_0848[1].JPG
1636826325684.png
 

Thread Starter

SamR

Joined Mar 19, 2019
3,807
Compared to some older more reputable makes, the specs aren't that bad. It has an external wall wart supply without a whole lot inside the unit. Basically a power switch for the small motherboard, the OLED display and the membrane switches and pushbutton on the front all in an unshielded plastic case. Having noise problems with my scope so I took my little pocket scope with about 18" of wire on the probe and started sweeping around for the source. First thought it might be the wireless dog fence RF transmitter, but it isn't. Checked around the LED bulbs and finally started checking around my bench. Noise is coming from the Cleqee even when it is "turned off". The wall wart power unit is always on and coming into the Cleqee even when the unit is off and making noise from its switch mode AC/DC voltage converter that has no shielding. Wish they had at least shielded and grounded the source and the unit. The older model is basically the same except for the display and only 15MHz, but not making noise or at least masked by the newer unit's noise.
 
Last edited:

KeithWalker

Joined Jul 10, 2017
2,091
It's a real shame that they cut corners like that to save a few pennies. I read the spec and it does appear to be a very versatile signal source.
 

Marley

Joined Apr 4, 2016
444
I have one of these like the lower unit. Cheap - for a reason!
The distortion on sine is terrible - especially at higher frequencies.
OK as a rough and ready signal source. No good for testing audio.
 

Thread Starter

SamR

Joined Mar 19, 2019
3,807
Yes, they do give out on their top end, that I did test. Somewhere around 80-90% of MAX frequency. Never noticed any distortion on low end or in the audio range.
 

sparky 1

Joined Nov 3, 2018
645
I am not sure about cleque, In feel tech it's trapazoid/rise sort of hidden.
I use the instek sfg-1013 a DDS, good for audio has -55dB sine, atten, The Voltage Freq display offset and sweep adj knob
is intuitive for me. A smooth quality constant current derived ramp can make huge difference that can vary from one application to another.
Often the type of adjustment needs to be tailored to an application as a dedicated circuit. At 6:30 He makes a sweep generator using older function generator. In video #104 he describes some of the less known aspects of sawtooth generators.
 
Last edited:

Thread Starter

SamR

Joined Mar 19, 2019
3,807
Already came to the conclusion that the only way to avoid the SigGen limitation was to build a signal source to fit my needs. W2AEW has some pretty good vids out and is a good and seemingly trustable source of ideas. I built one of the SigGen kits years ago but don't remember its specs and it is long gone anyway. Most of the kits run off of a 9V battery so limited to ±7 to 8V. Anything more than that is going to take a transformer PSU.
 
Last edited:

bloguetronica

Joined Apr 27, 2007
1,481
I suspect that changing the output op-amp could mitigate the clamping. But there are some questions:
- Have you checked if the promised specs are valid if and only if the output is open? Signal generators have a 50Ω output impedance, which is a resistor of the same value that is to be in series with the load. This is to avoid reflections by matching the load impedance. Correct specs should come from measures with 50Ω loads, and should state that. Probably the specs are wrong.
- Have you opened the unit and checked the rail voltages? Probably the supply in the unit doesn't cut it.

If points one and two check OK, then it is probably the output op-amp (or one in the chain) that is cheap, or a cheap version of what should be there. Voltage swing, as well as frequency response and slew rate can be limiting factors.
 

Thread Starter

SamR

Joined Mar 19, 2019
3,807
I can't get to the wall wart behind my bench but it is most likely a 12V one. That is the external power supply for the unit. Which is why the outputs are limited to ±10V. The SigGen has a Max Amplitude of 20V, however that 20V is between ±10V and cannot exceed that limitation! I was expecting it to do 20Vpp with a 10V offset to set that at 0-20V, but it won't do it because of the ±10V limitation. Any offset reduces the 10V, so if I use a 4V offset I end up with a MAX 12Vpp wave from -2 to +10V. Live and learn...
 
Last edited:

bloguetronica

Joined Apr 27, 2007
1,481
I have a suspicion that the generator itself has some sort of charge pump to generate the negative rail, internally. Or else the math doesn't add up. It has to have a 12V rail and a -12V one, besides ground. From rail to rail, it should have 24V, which is enough to be used to drive an op-amp in order to generate a 20Vpp signal.

Anyway, the offset limits the amplitude of the output signal, in the sense that there is less usable swing to keep the waveform unclipped, the more the offset deviates from zero (or center). That is to be expected. Also, +/-2V from the rails is pretty normal, unless the op-amp is meant to be a rail-to-rail variety. Then, you should be able to get to 500mV near the rails, depending on the output current. However, those op-amps have other issues. In a nutshell, I see nothing wrong with your generator. It is just the specs that are not well specified.
 
Top