Stage amp coupling. How to?

[JhonStan]

Hi guys, to use these two amplifiers (the first black and the second as final one silver) should use some type of coupling (resistor / series or parallel capacitor) or simply connect them one after the other directly? Because I tried the direct connection but with the oscilloscope in the output stage (silver with 50/75 Ohm load) I get a very distorted wave and I'm not sure it's just excessive gain. Thanks!

[TimFox]

I can't tell what's in the lower amplifier (contained in the metal box), but the upper amplifier obviously has an input coupling capacitor and an output coupling capacitor on the circuit board between the active device and the SMA connectors. 
(Some amplifiers have DC continuity at their input, such as a transformer primary, and require a coupling capacitor so as not to upset the DC quiescent voltages on the amplifier driving them.) 
Why do you think the distortion is not caused by too much signal (rather than too much gain) from too much input to the first amplifier? 
The full undistorted output capability of the first amplifier may well be too high for the second amplifier's input, in which case you need to reduce the input power to the first amplifier.

[JhonStan]

Now that I look at the black amplifier it has the input caps that at the output so I know it could couple to the silver one directly with a cable (I could take some photos but it is built so that when I open the panels of the connectors where there are the risk of disconnecting the connections because soldered from pcb to connector with a rigid wire). I lowered the input signal with a voltage divider formed by a resistor and a potentiometer to get lower voltages than the minimum I can get from the voltage generator which in this case (I have to check but I remember this) was at a 100mV sine wave at 10Mhz. The divider had R1 1KOhm and R2 100 Ohm potentiometer in order to have a maximum voltage of about 10mV then turning the trimmer I would have lowered the divider output voltage. Once this is done, I have always seen a distorted output sinuisoidal.

Here the voltage divider schematic used:

[TimFox]

What happens when you drive the larger amplifier directly or through your attenuator network into the 50 ohm load at your oscilloscope?  How much input voltage can it take before the output looks distorted?

[JhonStan]

The wave is distorted and this could also be the input signal too high (don't worry me for the moment) but also unstable. With the oscilloscope I connect the first channel to the signal generated by the sine wave generator and the second channel of the oscilloscope to the output of the amplifiers (the amps measured one at time). Obviously the signal of the second channel comes out amplified but the wave is not clear and goes back and forth. To me this feels like instability. Maybe I try to take a video or photo that I could be clearer.

[TimFox]

Sometimes, coupling between the two channels of the oscilloscope can cause instability or even oscillation when connected to the input and output of an amplifier.
Try this:
Connect only the amplifier output to one channel of the oscilloscope.
Drive the amplifier (one only) through your attenuator and vary the input signal until the output distorts.  Note the output level where distortion starts.
Then, connect that input signal directly to the oscilloscope (no amplifier) to estimate the input signal for distortion.
If the output is always distorting, even at a power far below its power rating, then that amplifier has a problem.

[JhonStan]

Sometimes, coupling between the two channels of the oscilloscope can cause instability or even oscillation when connected to the input and output of an amplifier.
Try this:
Connect only the amplifier output to one channel of the oscilloscope.
Drive the amplifier (one only) through your attenuator and vary the input signal until the output distorts.  Note the output level where distortion starts.
Then, connect that input signal directly to the oscilloscope (no amplifier) to estimate the input signal for distortion.
If the output is always distorting, even at a power far below its power rating, then that amplifier has a problem.

Hi,

i've done some measurement of the amp stages.

Small black
Maximum and minimum voltage to avoid distortion of the sine wave @10Mhz .
The measurement was made by connecting the amp output to the oscilloscope input.
Input 15mV - Output 64mv
Input 100mV - Output 452mv
10 db of gain is declared but it gains 4 times so it seems little to me.

Silver amp
Maximum and minimum voltage to avoid distortion of the sine wave @10Mhz
The measurement was made by connecting the amp output to the oscilloscope input.
Input 10mV - Output 5,2V
Input 5mV - Output 2.3V
Input 2mV - Output 700mV
Connecting in parallel a 50Ohm load to the output towards the oscilloscope.
Input 10mV - Output 1,2V
Input 40mV - Output 5,2V
Input 50mV - Output 6,4V
This amp appears to be working fine.I have not used the voltage divider because especially with very low voltages I have seen that the sinusoidal is disrupted (I attach photos). The function generator probably has some spurious adding to that of the divider resistor.I was a little disappointed by the fact that especially in the small amplifier it has little input sensitivity. I maybe is due to the fact that there is this background noise.

[TimFox]

A voltage gain of 4 V/V is a power gain of 4² = 16 W/W (assuming equal impedances at the two measurement nodes, which is 12 dB.
That variation from one period to another period on the 'scope face indicates a low-frequency component, not synchronized with the sine wave, so that each repetition displayed on the screen is slightly displaced vertically by the low-frequency component.

[JhonStan]

Could this low frequency component be a spurious generated by the power supply stage of the DDS generator, perhaps poorly filtered? Because there are no other signal sources besides that. I am that someone has changed the power stage because it was bad.

[TimFox]

That could be.  I have often seen similar displays when there was a lot of 60 or 120 Hz "hum" either at the generator output or picked up by the connecting cables.

[JhonStan]

I try to check the power supply stage and see if I find anything. The model is FY6800 also discussed on https://www.eevblog.com/forum/testgear/fy6800-power-supply/

[TimFox]

This is an example of when an analog CRO will give a better result than a digital DSO.
With an analog scope, where aliasing is not a problem, change the time base to one suitable for 50 to 120 Hz and set the trigger to "Line".
(With such a slow time base, the DSO sampling will probably give an alias on the display, since it is undersampled.)
The high-frequency sinewave will turn into a solid ribbon, but if you have superimposed low frequency that is line-synchronous, it will be obvious on the top and bottom edges of the ribbon.
It is possible that the low-frequency content will be a modulation, rather than an addition, but that can also be seen on line trigger.

[JhonStan]

I have another question if possible. Usually for the input these amplifiers work with loads of 50 / 75ohm. The question is: What happens if I use (in input) lower loads (eg 5/10 Ohm) do I burn something or reduce the input bandwidth? The question arises from wanting to use a probe that has very low resistance. So I would like to understand if in series with this probe I have to add a resistance to get to 50/75 Ohm total. I did some tests by simulating a 5 Ohm input load and injecting the 10Mhz sine signal at the input but I think the load is too low. Thanks!

[TimFox]

Please clarify:  to which port of the amplifier in question do you connect the 5 ohm resistor?
Usually, placing 5 ohms in parallel with the input port should not damage the amplifier, but 5 ohms connected to the output port could damage the amplifier.
Usually with a generator, the 50 ohm source impedance results from a series resistor from the actual generator to the output connector, and will not be damaged with a low-resistance load (although the voltage decreases accordingly).

[JhonStan]

I mean only for input.



TIM The problem is precisely this. The input signal is very low and if I add a resistor to reach 50 ohms, risk of damping the signal. For this reason if I can avoid increasing the incoming load is preferable. The alternative is to use an open circuit (no mass input signal) input but I don't know if it works.

[TimFox]

The 5 ohm resistor at the input will reduce the signal from a 50 ohm source (such as an RF generator) in a calculatable manner, but should not cause any problems.  If driven by another amplifier, however, the 5 ohm input shunt resistor may cause a problem with the driving amplifier.

[JhonStan]

If it reduces the signal it would be better for me to increase the input load somewhat without interfering with the input sensitivity of the amp stage. Maybe i need a first stage bye fet. These above are all at BJT I think.

[TimFox]

What are you trying to do here?  Do you need a higher load impedance for the source than the amplifiers 50 ohms?
Adding a resistor to a circuit may be needed for a particular purpose, but it doesn't improve the gain or signal/noise ratio.

[JhonStan]

A species of very low signal repeater with a tuning circuit like that of the crystal radio. The load (input probe) is the antenna only that instead of having a 50ohm antenna is 5ohm. Initially I thought that having less Ohm the signal entered stronger and being very low signals I thought that by inserting a 50ohm resistance could further lower the input signal.