mertkan65
Junior Member level 3
- Joined
- Nov 24, 2004
- Messages
- 30
- Helped
- 0
- Reputation
- 0
- Reaction score
- 0
- Trophy points
- 1,286
- Location
- Ankara/Turkey
- Activity points
- 279
Hi,
we can´t verify your layout, because you don´t show it.
It seems your input signal is distorted somehow. It is no clean sine.
I wonder: you run the amplifier with 140kHz , but the R1/C37 cutoff frequency is at about 80kHz.
You will not see max. gain (max amplitude).
For stability reasons I´d rather connect C37 directely to OUT of the LTC6090...but maybe this increases output distortion...
On the other hand then you could reduce it´s value.
Klaus
I would first assume that there is a wrong value placed somewhere in the layout of the last amp. I am assuming that you have eliminated a bad transistor as the culprit. I am also assuming that the layout for the last amp has been properly checked and verified.
I have got similar waveform before and I found that something is oscillating. From your figure it seems that the output has been modulated, and you can use the DFT to extract the frequency information. In addition, if something other than VCO is oscillating, the DC condition may be apart from the normal condition.
Maybe some parasitics (R or C) caused by PCB traces, that are different for this channel, versus other channels?
Hi:
As oscillation does not require an AC input, just short the input to a DC value, and measure the output spectrum, if it really oscillates then you can see clearly the oscillation frequency of the opamp. As for DC condition, what I mean is that you give the same voltage bias to the opamp then check the current. If the chip is oscillating then the current may be different from that of the normal amplification condition.
I believe that your issues are parasitic and layout related. Until you post a layout, I don't believe that you can get the help that you need.
I believe that your issues are parasitic and layout related. Until you post a layout, I don't believe that you can get the help that you need.
Agreed. Parasitics or something like that caused by layout (difference between different channels).
Matching parasitics is not easy, and we need to see the layout for that.
Hi,
Most users are happy with PDF.
Klaus
Hi:
As oscillation does not require an AC input, just short the input to a DC value, and measure the output spectrum, if it really oscillates then you can see clearly the oscillation frequency of the opamp. As for DC condition, what I mean is that you give the same voltage bias to the opamp then check the current. If the chip is oscillating then the current may be different from that of the normal amplification condition.
Hi,
Which of those amplifiers is the problematic one?
Klaus
We use cookies and similar technologies for the following purposes:
Do you accept cookies and these technologies?
We use cookies and similar technologies for the following purposes:
Do you accept cookies and these technologies?