Opamp capacitive load?

[FvM]

I have the values for this circuit, I guess I didn't think it would matter in determining the kind of load...

L202 22uH
L203 8.2uH
C200 82pF
C201 10uF
R200 560-ohms
Q200 2N3906

The component values tell, that the filter characteristic is a 3rd order chebyshev low-pass with pass band ripple < 1 dB. The input impedance is resistive/inductive over the full frequency range with a minimal magnitude of about 400 ohm. So you shouldn't have problems to drive it by an OP buffer. If I didn't miss something, you didn't yet tell about the used OP type nor about the exact problems observed with it.

 

[SkyHigh]

You mentioned you have FPGA and VDP etc. Are you building a configurable digital or software-defined video switch and repeater in one box or something with digitally-tunable filter?

OK, there is no issue if you need to use op-amp. you can still use op-amp as unity-gain buffer, then place the passive filter before or after the buffer.
Then connect to your load. Try it.

My suggestion is for you to use a trimmer for R200 (so you may need to tune it, until you got the right value, then change it to a fixed resistor later)
In video, most of the time, you need some manual adjustment to get it right and it is good to have option to allow that.

I have the values for this circuit, I guess I didn't think it would matter in determining the kind of load...

L202 22uH
L203 8.2uH
C200 82pF
C201 10uF
R200 560-ohms
Q200 2N3906

All the circuits shown above are follow-on stages from the device creating the composite video, which is originally a TM9918A VDP, and now my FPGA-based replacement. I'm just trying to generate composite video that will be compatible with the existing circuitry in all these various computers. The original 9918A has an internal MOS transistor driver with an open source, and expects a 490-ohm external resistor to develop the signal. I could not get a MOS transistor circuit to work for me, so I turned to an opamp since the loads are different in every computer the chip is used in.

Throw normal 75-ohm loads out the window, my circuit is not the end driver going to a monitor. I simply need to develop a linear 0V to 1.92V signal across a variable load, which seems to range from about 390 to 560 ohms *resistive*, but I don't know what the effect the inductors and capacitors will have on the opamp (which is why I posted.)

The first circuit seems to be closest to what the original 9918A datasheet specifies, and looks to be totally resistive, and should not be a problem for the opamp. The other two circuits add the extra inductors and capacitors which complicated the load determination for me.

A basic unity-gain opamp circuit seemed to the be thing I needed, but I made the mistake of reading the datasheet and all the warnings about capacitive loads.

 

[TQFP]

Some questions. Did you try out the recommended reference design given in the datasheet?

Not yet, but I'm doing that today.

You said you couldn't get the MOS transistor cct to work, which one do you mean? The recommended circuit or one of your posted images?

All my posted images are examples of actual circuits that *follow* the one I need to design. The original 9918A's composite output pin would be connected as the input to the circuits posted. I'm reproducing the 9918A with an FPGA, so I need to drive those circuits at least as well as the original 9918A VDP.

The 9918A used a MOS transistor with an open source, I know that from the datasheet and because I have the schematic for the actual chip internals. So at first I tried to just use a similar circuit, but it was getting complicated and growing beyond my available needs (I'm not doing color, i.e. no color burst or phase modulation) and physical space on my circuit board. So, I started looking for a simpler design and an opamp seemed to work nicely.

I have chosen to try an OPA358 (TI part), and I was just trying to make sense of the "capacitive load" section.

 

[TQFP]

If I didn't miss something, you didn't yet tell about the used OP type nor about the exact problems observed with it.

I didn't build it yet. ;-) I just wanted to understand what kind of loads those circuits were presenting so I could select an opamp. I'm wiring up an OPA358 now to see how it goes.

 

[SkyHigh]

I see. FPGA is used to behave like 9918A.

OK I think I know where is the problem. Let's put aside the op-amp.
Did you try to use an external MOS transistor (let it driven by your FPGA) and to make it work in Open Source like the internal MOS transistor in 9918A?
I may be simpler this way as it mimics closer to 9918A.

 

[TQFP]

You mentioned you have FPGA and VDP etc. Are you building a configurable digital or software-defined video switch and repeater in one box or something with digitally-tunable filter?

Nope. I'm just generating composite gray-scale video using two digital outputs and three resistors. Like I said previously, I have the composite part working as long as the load is fixed. But, the load is not fixed, it varies depending on the computer I'm putting my circuit in to, thus I went for an opamp solution to isolate my input from the variable circuits on the output.

My suggestion is for you to use a trimmer for R200 (so you may need to tune it, until you got the right value, then change it to a fixed resistor later)
In video, most of the time, you need some manual adjustment to get it right and it is good to have option to allow that.

I can't modify R200, it is not part of my circuit. It is part of the TI-99/4A home computer motherboard, and will exist in all the 99/4A computers I put my board in to. The other circuits are from other computers that I will also put my board in to, and thus I can not change any of them, they just represent some of the loads that will be hanging off of my circuit, and thus the opamp.

 

[SkyHigh]

Hmm.... I see. You have to tackle different loads (due to different computer to interface with your hardware design)

Why don't you tackle one interfacing computer at a time? Make one work first. Then move on and modify slightly for another computer?

 

[TQFP]

I see. FPGA is used to behave like 9918A.

Exactly. It is done too, and generates VGA just fine, but for some reason I got this idea to ALSO generate composite video like the original 9918A because it would be "easy"... oops. ;-) I found a few examples online using two digital outputs and two resistors. But it relies on fixed loads, which I don't have.

OK I think I know where is the problem. Let's put aside the op-amp.
Did you try to use an external MOS transistor (let it driven by your FPGA) and to make it work in Open Source like the internal MOS transistor in 9918A?
I may be simpler this way as it mimics closer to 9918A.

I tried the MOSFET, I even posted about it:

https://www.edaboard.com/threads/213565/#post903158

After two weeks of getting no where, I fired up LTspice and found that an opamp nicely isolates my input from the output loads (that I can not control), so I decided to try the opamp. Then I read about the capacitive loads... I think I read too much.

The problem I was having with the MOSFET is that I only have 0V and 3.26V to make the needed inputs, and the levels are very specific for the video:

-40 IRE, SYNC
0 IRE, BLANK
7.5 IRE, BLACK (for NTSC)
100 IRE, WHITE

I need to scale that between voltages I can generate with a voltage divider, while applying 0V or 3.26V (logic '1' from the FPGA). I'm going to ignore the "BLACK" level and just use 0 IRE for BLANK and BLACK (like PAL does).

Working out the voltage divider in to a fixed load is no problem, but in to the variable loads was not possible without an active device like the MOSFET or opamp. The problem with the MOSFET was the gate threshold voltage requires a bias on the input which complicates things, and then small range before total saturation. I also noticed that the MOSFETs were not really linear, and over the small range I'm dealing with, it matters.

 

[TQFP]

Hmm.... I see. You have to tackle different loads (due to different computer to interface with your hardware design)

Why don't you tackle one interfacing computer at a time? Make one work first. Then move on and modify slightly for another computer?

Like the original chip, I need the same circuit to work in all the systems where the original chip was used. I'm making a pin-compatible replacement, so different boards for different systems is not an option.

The LTspice simulation was showing the opamp working as a very nice linear unity follower over various resistive loads, and I only started wondering when I started reading about the capacitive loads. I'm thinking I just need to build and test at this point, and see what happens (which I'm doing now.) I just didn't know what kind of loads those existing circuits presented and if it was even going to matter. So I asked.

 

[FvM]

OPA358 and similar rail-to-rail video OPs should be fine to drive the said load. I presume, that the complete low-pass circuit has a bias resistor added at the transistor base.

If you have a negative supply voltage available, a npn/pnp emitter follower combination can also achieve a low voltage shift and moderate output impedance and linearity. But it can't compete with an OP buffer.

P.S.: It may be the case, that the filter circuits are designed for 75 ohm source impedance rather than a low impedance buffer source. In this case, you can connect three separate 75 ohm series resistors between the buffer and each load. Placing a common output resistor or a larger value than 75 ohm would however change the frequency characteristic of the filters and should be avoided in my opinion.

 

[SkyHigh]

OK, mind if I ask if you try placing a 330 Ohm in series at the output of your op-amp buffer and test it out to any of your interfacing load presented by your computers?

 

[TQFP]

I will as soon as I get done wiring it up. Why 300-Ohm specifically?

 

[TQFP]

If you have a negative supply voltage available, a npn/pnp emitter follower combination can also achieve a low voltage shift and moderate output impedance and linearity. But it can't compete with an OP buffer.

Nope, no negative voltage available. The opamp is nice though, it seems very easy to implement, and the parts are small, which is good.