Maybe the LM358 can't supply enough current? Seeing 100 ohms on the output of an op-amp raises a red flag.
For what it's worth, there's a laboratory DC amplifier circuit on page 274 of The Art Of Electronics, 3rd edition, that uses an LT1010CT as the output buffer for 50 ohm output impedance.

 

I finally realized that my brand new Hantek 6022BE USB scope is only good for +/-5Vpp input, although protected to 35V. Obviously, they are not highlighting this in their sales literature. To measure higher voltages, the probe always needs to be set to 10X and the 10X field selected in the sidebar. Thanks to all here for the advice received.

 

the probe always needs to be set to 10X

It's actually good practice to always use a 10X probe, unless you need the sensitivity at low voltage levels.
It will have less loading effect on whatever you are measuring.

What is your minimum scope sensitivity?

 

Good pointer. DC accuracy is listed as +/-3%. Lots of YT videos showing the scope in action. Good low frequency value for money. Slightly confusing for someone who started out on vacuum tube scopes.

 

A X10 probe can also reduce the chances of damaging your scope. When I first started working in electronics I was supposed to check the noise on a 12V power supply using X1 probe and the scope set to 5 mv/division. I accidentally bumped the 1200 volt supply with the probe tip. That would have deflected the trace vertically about 2400 miles. The 2N708 on the preamp's (Type 1A1) input was dead-dead-dead. Fortunately in those days Tektronix socketed their transistors and I had some of the right transistors at home. I believe a 10X probe would have limited the current enough to save that transistor's life.

 

Here is the final circuit I used, shown as assembled on a solderless prototyping board. Input signal is from PC sound card. It is driving a low resistance load (not shown). Simple, but adequate for my application.

 

Do you have a schematic?

 

As per earlier posting but with a few changed values to reduce the unneeded trimpot range.

 

Much better! Thank you!

Have you considered adding short circuit current limiting to the TIP41?

 

Generally I want max signal current through RL. I ensure via VR1 that the output swings to near 0V so there is minimal DC. Using the same adjustment, the amp can be offset if desired to accommodate a unipolar signal, e.g. from a lab function generator.

I also keep an eye on the ammeter and heatsink temperature. With 12V 2A supply I rarely consume more than an amp. So far, no problems, but your's is a good idea to file away just in case.

For slightly more amplitude, the LM358 can be replaced with a TLC727 or similar single supply, rail-to-rail op amp.