Hey mate, I've overheard that there are good deals on fishing rods at Amazon at the moment.
There are two flavors of the LM4562. One is before National was purchased by TI and the other is after. Within a year of TI purchasing National the fabrication process for the LM4562 was shut down by TI and the LM4562 was moved to a TI process along with other audio op amps. TI's reason was the National was a manual process and TI uses an automatic process.
The older National parts have issues with RF detection at the inputs. Not a resent problem with National IC's. It stems back to the late 70's and possibly longer.
I thought the new process might be more RF immune. That's why I asked.
Again, I don't know which these are. DigiKey says TI but they have National logos. As for RF detection, I added small capacitors everywhere, input to ground, across feedback resistor, input to input, etc. Connecting the 2 grounds with a capacitor did not help either. I don't think it was picking up RF and detecting it but was actually generating RF but I can't prove it. The end result is a single ground wire for the entire preamp section with the signal ground isolated with a 4.7 Ω resistor.
A simple ground wire with a 4.7Ω series resistor sounds like an antenna injecting RF into your circuit.
You can't use thin wires as RF grounds, you need wide conductors that will then have low inductance, and as a result, low impedance at RF frequencies. You can do this with printed circuit board traces, specifically ground planes or at least a copper pour that result in very wide, large "traces".
Two important things to keep in mind:
1) the inductance of a conductor increases according to its length divided by its width or diameter. Long thin wires have high inductance, and inductors have rising impedance with frequency - they work against RF bypassing.
2) the impedance of 1 ounce copper foil is 500µΩ per square. That means that a ground plane of 50 mm x 50 mm will have an impedance of 500µΩ from edge to edge. Similarly, a 1 mm long PCB trace (or an 18 AWG solid wire) will have an impedance of 500µΩ per mm. So, you can have the same 500µΩ impedance across a 50 mm square circuit board, or over 1 mm of wire - obviously the ground plane will be a much better ground at RF than a 2-4 cm wire, because the ground plane's length is low compared to its width.
Do not use a star ground for RF grounding purposes - it will do nothing, and will probably act like an antenna and inject RF into your circuit.
One simple way to remedy the situation is to get a piece of unetched single or double sided circuit board material, and use it as a ground plane. Mount your circuit board above the unetched PCB, and solder each ground connection down to the unetched PCB ground plane using the shortest and widest possible connection (perhaps some copper tape). In this way, your original ground system is shunted to the new ground plane, which will have integrity at radio frequencies.
Any bypass caps used to bypass RF from the input / output cabling can now be shunted directly to the unetched PCB 'ground plane' - ignore the existing ground of your circuit for this use.
You have it reversed. The added resistor did not cause the problem, it ELIMINATED it.
Referring to and thinking about the keantoken diagrams in post #9250:
The amplifiers shown really need to be broken into small signal and output transistors.
Let's look at the current path for the speaker.
Current starts at the power supply, runs though a wire and maybe a rail fuse, then through a power transistor, emitter resistor, and another wire to the speaker terminal. After returning from the speaker, it needs to get back to the power supply ground to complete the circuit. This is a path with high intermittent peak currents and no part of this path should be shared with anything else.
The small signal portion of the power amplifier is relatively constant current and should be connected with separate wires back to the power supply to avoid contamination from the high current speaker loop. The input ground can then be connected either to the small signal portion of the amplifier or the star ground since the connecting ground wire will be relatively quiet.
The amplifiers shown really need to be broken into small signal and output transistors.
Let's look at the current path for the speaker.
Current starts at the power supply, runs though a wire and maybe a rail fuse, then through a power transistor, emitter resistor, and another wire to the speaker terminal. After returning from the speaker, it needs to get back to the power supply ground to complete the circuit. This is a path with high intermittent peak currents and no part of this path should be shared with anything else.
The small signal portion of the power amplifier is relatively constant current and should be connected with separate wires back to the power supply to avoid contamination from the high current speaker loop. The input ground can then be connected either to the small signal portion of the amplifier or the star ground since the connecting ground wire will be relatively quiet.
Hi Fred, now a different approach: Feedback works by comparing the output voltage with the input voltage. In this case, the output voltage comes from both the amp and the speaker. Feedback works best when you connect the sense resistors close to the output, including ground, and equally the load.
Yes, country-of-origin is where the chip is packaged (and tested, usually). It is not where the wafers were fabricated. It's an oddity of the world of international customs and tariffs that the country of origin for a semiconductor is tied to the assembly site rather than the place in which the main value (the silicon chip) was fabricated.
Thanks for the comments, do you think you could post them here where we have had a bit of discussion already:
https://www.diyaudio.com/forums/solid-state/336947-wire-amplifier-2.html#post5769065
The general rule is that 'Country of Origin' is the country were the last significant work on the product is done.
Jan
Thanks. This is the first time I've seen a link to that thread. But, I need to spend time reading (and hopefully understanding) the 10 pages already there before adding any response.
forcing us to buy years old Motorola labeled parts(MJL) & old Sanyo stock from Asia on eBay. The resistor is not the problem. The long thin antenna that you think is a low impedance ground is the problem. If it's 1/4 wavelength long, what is it tuned to? ;-)