Note that op amps also tend to oscillate with capacitive loads.
They are not a magical way to eliminate oscillations.

 

Gosh, I haven't used one of those since 1975! Used to have a pair in push-pull at 144MHz pumping out about 400W...

I've never used them at all, never built anything with valves in but have been thinking of playing around with some just to get the knowledge, I can't tell you how many valves I smashed as a kid so they'd go "pop" in the street, I likely destroyed some rarities too - like an idiot.

 

I've never used them at all, never built anything with valves in but have been thinking of playing around with some just to get the knowledge, I can't tell you how many valves I smashed as a kid so they'd go "pop" in the street, I likely destroyed some rarities too - like an idiot.

Electrically, not that difficult. Mechanically, the forced air cooling was a pain, the components mentioned in the data sheet were way outside my means as a teenager, though I was lucky then to find some ex-Army surplus parts. All too often the tubes had been pulled from ex-military kit but the housings were scrapped... it was the housings that were valuable!

Here's an example that is quite similar to what I built....

 

Electrically, not that difficult. Mechanically, the forced air cooling was a pain, the components mentioned in the data sheet were way outside my means as a teenager, though I was lucky then to find some ex-Army surplus parts. All too often the tubes had been pulled from ex-military kit but the housings were scrapped... it was the housings that were valuable!

Here's an example that is quite similar to what I built....
View attachment 325660

That's a serious project, quite a lot going on there. May I ask your age? I'm 64 and as a kid (say thirteen or so) I often got given old copies of Practical Wireless and I recall that there was often several transmitter projects a year in that magazine. A very good mate of mine attended Riversdale College in Liverpool for two years in the late 70s (long closed down now) to study Marine Radio and Radar, ended working in Saudi for a few years for Marconi doing radar and stuff.

 

It looks like Power Supply Rejection Testing or something like that.
I would never use white noise. I would sweep from low frequency to high. It will take less power that way. I normally inject a smaller voltage or current. With the test equipment I have it works well.
Probably you were told to do it this way, but I have never seen it done like this.
I have an injection transformer made to do this job, but it will not work to DC. I can't remember but I think it stops working above 1mhz. I could make a higher frequency injection transformer.

 

That's a serious project, quite a lot going on there. May I ask your age? I'm 64 and as a kid (say thirteen or so) I often got given old copies of Practical Wireless and I recall that there was often several transmitter projects a year in that magazine. A very good mate of mine attended Riversdale College in Liverpool for two years in the late 70s (long closed down now) to study Marine Radio and Radar, ended working in Saudi for a few years for Marconi doing radar and stuff.

I'm 66. In the early 70's, from about 13 onwards, I was very into amateur radio, primarily on the 144MHz/2m band. The power amp was built, circa '75-76 for a radio competition. I belonged to a local amateur radio club in north London (and the local motoring/rallying club as did many other members, the two interests went side by side), and we used to go up to Yorkshire and set up huge aerial masts in farmer's fields and see with how many stations, how far and how rare, we could communicate. Sadly I have no photos of the amp, which was a collaberative effort with other club members (from memory we built 3 in total), or the competitions, just fond memories. In 1979, after I graduated, I got a design job at the Electronic Warfare Labs of Plessey Avionics, working under/with several club members on ground-to-air comms, with a trend to semiconductors, notably early high-power Motorola RF MOSFETs. Once it became a job, my interest in amateur radio, but not electronics in general, waned, and cars and girls took over....

 

What I am guessing that the actual specification is, involves a noise suscepability test for some device that is powered by 12 volts and draws 20 amps, and some noise spectrum is added to that DC signal at some peak to peak amplitude that the TS saw no need to tell us about. But I KNOW how specifications are written. So instead of asking how to achieve what is in the specification the TS decided what the answer was and described a completely wrong approach.

So if you have already picked the answer, don't ask me the wrong question!!!

Modulating a DC voltage with a noise signal is an entirely different task in almost every aspect.

 

Body
Hello,

I am working on a power electronics project where I need to mix a 12V DC signal with a 1V AC signal and then amplify the output to achieve 20A while maintaining a bandwidth of 100 MHz. My current approach involves using operational amplifiers, but I am unsure about the best way to design the circuit to handle such high current without distorting the signal.

<snip>

I only the other day from answers to my thread did I discover how you can boost the current from an op amp with transistors. A possible approach? Certainly more elegant than paralleled op amps.

 

The question posed in post #2 should have been "What is the purpose of adding the noise to the DC power. Because 12 volts DC at 20 amps is not a "signal" to be amplified and injected, it is a power supply. That suggests that the TS is in far over their head.

It makes no sense to AMPLIFY a DC power supply voltage. that just simply makes no sense at all.
ADDING a wide-band noise signal to a DC supply voltage is an actual test process that is actually done by test groups in organizations that DO KNOW WHAT THEY ARE DOING. Not beginners.So the only part that needs an actual amplifier isthe wide-band noise signal. The 12 volts DC just needs a 20 amp regulated 12volt power supply. That immediately removes the need for a monster amplifier. Now the wide-band signal only needs a 20 watt power amplifier to add it to the DC power.

UNLESS the intention is to produce a monster jamming wide-band noise transmitter, which is a federal crime in the USA, where the FCC might seek assistance from the USMC to help quiet the noise source.

 

MisterBill2,Irving, ronsimpson , ApacheKid , crutschow

I understand from all the feedback that there might be an issue with the way I phrased my question. As I'm not accustomed to forums and English isn't my first language, I may have missed some nuances based on some of the responses I've received.

My only goal is to understand and learn. I know I am a beginner, and I am here to advance in my project. I will take the time to consult with my tutors and write a detailed paragraph with all the information I am authorized to share to explain my entire engineering school project.

Initially, I posed a specific question to avoid the "here's my project, give me the solutions to finish it" approach. I prefer to ask if certain types of configurations that I might not be aware of exist so I can continue my research. But now, I will try to elaborate more on the problem to have a constructive discussion.

I appreciate your patience and engagement. I am here to learn about electronics, and any advice or discussions are very welcome. I will return later with as much detail as possible.

 

I would like to provide more context to clarify my project and the challenges I am facing, without disclosing sensitive details.

Project Background:
The project originated from the need to accurately test electrical disturbances, such as noise in power supplies of high-frequency electronic boards. The current testing methods lack precision in simulating predefined noise profiles, which limits the ability to reliably evaluate the performance and robustness of these boards under specific noise conditions.

Current Challenge:
Existing methods use transformers and the "sense" function to simulate electrical disturbances, but these have significant limitations:

Transformers: Can induce unwanted voltage oscillations and surges, and often fail to reproduce the full spectrum of electrical noise.
"Sense" Function: Useful for controlling input voltage, but can lead to stability issues and potential damage to the boards being tested.
Pi Filters at Input: The boards have Pi filters at the input, which can contribute to oscillation issues when noise is injected.
Project Goals:

Develop a Noise Injection System: Create a setup that includes two main electronic boards - one to generate noise according to a defined frequency and amplitude profile, and another to inject this noise into the standard power supply of the test benches.
Ensure System Stability: Design the system to avoid unwanted side effects such as oscillations or voltage spikes that could damage the boards under test.
Key Specifications:

Noise Profile: Inject white noise with a frequency range up to 100 MHz and an amplitude around 10^3 µVrms/√Hz into a 12V power supply, achieving a 12V output with 20A.
High-Frequency Performance: Maintain a high bandwidth to accurately simulate operational conditions.
Avoiding Transistors: Based on guidance and an application note, using transistors directly is discouraged due to potential oscillation issues.

I am open to exploring any configuration or approach that meets the project requirements and overcomes the limitations of current methods.

I apologize if my explanation is not perfect; it is based on my current understanding. This is really the maximum detail I can provide. I appreciate the constructive discussion and any guidance or suggestions you can provide to help me advance in my project.

 

Avoiding Transistors: Based on guidance and an application note, using transistors directly is discouraged due to potential oscillation issues.

If taken literally, this really means you need to use vacuum tubes. All opamps use transistors as their active device.

 

I would like to provide more context to clarify my project and the challenges I am facing, without disclosing sensitive details.

Project Background:
The project originated from the need to accurately test electrical disturbances, such as noise in power supplies of high-frequency electronic boards. The current testing methods lack precision in simulating predefined noise profiles, which limits the ability to reliably evaluate the performance and robustness of these boards under specific noise conditions.

Current Challenge:
Existing methods use transformers and the "sense" function to simulate electrical disturbances, but these have significant limitations:

Transformers: Can induce unwanted voltage oscillations and surges, and often fail to reproduce the full spectrum of electrical noise.
"Sense" Function: Useful for controlling input voltage, but can lead to stability issues and potential damage to the boards being tested.
Pi Filters at Input: The boards have Pi filters at the input, which can contribute to oscillation issues when noise is injected.
Project Goals:

Develop a Noise Injection System: Create a setup that includes two main electronic boards - one to generate noise according to a defined frequency and amplitude profile, and another to inject this noise into the standard power supply of the test benches.
Ensure System Stability: Design the system to avoid unwanted side effects such as oscillations or voltage spikes that could damage the boards under test.
Key Specifications:

Noise Profile: Inject white noise with a frequency range up to 100 MHz and an amplitude around 10^3 µVrms/√Hz into a 12V power supply, achieving a 12V output with 20A.
High-Frequency Performance: Maintain a high bandwidth to accurately simulate operational conditions.
Avoiding Transistors: Based on guidance and an application note, using transistors directly is discouraged due to potential oscillation issues.

I am open to exploring any configuration or approach that meets the project requirements and overcomes the limitations of current methods.

I apologize if my explanation is not perfect; it is based on my current understanding. This is really the maximum detail I can provide. I appreciate the constructive discussion and any guidance or suggestions you can provide to help me advance in my project.

Also I'm not English. If I correctly understand Your problem: Take a stable 12V 20A power source with necessary power stability response (here your bandwidth 100MHz) AND try to modulate (inject, if You like. Yes, reverse engineering) circuits voltage stabilizing part with Your white noise. MR

 

Noise Profile: Inject white noise with a frequency range up to 100 MHz and an amplitude around 10^3 µVrms/√Hz into a 12V power supply, achieving a 12V output with 20A.

White noise from DC to 100mhz, there will be almost zero energy in the 50mhz to 50.01mhz area so measuring the output is hard. You are injecting a signal at all frequencies. I inject a small signal at a frequency that moves from DC to 100mhz. Because all the energy is at one frequency at any moment, the signal is easy to see, and the power used to inject is very small.

How do you make white noise at DC? Usually, we do not test below 50hz.

I know this has been said before, several time.

 

The question could have been presented like this:
I need to test a system to discover the effect of noise on the power supply input of a system. The power supply voltage is "VV" volts DC, and the current is "AA" amps. The noise test specifies "Noise frequency spectrum limits" at a level of "NV" volts relative to the power source negative side.
What is the best (cheapest?), (Fastest), ( Most accurate) method that somebody has experience with??

The only product information given is the voltage and the current, which do not really give away much detail.

Telling us that you needed an amplifier to provide both DC and AC greatly limits the options.

 

Noise Profile: Inject white noise with a frequency range up to 100 MHz and an amplitude around 10^3 µVrms/√Hz into a 12V power supply, achieving a 12V output with 20A.

I have some issues with this element of the ongoing problem statement.

Firstly, white noise, by definition, is noise with constant power spectrum i.e. a flat frequency spectrum over the bandwidth in question. At a guess, one could argue for white noise into a specific resistive load, say 0.6ohm (being 12v @ 20A). Since V/√Hz = √(P/Hz x R), P/Hz = V/√Hz^2/R = .001 * .001/.6 = 1.666μW/Hz. Over a 100MHz bandwidth that's something like 167W if my math is right.

Secondly, top of the range lab supplies from the likes of Rigol, Agilent/Keysight, etc have a noise figure of 350uVrms over a 20MHz band width, roughly equivalent to 350/√20e6*√100e6 = 783uV√Hz, over 100MHz BW so roughly in the ballpark of your test signal - how are you going to isolate that?

 

Given that the noise will need to be isolated from the power supply output by a filter, that filter will also tend to block the noise contribution from the power supply. AND, given that the test requirement may require verification of the noise bandwidth, some adjustment may be required. Seldom is it wise to simply push ahead without being aware of just what one is actually doing.

 

Also I'm not English. If I correctly understand Your problem: Take a stable 12V 20A power source with necessary power stability response (here your bandwidth 100MHz) AND try to modulate (inject, if You like. Yes, reverse engineering) circuits voltage stabilizing part with Your white noise. MR

With which component do you make the mixture of the tension 12V 20A and the noisy signal?

 

White noise from DC to 100mhz, there will be almost zero energy in the 50mhz to 50.01mhz area so measuring the output is hard. You are injecting a signal at all frequencies. I inject a small signal at a frequency that moves from DC to 100mhz. Because all the energy is at one frequency at any moment, the signal is easy to see, and the power used to inject is very small.

How do you make white noise at DC? Usually, we do not test below 50hz.

I know this has been said before, several time.

Thank you for your suggestions regarding the approach of frequency sweeping versus white noise injection. Currently, I generate the white noise using MATLAB to create a signal that conforms to a specific frequency template, and then I perform an FFT to analyze the spectrum. This method allows me to cover the range from 50 Hz to 100 MHz, in accordance with the requirements of my project.

I understand the advantage of frequency sweeping, especially in terms of simplifying measurement and visualizing disturbances at specific frequencies. That said, I have very advanced measuring equipment that can detect very low signal levels, so the low energy density of white noise is not an obstacle for us in terms of measurement capability.

However, I am curious to know if switching to a frequency sweeping method could impact other technical specifications of our project, particularly for our mixing board that currently processes white noise. Are there specific considerations or adjustments we should consider for the mixing board if we opt for frequency sweeping? Could this change the bandwidth or signal processing requirements in a way that might influence the overall design or functionality of the board?

 

The question could have been presented like this:
I need to test a system to discover the effect of noise on the power supply input of a system. The power supply voltage is "VV" volts DC, and the current is "AA" amps. The noise test specifies "Noise frequency spectrum limits" at a level of "NV" volts relative to the power source negative side.
What is the best (cheapest?), (Fastest), ( Most accurate) method that somebody has experience with??

The only product information given is the voltage and the current, which do not really give away much detail.

Telling us that you needed an amplifier to provide both DC and AC greatly limits the options.

Thank you for your perspective on how I could have framed my question. Initially, I hesitated to present my query in such an open-ended manner because it seemed too much like asking for a complete solution without demonstrating my own efforts. However, I have been conducting research on my own, and my intention was to find out if there exists a specific application or solution based on my conclusions. This approach allows me the possibility of being wrong, which is part of the learning process. My goal isn't to have you solve my project without any input from my side but to verify and refine my findings through expert feedback.

I now realize that providing more detailed information could help you better understand the full scope of my problem and assist more effectively. This acknowledgment comes from recognizing that my initial assumptions and specifications might be incorrect, which could prevent you from grasping the entirety of the issue and proposing the most suitable solution.