That amplifier is attempting a "DC to Daylight" passband, and so you might consider limiting its response to the audio band. See "radio production environment" filtering.
The Audiosector product is an amplifier kit--it is up to the purchaser to add any and all support components that are either needed or wanted. Filters are one of those things that you add on yourself. So are speakers.
A weaker filter is better than an omitted filter. Simply adjust your filters until the audio band passes un-hindered.
Although you should use the Output RC (speaker jack zobel), you do need a more appropriate cap for that. In this case, overspending would only get you the wrong part. You need an inexpensive polyester dip cap for that job--exactly the same thing that's found on the outputs of the t-amp. That's a high-ESR cap. And that's when the national semiconductor data sheet resistor value actually makes sense.
At the chip itself, you can put a 220pF cap from the chip's IN+ to the chip's IN- telling it to never amplify radio signals. And you can add a 100pF from + to - at the RCA jack. (of course you could use 330pF for both of those caps if you want)
You can add a 2uF to 4.7uF (range) polyester tweeter cap from V+ to V- at each amplifier board (only one cap per amplifier board) which will short noise during its passband. At best, this can only clean up mid-range and higher frequencies of the audio band.
You can put an ac rated 5nf polyester cap in parallel with each of the secondaries of the transformer. There are two secondaries, so this takes 2 caps.
You can swap the MUR860's out for MR756 instead, which is still a fast diode but not nearly as apt at causing noisy voltage spikes.
And, you could go to decibel dungeon for the example of how to connect earth ground to chassis ground without making a loud ground loop noise.
Peter also specifies an input transformer for his AudioSector amplifier. This isolation may be able to stop ground loops as well. I don't know, but you could use the AudioSector support portion of DIYaudio.com and find out.
Of course you do have the option of using regulated power on all of your audio components. The capacitive multiplier regulated power will stop power noise with a performance that's far better than. . . that would be a long list indeed.
Well, that's all the noise I can think up tonight.
The Audiosector product is an amplifier kit--it is up to the purchaser to add any and all support components that are either needed or wanted. Filters are one of those things that you add on yourself. So are speakers.
A weaker filter is better than an omitted filter. Simply adjust your filters until the audio band passes un-hindered.
Although you should use the Output RC (speaker jack zobel), you do need a more appropriate cap for that. In this case, overspending would only get you the wrong part. You need an inexpensive polyester dip cap for that job--exactly the same thing that's found on the outputs of the t-amp. That's a high-ESR cap. And that's when the national semiconductor data sheet resistor value actually makes sense.
At the chip itself, you can put a 220pF cap from the chip's IN+ to the chip's IN- telling it to never amplify radio signals. And you can add a 100pF from + to - at the RCA jack. (of course you could use 330pF for both of those caps if you want)
You can add a 2uF to 4.7uF (range) polyester tweeter cap from V+ to V- at each amplifier board (only one cap per amplifier board) which will short noise during its passband. At best, this can only clean up mid-range and higher frequencies of the audio band.
You can put an ac rated 5nf polyester cap in parallel with each of the secondaries of the transformer. There are two secondaries, so this takes 2 caps.
You can swap the MUR860's out for MR756 instead, which is still a fast diode but not nearly as apt at causing noisy voltage spikes.
And, you could go to decibel dungeon for the example of how to connect earth ground to chassis ground without making a loud ground loop noise.
Peter also specifies an input transformer for his AudioSector amplifier. This isolation may be able to stop ground loops as well. I don't know, but you could use the AudioSector support portion of DIYaudio.com and find out.
Of course you do have the option of using regulated power on all of your audio components. The capacitive multiplier regulated power will stop power noise with a performance that's far better than. . . that would be a long list indeed.
Well, that's all the noise I can think up tonight.
I get tired of people telling me "do not kill yourself" with AC related things. I have never even had a close call or accident, spark, nothing. Just because someone is working with AC does not mean they might kill themselves. Unlike a fair amount of people I have had some formal teaching on AC and DC.
Daniel the capacitor on the pins of chip is what PD recommended and I will probably do. Several of the cap mods you are mentioning I have not been interested in due to PD's recommendations. My true biggest problem is being so close to a radio station.
I am also a bit of a minimalist when it comes to building equipment.
Daniel the capacitor on the pins of chip is what PD recommended and I will probably do. Several of the cap mods you are mentioning I have not been interested in due to PD's recommendations. My true biggest problem is being so close to a radio station.
I am also a bit of a minimalist when it comes to building equipment.
Cap across the in+ and in- of the chip itself is a big head start.
You should also ask Peter about a small value capacitor or RC as an RF blocker at the input jack.
And you should also ask Peter about using a weaker speaker output RC instead of an omitted speaker output RC.
These steps are usually easier than relocating a house.
Well, here's the manufacturer's datasheet: http://www.national.com/ds/LM/LM3875.pdf
According to National Semiconductor (the actual manufacturer of LM3875), that resistor does prevent noise. . . but only when the amplifier is off or run at voltages less than the output of an AA battery.
This isn't useful during playback.
See for yourself. Its RB, on page 6.
According to National Semiconductor (the actual manufacturer of LM3875), that resistor does prevent noise. . . but only when the amplifier is off or run at voltages less than the output of an AA battery.
This isn't useful during playback.
See for yourself. Its RB, on page 6.
That resistor will provide some protection for the chip, thus reducing unwanted noise caused by exploding silicon etc. He's not kidding about "...if you plug equipment you are unsure of into your amp..."
Here's some history to put things in perspective:
Maybe that's worth repeating:
There's not much difference between the kind of mistake that kills a volume pot and the kind of mistake that kills you.
Please take safety seriously. Your comment about not being scared of electricity doesn't suggest a healthy attitude.
Here's some history to put things in perspective:
That's also why there's some concern about safety. There's not much difference between the kind of mistake that kills a volume pot and the kind of mistake that kills you.
Maybe that's worth repeating:
There's not much difference between the kind of mistake that kills a volume pot and the kind of mistake that kills you.
Please take safety seriously. Your comment about not being scared of electricity doesn't suggest a healthy attitude.
Last edited:
When you are experimenting or servicing any equipment connected to mains power there is always a possibility of hurting or possibly killing yourself. This may be a hobby for you and nothing may have happened YET, but that doesn't mean that the risk isn't there.
I find this to be an extremely arrogant statement on your part.
I have had years of formal teaching in electronics, and experimental physics. This doesn't mean that I shouldn't take advice from people who don't have the same background as me. I'm certain there are many out there who may not have sat through a class or two on electronics, but have a better understanding of the subject than I do.
If you think you know better than everyone on a DIY website, why bother asking here in the first place?
I have heard countless stories involving careless and stupid mistakes that have hurt and even killed people. I've had to call 911 for someone who made a simple mistake in a lab that caused an explosion, and that person ended up blind in one eye and burns covering their face and hands. I wish I had seen what they were about to do so I could have had an opportunity to stop them.
I've also heard stories like this directly from the mouths of people who are considered to be some of the most intelligent people on this planet, including Nobel laureates. If these accidents can happen to them it can happen to anyone in a lab, and also anyone tinkering around in their apartment.
Of course, just because it can happen doesn't mean it will happen.
However, you should be grateful that people are warning you. If you've had any formal training you should know that what they are saying is true, and that it shouldn't be take lightly.
The way you're talking in this thread makes me believe that you are more interested in having your noise problem fixed, rather than carefully experimenting to figure out what is causing the noise and fixing the problem in a safe way.
I could be completely wrong, but that is the way you come across. I believe this is why people have and will continue to give you warnings.
Ah, but he has just enough understanding to design and build mains power conditioners...! And sell them.
Destroyer OS,
I am an old retired service technician. I traveled all over eastern Ontario fixing problems that others could not. Several ( >15 years ago) I went to an account to help solve a problem, I had to wait an extra day to gain access to the equipment as a certified electrician was killed servicing an electrical panel that day. He forgot to stand beside a breaker panel when he turned off the power. The panel blew and killed him. He was not the first that I heard of that happeneing. Always be extra safe you could be dead in a second.
I am an old retired service technician. I traveled all over eastern Ontario fixing problems that others could not. Several ( >15 years ago) I went to an account to help solve a problem, I had to wait an extra day to gain access to the equipment as a certified electrician was killed servicing an electrical panel that day. He forgot to stand beside a breaker panel when he turned off the power. The panel blew and killed him. He was not the first that I heard of that happeneing. Always be extra safe you could be dead in a second.
R1, Audiosector, a resistor in series to the input signal
RB, National Semiconductor, a resistor in series to the input signal.
This resistor connects to the inverting input. The audiosector diagram here: http://www.audiosector.com/nigc_kit-users_guide.pdf on page 3.
I read the National Semiconductor datasheet as if this resistor is supposed to assist with decreasing noise that may happen after the amplifier is turned off.
Here is a direct full quote from National Semiconductor, (the maker of LM3875TF):
"Prevents currents from entering the amplifier's non-inverting input which may be passed through to the load upon power-down of the system due to the low input impedance of the circuitry when the under-voltage circuitry is off. This phenomenon occurs when the supply voltages are below 1.5v"
As LM3875 cannot use 1.5v rails during playback, in fact, that resistor doesn't reduce noise during playback, but the mention in the data sheet has led to a misconception.
This resistor works when the amplifier's power switch is OFF and during the timeframe where the power capacitors are draining of voltage. The point being: The power switch is off. The resistor is useless during playback.
Exploring the rest of the input circuit for causes of DC offset:
(exploring mildly)
In this case, Audiosector chose to move the load (their R2) for direct contact with the chip, which would have decreased DC offset by 1/10th, except that they also chose to lighten the load from NatSemi's 10k up to an arbitrary 22k, greatly increasing DC offset.
Audiosector's idea is much better than what will happen when NatSemi's 10k pot gets dirty and releases heavy duty bursts of DC offset. Because of the DC offset issue, I will have to assume that Audiosector intended the customer or end user to directly wire a 20k pot, which would sum with their 22k resistor and restore the datasheet spec input load of approximately 11k (load at DC), canceling DC offset down to a figure that's hard to measure.
I think that Audiosector intended 20k//22k as the input load. Part of that is added by the customer or end user. It isn't nice to assume the customer will do this!
Unfortunately Audiosector and Chipamp.com customers also keep mistakenly placing the input filter cap inbetween pot and load (or omitting the pot's load by simply omitting the pot), thereby raising the load on DC back up to 22k and then complaining about DC offset (and sometimes noise). Well, that cap (or isolation transformer) goes at the RCA jack.
And now the filters:
Its almost a mystery of why customers would choose to omit rather than fine-tune filters, however. . .
If we're talking about Minimalist Design, we're also talking about: Muntzing - Wikipedia, the free encyclopedia <See link. Your RF filter has been cut off. The link explains it.
P.S.
That's about the best I can do on a response while also aiming for a really wide audience. Perhaps someone with finer writing skills would like to compress these materials into something more focused.
I should screw up and do something stupid that kills me because apparently if I make a post I have to read all that ****.
Danielwritesbac I was going based on what Peter Daniels said about R1.
Hey Godfrey got any good quotes about not feeding trolls?
smgidcumb my comment about having some formal training is not to insult others. You turned it into an insult. It was just stating that even if how I explain things may be confusing, the fact of the matter is I am not just throwing wires in a box randomly. You want an insult? I think your jerk because you are trying to speak for everyone else via trying to vilify me.
multisync nice anecdotal story. Anecdotal hahahaha.... People die all the time but I am amazed they do not die more. Big deal. Why am I not dead? Who knows. Old hat.
I respect electricity. If it killed me no big deal. If I get shocked then well that totally sucks. I am very careful and meticulous to check things over and cautious as to how I test things. I do not put myself in situations that are dangerous. It would be fair to give me a warning "careful working with AC, very dangerous" but thus far you have all been pushing your morality on me. Get off me. Either you want to help me further my power conditioners or you want to be a troll. Trolls are not welcome.
Besides that just in case you missed it I solved my noise problem. It was just the faulty earth ground in the building. I have no idea why some of you keep repeating that I have a problem. Now I am just concentrating on making the power conditioner better. I think there may be a gap of distinction that many of you are not aware of in that what a good power conditioner can do will blow you away; yet once removed from the system you will not be able to articulate specific noises you hear but will know that it does not sound as good.
Danielwritesbac I was going based on what Peter Daniels said about R1.
Hey Godfrey got any good quotes about not feeding trolls?
smgidcumb my comment about having some formal training is not to insult others. You turned it into an insult. It was just stating that even if how I explain things may be confusing, the fact of the matter is I am not just throwing wires in a box randomly. You want an insult? I think your jerk because you are trying to speak for everyone else via trying to vilify me.
multisync nice anecdotal story. Anecdotal hahahaha.... People die all the time but I am amazed they do not die more. Big deal. Why am I not dead? Who knows. Old hat.
I respect electricity. If it killed me no big deal. If I get shocked then well that totally sucks. I am very careful and meticulous to check things over and cautious as to how I test things. I do not put myself in situations that are dangerous. It would be fair to give me a warning "careful working with AC, very dangerous" but thus far you have all been pushing your morality on me. Get off me. Either you want to help me further my power conditioners or you want to be a troll. Trolls are not welcome.
Besides that just in case you missed it I solved my noise problem. It was just the faulty earth ground in the building. I have no idea why some of you keep repeating that I have a problem. Now I am just concentrating on making the power conditioner better. I think there may be a gap of distinction that many of you are not aware of in that what a good power conditioner can do will blow you away; yet once removed from the system you will not be able to articulate specific noises you hear but will know that it does not sound as good.
Well, since R1 doesn't do much of anything, let's look at R2, which is considerably more fun.
If you measure across R2, does your meter say 22k or more like 11k? (I'm hoping for 11k).
And, do you have a potentiometer on that amplifier? (I'm hoping yes).
I will have to open it up. My pot is a stepped attenuator outside of the amplifier actually.
Here is the circuit I want to add to my PC that just has a bank of capacitors. I got a feeling there may be something I do not know but what I am thinking is it will not only attenuate but kill most of the signal further down the line by eliminating the draw that anything else outside of the unit may see.
Here is the circuit I want to add to my PC that just has a bank of capacitors. I got a feeling there may be something I do not know but what I am thinking is it will not only attenuate but kill most of the signal further down the line by eliminating the draw that anything else outside of the unit may see.
Good luck with that - Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.