I will try to get those soon. No, as mentioned earlier, two amp boards (b1, b2) overheat a lot and have hum. Output significant voltages with no input connected. Third board (soldered by someone else) exhibits hum. Can't remember if it exhibits voltage issues. No overheating.
Speaker is an old Realistic Minimus 7. Mfgr. rated @ 8 ohms. Original music source was Sony CD player. But remember, DC voltages on output of two amp boards (b1 and b2) wwere with no signal source connected.
Gosh, I hope I'm remembering all this correctly.
Speaker is an old Realistic Minimus 7. Mfgr. rated @ 8 ohms. Original music source was Sony CD player. But remember, DC voltages on output of two amp boards (b1 and b2) wwere with no signal source connected.
Gosh, I hope I'm remembering all this correctly.
Last edited:
So tell us, for your test, did you ground everything, or did you ground nothing?
Show us what you did.
Last edited:
Everything audio was grounded, under the supervision of my friend, who has built a number of amplifiers from scratch (with no help, I might add, and before the digital age).
I will try to photo once my camera's working. Nikon P7000 has a problem with lens blades not opening and closing properly and focusing going out of whack. I have a new lens assembly here, but if you saw how tiny the screws were and how many of them were...it could scare a neurosurgeon.
I will try to photo once my camera's working. Nikon P7000 has a problem with lens blades not opening and closing properly and focusing going out of whack. I have a new lens assembly here, but if you saw how tiny the screws were and how many of them were...it could scare a neurosurgeon.
well if the amp has no connection to a source device and the input is not intentionally shorted to ground for test purposes then it "humming" and doing weird things like "heating" is to be expected. (could well be high frequency oscillation/instability.any zobel network on the output?)
if you don't fully disclose test setup conditions it's arbitrary.
hope they're not toast!
if you don't fully disclose test setup conditions it's arbitrary.
hope they're not toast!
Yes. That is my design and build of an LM3886 stereo amplifier. Those test are one channel and a very good star ground. Note the very low impedances measured. This is good grounding results. Good grounding is easy however, this kit amp is not very well grounded and would suppose it oscillates easily.
Using this amp, do not connect the input hot to ground will less than a 20k ohm resistor or whatever is the schematic. Shorting the input hot to "some ground" will probably make it oscillate as there is no real ground point. Also lightly load the output if it still oscillates like a 100 ohm resistor.
To someone's question and grounding I have seen this being better than most... Not in kits however, in professional or much commercial gear, yes I have seen a lot of grounding better than this. Most of it actually. If anyone cares and wants to go to the work of communicating with the kit maker, I will correct the board grounding for free. When someone lets me know the kit maker agrees, I will contact them.
Good luck.
really? this is a fantastic offer, I'm sure chipamp.com will be contacting you very soon🙄 haha
anyway why don't you say what the problem is 1st, and then propose a better solution. (I don't believe you have even studied the thing.) their star ground is a ground plane on the PCB, and is separate all the way to the bulk caps. (makes twisted pwr cabling very effective E.g. separate returns ).
Ground planes do not work for analog. I went over this in an earlier post. I also suggested changes to mostly fix the grounding even with an image of where to cut the stupid ground plane that does not work to convert to star ground.
In short, did all this already. Did you read all my post infinia?
In short, did all this already. Did you read all my post infinia?
well if it so easy to do, why don't you shake up the kit world and offer yet another, Im not sure you have convinced any buyers in your posting on this thread. hum and low distortion are 2 separate problems, ones a system interconnect issue and the other has been shown to improve with a central power ground plane , as demonstrated on your linked thread post 29 ( didn't you understand it?).
Hum is distortion so am not at all certain what you are getting at other than showing confusion. Hum is a great source of IM distortion. Did you know that? Have never seen anything analog using a "central ground plane" being any sort of solution to analog grounding. Where did that notion come from? Bob Pease would not agree with you and neither would Walt Jung. Yes I have had many conversations with both these guys erstwhile. It seems infinia just is just looking for ways to defend a poor grounding scheme and clearly does not understand good grounding. Nothing I build ever hums or oscillates. That chip amp version of mine has 250dB of open loop gain and not even a hint of oscillation. A big part of that is proper star grounding. There is no "ground plane" involved.
Ground point is akin to a mathematical "point" and located where the designer defines it. In analog design there is no such thing as a "ground plane" because in analog it is the total impedance in the signal path which matters and not simply the resistance between two points. The impedance across a ground plane varies a lot when testing between any two points on the plane which is why it is so important to use star grounding and a defined single point ground. Assuming a ground plane will all be at the same AC signal potential is a shortcut to oscillation and also very wrong. Analog is all about AC signals. Low noise AC signals is all about have a single point where all signals and circuits can find a "reference" voltage of ground. A ground plane design has varying voltages all over the area of the ground plane making any point on the plane a different reference voltage. Fine for digital. Really lousy for analog.
Am not going to try and defend this anymore. Either you understand this concept and its' importance or you do not. There is no reason to explain this further as all important aspects of implementation and theory of grounding have been covered in my post. If you want to use a ground plane for analog design, be my guest and have poor noise and hum performance. As a note, there was a U.S. Military guide on equipment grounding used for some time. I wrote that. I have no idea where this crazy notion a "plane" is somehow the same as a "point" came from. No one would say that based on math models or physical presentations.
Ground point is akin to a mathematical "point" and located where the designer defines it. In analog design there is no such thing as a "ground plane" because in analog it is the total impedance in the signal path which matters and not simply the resistance between two points. The impedance across a ground plane varies a lot when testing between any two points on the plane which is why it is so important to use star grounding and a defined single point ground. Assuming a ground plane will all be at the same AC signal potential is a shortcut to oscillation and also very wrong. Analog is all about AC signals. Low noise AC signals is all about have a single point where all signals and circuits can find a "reference" voltage of ground. A ground plane design has varying voltages all over the area of the ground plane making any point on the plane a different reference voltage. Fine for digital. Really lousy for analog.
Am not going to try and defend this anymore. Either you understand this concept and its' importance or you do not. There is no reason to explain this further as all important aspects of implementation and theory of grounding have been covered in my post. If you want to use a ground plane for analog design, be my guest and have poor noise and hum performance. As a note, there was a U.S. Military guide on equipment grounding used for some time. I wrote that. I have no idea where this crazy notion a "plane" is somehow the same as a "point" came from. No one would say that based on math models or physical presentations.
sumaAguy why did you link to a thread ( for an example of a good chip amp grounding) that shows with measurements how a ground plane reduces parasitics that contribute to lowering distortion, yet you now disagree? any PCB examples of your ideal grounding scheme?
it's also strange to me , that you immediately come to the conclusion that "bad layout" is the cause of the OPs multiple problems, when hundreds of problem free builds have been built beforehand. in my thinking his reported "issues" clearly points to newbie operator errors
BTW power line related hum that is coming from the source grounding, is a different mechanism to IMD, but a single THD # doesn't differentiate them, they are not the same thing! that is why trouble shooting the input connections is a key 1st step.
it's also strange to me , that you immediately come to the conclusion that "bad layout" is the cause of the OPs multiple problems, when hundreds of problem free builds have been built beforehand. in my thinking his reported "issues" clearly points to newbie operator errors
BTW power line related hum that is coming from the source grounding, is a different mechanism to IMD, but a single THD # doesn't differentiate them, they are not the same thing! that is why trouble shooting the input connections is a key 1st step.
Well, my friend brought a really old oscilloscope over on Saturday evening and tried to test the gear. Unfortunately, it looks like the probes he brought weren't compatible somehow with the scope. No readings could be taken.
We'll try again as soon as he has time. Wish I had the skills. 🙂
We'll try again as soon as he has time. Wish I had the skills. 🙂
For good chip grounding link I have no idea where the notion that design has a ground plane. It does not. It is proper star ground. Maybe you did not understand?
For "bad layout"--- Take a marginal design and some poor guy gets a marginal part or parts installed into the marginal design and "surprise," it's an oscillator. This is easily demonstrated by doing a ground plane op amp test bed with marginal design and then swap in a bunch of different opamps types. Some will oscillate, some will not. But you have not probably done such as I have. Just like my chip amp whips everyone else's easily. Hence the specs.
Nit picking different ways hum can enter and always causes IM seems pointless. Was there a point? Back to star ground and problem gone, pointless.
Hmm...I finally get around to trying to ground the things better, and I can't find good diagrams (NOT schematics) or verbal instructions on star grounding of these things.
Can someone please point me to a good diagram/picture?
And sumaudioguy, would you please, (this is the second time I've asked) stop hijacking my thread.
Can someone please point me to a good diagram/picture?
And sumaudioguy, would you please, (this is the second time I've asked) stop hijacking my thread.
Last edited:
Hi HW
its pretty easy didn't you review any chip amp build threads before building this?
connect 2 pairs of wires to the power supply
connect 1 pair " to speaker connector
connect another pair " to a volume controller or source thru a RCA connector.
I suspect you may be new to measuring electronics and audio things, it helps to have someone show you basic operation of DMMs? find Utoop videos >
before any of this you should of connected the transformer to the power supply PCB and tested it alone .
fuses may blow here , resolve 1st then move on.
a few ppl have asked to see your test setup, pictures ?
its pretty easy didn't you review any chip amp build threads before building this?
connect 2 pairs of wires to the power supply
connect 1 pair " to speaker connector
connect another pair " to a volume controller or source thru a RCA connector.
I suspect you may be new to measuring electronics and audio things, it helps to have someone show you basic operation of DMMs? find Utoop videos >
before any of this you should of connected the transformer to the power supply PCB and tested it alone .
fuses may blow here , resolve 1st then move on.
a few ppl have asked to see your test setup, pictures ?
Last edited:
Infinia:
Thanks for that. I'm sorry, I don't understand at all. Can you point me to a picture or something? I read Decibel Dungeon's page had this, but I can't see anything there that I can understand.
Thanks for that. I'm sorry, I don't understand at all. Can you point me to a picture or something? I read Decibel Dungeon's page had this, but I can't see anything there that I can understand.
IDK ill be searching the treads just like anyone else ?
isn't stuff labeled on the boards for the power supply and amps 4 PS wires?
come on man most folks build it from the chimpamp.com manual, print one out, sit down and read it.
isn't stuff labeled on the boards for the power supply and amps 4 PS wires?
come on man most folks build it from the chimpamp.com manual, print one out, sit down and read it.
Hi,
Attached is a drawing showing how to do the start ground for a LM3886 amplifier built. I used it in my 3 LM3886 amplifier built with an output offset of 2mv. You can use it as a reference.
Attached is a drawing showing how to do the start ground for a LM3886 amplifier built. I used it in my 3 LM3886 amplifier built with an output offset of 2mv. You can use it as a reference.
note ^ Marks figure is not the ideal way nor what chimpamp.com recommends E.g. defeats what has been carefully designed for. also his amperage is for 230V and doesn't show speaker or input signal grounds. it's one way to do it , but not what i'd recommend.
the star ground for power is on each amp PCB you should use 2 returns one for + and -, that way you have 2 twisted pairs for each amp. Marks diagram takes the star ground off the PCBs and can possibly cause imbalance. ( if you do decide to do it that way make sure to use double thickness ground wires and measure distances of each piece E.g. keep symmetry.)
Mark why do you do it like that? my answer > b/c the chassis wire is one, instead of one for each PCB I bet. and then the infamous ground breaker gizmo gets inserted there next. LOL
the star ground for power is on each amp PCB you should use 2 returns one for + and -, that way you have 2 twisted pairs for each amp. Marks diagram takes the star ground off the PCBs and can possibly cause imbalance. ( if you do decide to do it that way make sure to use double thickness ground wires and measure distances of each piece E.g. keep symmetry.)
Mark why do you do it like that? my answer > b/c the chassis wire is one, instead of one for each PCB I bet. and then the infamous ground breaker gizmo gets inserted there next. LOL
Last edited:
- Status
- Not open for further replies.