I know what the arguments are for ones that are two small, but what detriments arise if you have too much capacitance? A couple of times I have heard people say that it makes the response a little sluggish.....
As a goal, how much noise should I hear through the 10'' woofer I'm using to test the noise? Without caps it is at the volume level of a loud conversation, with 1 4700uf cap it is like a quieter regular conversation; should it be at a whisper or less even?
The caps near the power input pins do not help with the noise; after looking at how the power moves throughout the circuit I have decided that they are there so that the signal coming from the speaker pushes at the capacitors, forcing more power on the other end into the circuit so that it has enough juice for heavy bass and loud impacts, and when it pulls back it forces the other end to pull in a large number of electrons to keep as storage for when the next hit comes.
What are the advantages between a single capacitor in parallel and two in series parallel as shown on the NSC pdf?
As a goal, how much noise should I hear through the 10'' woofer I'm using to test the noise? Without caps it is at the volume level of a loud conversation, with 1 4700uf cap it is like a quieter regular conversation; should it be at a whisper or less even?
The caps near the power input pins do not help with the noise; after looking at how the power moves throughout the circuit I have decided that they are there so that the signal coming from the speaker pushes at the capacitors, forcing more power on the other end into the circuit so that it has enough juice for heavy bass and loud impacts, and when it pulls back it forces the other end to pull in a large number of electrons to keep as storage for when the next hit comes.
What are the advantages between a single capacitor in parallel and two in series parallel as shown on the NSC pdf?
Even when I wire all 3 (I found 3) 2700 and the two 4700uf together in the way that is prescribed in the pdf (series parallel) the hum is only marginally reduced from using only one 4700 in parallel. It seems like I cant get the hum any lower...so this amount must be ok then?
I'm also having a REALLY hard time soldering copper wire to aluminum (ground). Is it ok to use conductive glue for this?
Synesthesia said:
The charging current rises, which may lead to fuse blowing, when you power up the amplifier. There is also a point, where increasing capacitance further doesn't improve things any further.
Synesthesia said:
That is yours to decide. It is also hard to know, how much noise that really is, because woofers have different efficiencies. You should try with the speakers that you will eventually use, and decide according to how much noise you find bothering. Don't you have any commercial built amp for comparison?
The important point is that one side of the capacitors is grounded. The ground is the 0V reference. A filter only from rail to rail cannot have too much effect because it is referencing to a dirty source.Synesthesia said:
Your hum probably comes from a ground loop then. If you look up star grounding, you may find a solution.Synesthesia said:
No personal experience on that. Try it, and measure the resulting connection resistance. Depends probably on the purpose of that connection (grounding?). Connections to aluminium are better done with screws or bolts.Synesthesia said:
lol, Do not use glue! if you need to ground something to chassis, what you have to do is "bolt it" to chassis.just use a drill an make a small hole (big enough for the bolt) make sure you scrape it all around to get the best contact as possible and then tight it as hard as you can. Good luck!
PS: use dented washers.
PS: use dented washers.
I looked up star grounding but I cannot really utilize it if I only am dealing with one ground for the psu (my transformer does not have a ground). That ground then connects (since it is not currently housed in a chassis for ease of work) to a piece of aluminum. Since there is only one connection and the parts are housed outside of the grounding metal I cannot see how a ground loop can occur in this situation.
Interesting.
So what can cause different potentials? Is it a matter of the ground positioning and what other grounds are around it?
Interesting.
-wikipedia
So what can cause different potentials? Is it a matter of the ground positioning and what other grounds are around it?
Synesthesia said:
Your signals must be grounded somehow. You can get ground loops e. g. if the input connectors are grounded on the PCB through a wire and at the same time through being screwed to a grounded chassis. Another source for ground loops ocurrs, when you connect the ground to safety earth (PE) while another component in your chain already has such a connection.
You may want to read the datasheet for your IC from here, where you will find a good explanation about that issue.
all circuits by definition must have a flow and return loop.
Much of the confusion with the term "ground" is that the same word is used for many different parts of many circuits.
Think in loops and ensure each loop works correctly.
Much of the confusion with the term "ground" is that the same word is used for many different parts of many circuits.
Think in loops and ensure each loop works correctly.
Yes, and then connect all those loops to what you should from then on refer to as the ground. Make sure that each loop has only one connection to ground.
I'm not sure about toroids, but it is a good idea to connect the metal case of EI transformers to ground... some of mine even have a screw and nut marked specificaly for this..
Now I'm really starting to get annoyed. I hooked up 2 12v high ma batteries in series to the chip....NO SOUND. Not even any noise. I also cannot reduce the hum of the psu no matter how hard I try. I am really tired of failing at this.....I will just have to get the kit layout board. The only other problem is that I don't know if either chip is busted or not. What a pain....makes me not want to DIY amplifiers and just do speakers (once I get my resources to build them back) since I only had to fix one problem on my first speaker ever. I had no guidance like with this or any support...and I rushed it! Yet the only problem was with the zobel network configuration...which I identified as the problem within 5 minutes and fixed without any damaged caused. Anyways....since no one on any forum is willing to vouch for the sound quality of their amplifiers I am starting to doubt it will even sound as good as the ones on my AVR745.
.....with pcb's, do you only put the wire through its designated hole, or are you expected to bend it onto the contact surface?
Can a damaged LM3875 hurt any other components? If so, is there any way to check if it is damaged besides testing it with the other parts?
I also have to way of checking if my capacitors are working as they should be without hooking it up to everything and possibly damaging everything. I can't find a DMM that measures in the thousands of microfarads; is there a way to check for damage by checking the resistance across the capacitor? I usually get around 600, then 1500ohms, then nothing in a period of around 3 seconds.
.....with pcb's, do you only put the wire through its designated hole, or are you expected to bend it onto the contact surface?
Can a damaged LM3875 hurt any other components? If so, is there any way to check if it is damaged besides testing it with the other parts?
I also have to way of checking if my capacitors are working as they should be without hooking it up to everything and possibly damaging everything. I can't find a DMM that measures in the thousands of microfarads; is there a way to check for damage by checking the resistance across the capacitor? I usually get around 600, then 1500ohms, then nothing in a period of around 3 seconds.
Synesthesia said:
High ma?
Read the forum again. There are countless people telling how much they love the sound of their LM3875s.Synesthesia said:
You can bend for stability during soldering, but it is not necessary for other reasons.Synesthesia said:
Synesthesia said:
OK, sounds like you need some help, I remember when I built my first GC. I bought 6 chips and killed 3. One exploded, one melted and one I broke a legg off...
So, do not feel bad.
Now! lets get your amp up and working.
How?
We are going to break it down to the most basic form, then you can add parts to build up to the full schematic....
You need
Transformer
Rectifier
chip
input cap
the resistors from your schematic...
Connect the transformer to the bridge, if it has 2 seperate secondaries, tie the middle two together and connect that to the earth lug on your case, alternatively if it has only 3 wires, run the middle one to the ground lug. (We are sacrificeing non-essential parts, not safety).
Andrew will chip in to remind you to add a light bulb in series with the live wire on the mains before the transformer during testing.
Seperate the two DC wires comming from the rectifier, connect your multimeter's black probe to the ground tab, and connect the other lead to one of the two DC wires... make sure the other one is not going to touch anything while it is powered on.
Hands off the circuit, meter connected as described, switch on the power at the wall, observe the voltage and polarity... if you used couloured wireing make sure it corresponds to the polarity, else make a label for the wire... A piece of cellotape will do.
Now connect the red probe to the other DC wire...
Hands off and power on.
It should be the oposite polarity and a similar voltage to what was on the other wire... (it may differ a little as we do not have caps on).
Ok Power off and put this part away. we now have a working, though far from optimal PSU.
Lets get to the chip.
Open this page http://www.diyaudio.com/forums/showthread.php?s=&threadid=64582&highlight=
Download the picture of the LM3875 there, it has the pins clearly marked. Break off the other pins... Make sure you now know what connects to which pin
The 5 Pins are
In+
In-
V+
V-
Out +
(Out- will be the ground tab from the first step I discussed)
Make sure your input RCA socket is isolated from the case
I attach the schematic for you... you will see a circle in it, these parts and wires are connected to the Back of the RCA socket, with the resistor from + to - in. From the RCA- a thin wire should then run to pin 7 and another to the ground tab of the case...
Ok, I'm tired of typeing now, lets see if you get this far first...
So, do not feel bad.
Now! lets get your amp up and working.
How?
We are going to break it down to the most basic form, then you can add parts to build up to the full schematic....
You need
Transformer
Rectifier
chip
input cap
the resistors from your schematic...
Connect the transformer to the bridge, if it has 2 seperate secondaries, tie the middle two together and connect that to the earth lug on your case, alternatively if it has only 3 wires, run the middle one to the ground lug. (We are sacrificeing non-essential parts, not safety).
Andrew will chip in to remind you to add a light bulb in series with the live wire on the mains before the transformer during testing.
Seperate the two DC wires comming from the rectifier, connect your multimeter's black probe to the ground tab, and connect the other lead to one of the two DC wires... make sure the other one is not going to touch anything while it is powered on.
Hands off the circuit, meter connected as described, switch on the power at the wall, observe the voltage and polarity... if you used couloured wireing make sure it corresponds to the polarity, else make a label for the wire... A piece of cellotape will do.
Now connect the red probe to the other DC wire...
Hands off and power on.
It should be the oposite polarity and a similar voltage to what was on the other wire... (it may differ a little as we do not have caps on).
Ok Power off and put this part away. we now have a working, though far from optimal PSU.
Lets get to the chip.
Open this page http://www.diyaudio.com/forums/showthread.php?s=&threadid=64582&highlight=
Download the picture of the LM3875 there, it has the pins clearly marked. Break off the other pins... Make sure you now know what connects to which pin
The 5 Pins are
In+
In-
V+
V-
Out +
(Out- will be the ground tab from the first step I discussed)
Make sure your input RCA socket is isolated from the case
I attach the schematic for you... you will see a circle in it, these parts and wires are connected to the Back of the RCA socket, with the resistor from + to - in. From the RCA- a thin wire should then run to pin 7 and another to the ground tab of the case...
Ok, I'm tired of typeing now, lets see if you get this far first...
Nordic said:
This IS a good idea.
As I have told elsewhere the electromagnetic field surrounding E-Type trafos
is larger than for Toriods.
Also Torioids (round core transformers) have such eletromagnetic distrurbance field around them
but this is SMALLER but STRONGER.
Nordic said:
I have never used any bulb for getting lower voltage from my transformer when setting up my circuits for first time.
-----------------------
I understand that AndrewT will always do this for his Amplifier Projects. But then again, this man from Scotland never does any DIY constructions of his own.
So, what he would know about this .. I do not know until proven 😕
If he ever did some amplifier, it must be the most well kept secret in this world 😀
Audio Build Regards
Lineup Practical Audio Construction Projects - With some Images
😎
It has nothing to do with voltage, the bulb is a current limit.
On idle, the amp, if all is well will only draw some milliamps and the light will only light dimly if at all, but will stil pass enough current to makr the amp run.
If there is more significant current draw on the amp, which would indicate an error, the light will shine.
At the same time due to the law of resistances in series the amp will see no more current than the lightbulb., this gives you enough time to unplug the amp before smoke starts flying...
On idle, the amp, if all is well will only draw some milliamps and the light will only light dimly if at all, but will stil pass enough current to makr the amp run.
If there is more significant current draw on the amp, which would indicate an error, the light will shine.
At the same time due to the law of resistances in series the amp will see no more current than the lightbulb., this gives you enough time to unplug the amp before smoke starts flying...
lineup, please leave the personal shots out of this. No further warnings.
yes, ma, miliamps. I've seen some of the best amplifiers in the world run off of batteries.
Even with allot of hum, should I hear any music at all?
I've read that (and I think you guys mentioned the same thing) that the rca plug should not be grounded by just contacting the chassis through its hole. Should I perhaps run electrical tape through the hole to isolate it? For a neater appearance in the the future though, where can I find well isolated rca plugs (female)?
I will try everything suggested and hope that all goes well.
BTW, I finally layed down the cash and bought the Ultimate Ears Triple fi earphones. I must say that they sound unbelievable.
Even with allot of hum, should I hear any music at all?
I've read that (and I think you guys mentioned the same thing) that the rca plug should not be grounded by just contacting the chassis through its hole. Should I perhaps run electrical tape through the hole to isolate it? For a neater appearance in the the future though, where can I find well isolated rca plugs (female)?
I will try everything suggested and hope that all goes well.
BTW, I finally layed down the cash and bought the Ultimate Ears Triple fi earphones. I must say that they sound unbelievable.
- Status
- Not open for further replies.