Stabist said:
Well - my experience with samples - I've sent a message with order for 4 LM3875
That's what I've got as a reply today:
"Dear Primoz,
Thank you for the sample request from National Semiconductor.
We regret that we are unable to fulfill the sample request as follow:
Part Number : LM3875T for 4 pieces.
Company Name: Primoz.
Application : Consumer.
Please contact your local distributor to obtain sample:
http://www.national.com/contacts/
Regards,
Regional Administrator"
Hmm
The thing is - all others from our Country - that got same message - couldn't get samples - 'cause local distributor isn't prepare to send them ...
So ... I'll guess I'll have to find another way to get them ...
Hi Stabist
I had exactly same experience here in South Africa, several times, so I had no other choice to order from RS. Their price here is about EUR9.40 for the LM3875. Yesterday I tried again to get 4 of these samples from National and 1Hr. later I got e-mail to say my order was on it's way
So try again, maybe you are also lucky 
Bobken said:
....
For anyone else who is serious about obtaining the best sound in their system, I would strongly recommend that they substitute something else (like MCBs) for any RCDs in the mains supplies to their equipment, and I don't believe that anyone doing this will regret it!
....
Regards,
What is MCB? Or RCD?
http://www.google.com/search?hl=en&ie=UTF-8&oe=UTF-8&q=MCB&btnG=Google+Search
argo
argo said:
What is MCB? Or RCD?
http://www.google.com/search?hl=en&ie=UTF-8&oe=UTF-8&q=MCB&btnG=Google+Search
argo
These are electrical mains safety devices:
MCB
Minature Circuit Breaker. Up to 63A (UK).
RCD
Residual Current Device. Protects the user against short circuit (earth faults) and earth leakage caused by damaged cable or faulty equipment. A RCBO is a combined MCB and RCD, protecting against earth leakage/short circuit and overload. Known as a GFI (Ground Fault Interruptor) in the USA.
How they can improve sound, I don't know.
i'm thinking of building a "monster clone" using the bpa-200 application notes, but i want to try running them all in inverting mode. could i just use a buffer to change the phase on half of them? will this cause any other problems?
i have built two gainclones already, and the inverting option wins hands down, IMO. anybody tried the bpa-200 schematic with the lm-3875?
i have built two gainclones already, and the inverting option wins hands down, IMO. anybody tried the bpa-200 schematic with the lm-3875?
JDeV said:
Hi,
The original post which gave rise to my initial comments, was did these devices (RCDs) have any affect on the sound?
My response was that they do, since this is the case, but the later comment to which I then responded, had suggested that using an RCD would not be a bad thing, universally.
However, as I tried to point out , because of the diffferent construction of these units, an MCB is *much less harmful* to the sound than an RCD, and RCDs simply spoil the sound in audio applications. This is due to the increased mains source impedance, which is unaviodable with these devices (or at least with all of them I have so far seen.)
Please note that I did not say (or imply) that either device would in any way *improve* the sound, as introducing *any* kind of interrupter is likely to harm the sound rather than to improve it.
In fact, if one was prepared to do without the protection of either *****which I do NOT recommend under any circumstances*****
there would most likely be a further improvement in the sound even over when using the rather better MCBs.
Regards,
Bobken said:
RCDs simply spoil the sound in audio applications. This is due to the increased mains source impedance, which is unaviodable with these devices (or at least with all of them I have so far seen.)
Regards,
How does a slight change is AC line impedance manage to wind its way through power transformers, rectifiers, filter caps, and in many cases voltage regulator circuits inside of our electronics? All I can imagine it would do is to somewhat lower the line voltage, but it would have to be a very small amount because the devices cannot dissipate much power. So, lets say it lowers the supply voltage by a volt. Now you take that volt and divide it by the drop through the power transformer- if there's a 240V in and 30V out, you're now talking about 1/8 of one volt. 1/8 of one volt lower going into the regulator circuits isn't going to affect their output one bit, unless the circuit is really badly designed.
How does the line voltage amount compare to the drop caused by soft-turn on circuits using thermistors that appear in many unregulated power amplifier supplies?
It seems EXTREMELY unlikely to me to have ANY audible effect whatsoever.
MR
Hi,
It is not as result of any voltage drop, which, in any case will vary by several volts depending on the time of day.
It results from the effect of the choke used in the construction of the RCD device, together with the coils of tiny wire used in series with the live and neutral lines, which form the current sensing arrangement, as I said earlier.
I simplified the description for brevity, but it is all related to an increase in the dynamic impedance of the mains supply.
As I also said earlier, it has the same effect as the 'mains conditioners' I have tried, which all tend to suppress dynamics in the sound.
The matter has been well-researched and covered a few times in audio periodicals, and I am surprised that it is not apparently very well-understood in DIY audio circles.
It may well "seem unlikely" to you, but a simple test will reveal the effect to anyone who tries what I suggested, if they already have RCDs in their mains supplies to their audio equipment.
Regards,
It is not as result of any voltage drop, which, in any case will vary by several volts depending on the time of day.
It results from the effect of the choke used in the construction of the RCD device, together with the coils of tiny wire used in series with the live and neutral lines, which form the current sensing arrangement, as I said earlier.
I simplified the description for brevity, but it is all related to an increase in the dynamic impedance of the mains supply.
As I also said earlier, it has the same effect as the 'mains conditioners' I have tried, which all tend to suppress dynamics in the sound.
The matter has been well-researched and covered a few times in audio periodicals, and I am surprised that it is not apparently very well-understood in DIY audio circles.
It may well "seem unlikely" to you, but a simple test will reveal the effect to anyone who tries what I suggested, if they already have RCDs in their mains supplies to their audio equipment.
Regards,
Newbie Gainclone building question
Having studied the many great suggestions on this list I think I'm ready to give it a try. Would appreciate help on two questions:
1.) Some designs suggest adjusting one of the resistors to minimize DC offset. Does this mean minimize DC voltage measured across the output? Under conditions of no input? And what value would be considered "good"?
2.) Some comments have suggested that some of the the gainclone variations might be picky about the inpedence of the pre-amp that feeds it. If I want to feed directly from the output of a CD player, or from the "pre-amp out" of a good but not great quality receiver, do I need to worry about this? Or are these kind of signal sources within the usual range of the gainclone designs?
Thanks!
- David
Having studied the many great suggestions on this list I think I'm ready to give it a try. Would appreciate help on two questions:
1.) Some designs suggest adjusting one of the resistors to minimize DC offset. Does this mean minimize DC voltage measured across the output? Under conditions of no input? And what value would be considered "good"?
2.) Some comments have suggested that some of the the gainclone variations might be picky about the inpedence of the pre-amp that feeds it. If I want to feed directly from the output of a CD player, or from the "pre-amp out" of a good but not great quality receiver, do I need to worry about this? Or are these kind of signal sources within the usual range of the gainclone designs?
Thanks!
- David
Re: Newbie Gainclone building question
Hi,
In the absence of any other responses so far, I hope the following will be helpful, but I haven't (yet!) built a Gainclone, myself.
1) When measuring and adjusting DC offsets like this, it is generally advisable to short the input to ground rather than leave it open circuit as you suggest. Otherwise, if there is no positive 'reference' at the input, the entire circuit can 'float' and you can end up 'chasing your tail' when making adjustments which will probably vary, otherwise, whilst doing this.
An offset of up to 100mV (or even higher) is unlikely to harm any normal speaker but, because any offset will tend to displace the cone of a conventional speaker when 'at rest' with consequent possible slightly 'lop-sided' operation, it is better to keep DC offsets as low as is practical.
For what it is worth, being a perfectionist, I like to stay within say 10mV, but it is not worth losing any sleep over an offset which is several times higher than this. I belive Peter Daniel has made reference to the offsets he has experienced somewhere in these threads, and anything he suggests for a Gainclone is going to be fine.
I have heard of people accepting 500mV (half a volt!) here, but it would not do for me!
2) I cannot speak from experience here, but it will also depend to some extent on whether you are using a volume control integral with the gainclone, as I have seen suggestions both with, and without this (i.e.fixed gain).
In general, it is a good idea to ensure that the input impedance of any succeeding gear is at least several times (say up to 10 times) higher than the output impedance of the source equipment, to ensure adequate signal transfer.
I don't know of any reason why any gear should be 'picky' about this matter, if sensible guidelines like this are observed, but if there are any peculiarities with the Gainclone, I hope someone else will add to this query.
I hope this helps.
Regards,
Edit: I missed this bit before, but you are right in that output DC offsets are measured between the output and ground.
David B said:
Hi,
In the absence of any other responses so far, I hope the following will be helpful, but I haven't (yet!) built a Gainclone, myself.
1) When measuring and adjusting DC offsets like this, it is generally advisable to short the input to ground rather than leave it open circuit as you suggest. Otherwise, if there is no positive 'reference' at the input, the entire circuit can 'float' and you can end up 'chasing your tail' when making adjustments which will probably vary, otherwise, whilst doing this.
An offset of up to 100mV (or even higher) is unlikely to harm any normal speaker but, because any offset will tend to displace the cone of a conventional speaker when 'at rest' with consequent possible slightly 'lop-sided' operation, it is better to keep DC offsets as low as is practical.
For what it is worth, being a perfectionist, I like to stay within say 10mV, but it is not worth losing any sleep over an offset which is several times higher than this. I belive Peter Daniel has made reference to the offsets he has experienced somewhere in these threads, and anything he suggests for a Gainclone is going to be fine.
I have heard of people accepting 500mV (half a volt!) here, but it would not do for me!
2) I cannot speak from experience here, but it will also depend to some extent on whether you are using a volume control integral with the gainclone, as I have seen suggestions both with, and without this (i.e.fixed gain).
In general, it is a good idea to ensure that the input impedance of any succeeding gear is at least several times (say up to 10 times) higher than the output impedance of the source equipment, to ensure adequate signal transfer.
I don't know of any reason why any gear should be 'picky' about this matter, if sensible guidelines like this are observed, but if there are any peculiarities with the Gainclone, I hope someone else will add to this query.
I hope this helps.
Regards,
Edit: I missed this bit before, but you are right in that output DC offsets are measured between the output and ground.
Re: Newbie Gainclone building question
You're referring to the resistor from the non-inverting input to ground in the inverting version, right?
If so, then yes, that resistor does help cancel out DC offset. The DC offset is produced by the voltage drop caused by the input bias current across the equivalent impedance at the inverting input, which causes a differential voltage between the inverting and non-inverting inputs which produces a DC offset voltage at the output.
The "ideal" value should be the parallel value of the feedback and input resistors. But unless your design uses a really high impedance input resistor, I wouldn't sweat it.
The LM-3875's input bias current is rated at a typical 0.2 uA. If the equivalent impedance at the non-inverting input is say, 10k ohms, you can expect a DC offset of about 2 mV, which is no big deal.
Problem is that the bias compensation resistor (the one going to ground from the non-inverting input) can actually cause stability problems as it forms a high impedance path to ground and can act like an antenna and pick up interference from the opamp's output which is fed positivley (i.e. positive feedback) to the input causing oscillation.
That's why if you use the resistor it should be bypassed with the 0.1uF capacitor to provide a low impedance path to ground at high frequencies.
So unless you're going to sweat over a couple millivolts of offset, just leave the resistor and capacitor out completely and tie the non-inverting input directly to the single (i.e. star) ground point.
se
Edit: By the way, if you're using a power opamp that has a FET input, then it's really nothing to worry about as the input bias currents will be easily a magnitude or two lower versus a device with bipolar inputs.
David B said:
You're referring to the resistor from the non-inverting input to ground in the inverting version, right?
If so, then yes, that resistor does help cancel out DC offset. The DC offset is produced by the voltage drop caused by the input bias current across the equivalent impedance at the inverting input, which causes a differential voltage between the inverting and non-inverting inputs which produces a DC offset voltage at the output.
The "ideal" value should be the parallel value of the feedback and input resistors. But unless your design uses a really high impedance input resistor, I wouldn't sweat it.
The LM-3875's input bias current is rated at a typical 0.2 uA. If the equivalent impedance at the non-inverting input is say, 10k ohms, you can expect a DC offset of about 2 mV, which is no big deal.
Problem is that the bias compensation resistor (the one going to ground from the non-inverting input) can actually cause stability problems as it forms a high impedance path to ground and can act like an antenna and pick up interference from the opamp's output which is fed positivley (i.e. positive feedback) to the input causing oscillation.
That's why if you use the resistor it should be bypassed with the 0.1uF capacitor to provide a low impedance path to ground at high frequencies.
So unless you're going to sweat over a couple millivolts of offset, just leave the resistor and capacitor out completely and tie the non-inverting input directly to the single (i.e. star) ground point.
se
Edit: By the way, if you're using a power opamp that has a FET input, then it's really nothing to worry about as the input bias currents will be easily a magnitude or two lower versus a device with bipolar inputs.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.