First gaincone LM3875. Some questions.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello! I came here with big intentions but after some reading became smarter.
I'm planning to build 2 channel LM3875 gainclone amplifer for diy speakers (theyr not built). it is useful to know that i've completed opamp project in past (cmoy). My questions are pretty basic.
Can i use just mp3 player, cd player, phone mp3 for imput or i need preamplification?
Is it strongly recommended to use quality components like quality resistors, caps and what about grounding off transformer?
How much will affect speakers sound quality and as theyr not built yet then witch ones is to build for best result? (i'm planning to build with 2 drivers)
What about speaker, amplifier power levels.

Thank you for reading and i would be more than happy to read your post!
 
Administrator
Joined 2007
Paid Member
Good quality commercial grade parts are fine, don't go mad on exotic stuff.

Power output isn't hugely important... to understand why you need to appreciate that to get a 3db increase in sound level requires a doubling of power, and that in itself gives a barely noticeable increase in "sound". Of far more importance is the efficiency of the speaker, a large floorstander @ say 93db watt will be far louder with a 10 watt amp than an 82db watt speaker with a 200 watter (I have just picked those figures out of the air before anyone does the maths :))

Grounding is of the utmost importance... and it's where 99% of DIY builds fail... and it's to vast a subject to sum up in a few lines here :)
 
93dB speakers and LM3875 will be really loud. The amp can use the signal directly out of the source, but with an active preamplifier it's better. Even a buffer does a great job.
where does all this come from?

50W adds about 17dBW to the basic 93dB/W @ 1m
This will result in ~105dB @ 2.5m listening distance from a pair of 50W amplifiers driving a pair of these 93dB speakers.

105dB is not really loud for transient peaks.

"an active pre-amp is better"
How? better sound/better voltage/better distortion/better bass/better spoken voice/better what.

"even a buffer does a great job"
How? define a great job.
 
Maybe for you 105dB isn't too loud, but for me it is. The usual listening level is around 80-84dB.

OK, I must admit that I was too short in explanations.
An active preamplifier before the 3875 will adapt the impedance between signal source and 3875's input impedance. If the latter runs in inverted mode, the input impedance will be pretty low. One possibility would be to increase both feedback resistors, but this will mean higher noise (the higher the resistor value, the higher its noise will be). If needed, the preamplifier can also have tone control, a voltage gain, balance, RIAA correction etc. If the signal source has a high output impedance, the sound will be pretty "blurred", loosing the transparency, along with a voltage decrease.

A voltage buffer i.e. voltage repeater has a voltage gain 1 (well... around 1) and it has the only purpose to adapt the impedances of signal source output and amplifier input. A standard CD/DVD player pretty much doesn't need either of those, since their output is fine, usually 1kOhm/2Vrms. With a tape deck or a turntable the situation is way different; even though it will work, the levels will be unsatisfying and it will sound like crap.
 
Maybe for you 105dB isn't too loud, but for me it is. The usual listening level is around 80-84dB.
exactly.
Average listening level is usually around 20dB below peak level.
80 to 84dB at 2.5m from a pair of speakers is about 10dB above conversation level. I would call that comfortable.

95dB is loud by my definition. and 20dB above that is 115dB. about 10dB more than 50W into 93dB/W can achieve.
 
fistly in conclusion "you" recommend building preamp. I quote "The amp can use the signal directly out of the source, but with an active preamplifier it's better. Even a buffer does a great job."
Is it needed? i've build my cmoy and in result i got small buzz and different gains due stereo pote. (my cmoy was in hammond case, witch was fully grounded to bat minus.
secondly its art to build good grounding. can you give me some tips.
Nice discussion but sorry. i'm feeling bad for not getting all what your saying.
So preamp it is! what chipto use and what about psu.

maybe you add link to project that will suite my needs. can use with mp3 player if wanted. psu and amp in one case. maybe someone have good 2 driver speaker project, refer link.
all i want is to get into diy audio. wost that can happen to me is building 150$ amp what will not satisfie me.
 
Last edited:
Administrator
Joined 2007
Paid Member
secondly its art to build good grounding. can you give me some tips.

Yes... think of all ground wires as having resistance, then ask yourself, can the current that flows in that wire or that piece of PCB print cause a volt drop that can influence the circuit in ANY way at all. Think of it as a DC coupled amp and analyse it at DC... and by common sense you can make the effect of the wiring "dissappear"

It's not an easy concept to grasp I admit, read up on it all, look at recommended wiring layouts and try and see what is being achieved.
It took me years to fully appreciate the real importance of this, but it's worth getting correct.
 
i reconsidered that. Peters circuit is just to simple for a good result. i zoomed around net and found this interesting website. http://www.fedde.nu/audio/chill-amp2.html Chillamp.
There is some problems. As you know i have intentions to build amp in one case. Aslo i dont like two transformers build. is there a chance to get rid of two transformer build. i mean for example to build cillamp 2 with chillamp1 psu.
I you have some reccomendations then i wuld more than happy to read them.
 
I like to think that i already have some experience at amp building. Secondly i will have my technical education diploma soon.
And here are some explanations:
Firstly Peter's kit is for total beginners. Peter's implementation is a stripped down version with no components to control chipamp stability.
It has no protection for DC output offset.
It can very easily destroy speakers if any fault either in the chipamp or in the source develops.

Peter's implementation uses no smoothing capacitors at the Power Supply. He has small capacitors at the amplifier.
This set up particularly suits very efficient speakers that take tiny currents from the chipamp.
im really looking for something that will challange me but at same time wont require much effort to build. I mean simulation circuit on cads and so on.
First impressions about peters was simple and easy but then i realized that i maybe need something else.
Im not saying that peters amp is bad it just doesn't suit me.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.