CA 340A SE LM3886 based amp - Upgrade advice please.

Status
Not open for further replies.
Hi Barry.

Sorry for the late reply, I missed your post.

I lifted the input caps and there was no buzz so its the pre section.

So far I have replaced the following components:

-Local decoupling caps on input selector chip and opamp.
-Preamp supply diodes, caps, and regs.
-Tried a different opamp.
-Tried a different pot.
-Main caps and diodes on poweramp psu.
-Checked ground traces and circuit continuity, all seem ok.
-Checked soldering, all seems ok.

There are details on this thread:

CA 340A SE fault finding.

At the moment my main hunch is the input selector chip. Other than that Im suspicious of a dud resistor somewhere.

Mike.
 
Do I need to glue my caps for any reason or is it a non-issue?

Also I have a question about the volume pot:

Currently there is a 20k pot which I'd like to replace at some point. I would like to use the Alps Blue motorized but I think its only available in 10,50, & 100k values. Which of these would be suitable and why?

Mike.
 
The glue may help with flow soldering in mass production.
From my experience hand soldered large capacitors don't need glue.
Never had a failing solder joint with them.
If the pcb is subject to vibration, things may be different.
But I think this is a non issue with your amplifier.
regards
 
Currently there is a 20k pot which I'd like to replace at some point. I would like to use the Alps Blue motorized but I think its only available in 10,50, & 100k values. Which of these would be suitable and why?

Anyone care to comment please?

Why do we get different value pots and what difference do they make to the circuit and sound?
 
High value pots will create more noise, but are useful for amplifiers with high input impedance. As a general rule, the pot should be 1/10th of the input impedance.

Why? I don't know, but maybe this will push somebody to elaborate. In your position I'd go for the 50k.
 
If the impedance of the pot is very low it will present a load that the previous stage is likely incapable of driving properly (possibly because it cannot supply sufficient current or the output impedance is significant compared to the pot impedance), so in order to present an easier load a high value pot should be chosen. Unfortunately this brings another problem in that if a high value pot is chosen (100k or 250k say) the input impedance of the following stage becomes significant compared to the value of the pot and this will interfere with the operation of the pot. A happy medium must therefore be selected - impedance matching.
 
there is aditionally the impact on common-mode distortion and the possible interaction with balance controls.
If I were up to design a preamp, I would use a 10k volume pot.
The 20k may be just right for your circuit or not - how should I know? 😕 (<- kindly asking for a schematic 😉 )
regards
 
mikesnowdon said:
I just found the specs for my TDL speakers:
TDL RTL2

How does this affect the power supply considerations?

Earlier it was suggested that a 22v trafo would be best but we assumed the RTL's were 4 ohm, apparently they are 8ohm. Any ideas?

Mike.
Yes, the whole design changes when trying to use a chipamp to feed either a 4ohm or 8ohm speaker.
You can aim for 60W into 8r0 and use two of these to feed the separate terminals on the speaker. That's your next project, to bi-amplify (not go active).
 
So its something like 'raising the voltage to get more useable power into an 8 ohm load' right? someting like 24/25v with a single LM3886 per channel? This may not be an issue as the service manual states a 24v supply before rectification.

Andrew,

How would I obtain a 60w output? The amp as standard has a manufaturer specced output of 45w/pc. I dont know how accurate this is.

I like the idea of bi-amping. The heatinks allready have tapped screw holes for an additional chip (see earlier pics). Food for thought.

Mike.
 
Status
Not open for further replies.