LM3875 Poweramp- reflections on components...

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi,

What's "lawfaked" mean?

Lawfaked refers to making a linear LAW pot behave more or less like a log LAW one.

What the LAW means is that at halve rotation you have around 20db attenuation (log) vs. 6db attenuation (lin) and at further rotation the attenuation becomes progressively more.

A "usual" methode of lawfaking to take a resistor appx. halve the value of the "nominal" potentiometer value if a log law pot was used and use a linear pot with around 10 - 20 times the value of the resistor. The resistor goes from wiper to ground, the pot is connected normally.

If the resistor is 10k and the pot is 100k the attenuation at halve rotation is around 17db, close to the normal log curve, the minimum input impedance (volume all the way up) is around 9kOhm, with a few db attenuation it is above 10k.

If we made the resistor 22k and the pot 470k we would have 22db attenuation at halve rotation and a minimum impedance of 21K.

Clearer?

Sayonara
 
Kuei wrote:
“Transformer based Attenuation”

What is the price?

Kuei wrote:
“Good quality Cermet/Carbon track Linear Pot lawfaked”

The problem with it is that lawfaking resistor should have small enough value (to be effective). That means at higher levels you’ll get relatively low input impedance. Further, that means tubes can not drive it.

Kuei wrote:
“Single Series Resistor Single Shunt Resistor Attenuator”

Is this volume control or what?

Kuei wrote:
“True "Ladder" Attenuator”

It needs one whole day of soldering to sort it and cost is not negligible. It may look nice you have only one resistor at signal path (and only one shunt resistor), but truth is also you have two switches at signal path. I’ve never compared it directly with something else (just listened with and without it, difference was very small), but some reports I can not ignore say it is not any better than Black APLS.

Kuei wrote:
“Series Resistor Chain Attenuator”

Cheaper and much easier to sort than ladder attenuator. Not bad idea, if you have small enough number of resistors and if you can live without continuous control.

Kuei wrote:
“Conductive Plastic Log Pot (linear w. Lawfaking not applicable due to distortion)”

??? :scratch:

Bob, where is that thread?

Pedja
 
Hi Pedja,

I don't agree with the 'order' of choices mentioned by KYW, but as I have covered the subject almost to death in the other post, I didn't wish to waste any more time on these points or any further unnecessary arguing here.

The thread I posted in a lot was "passive preamp" and within that post, there was a reference by Dave the Moderator, to an earlier Post of some ten pages which also covered all of the relevant points about passives too.

Regards,:)
 
Philo said:
Got it. Then the 10K 4th lead on my new ALPS pot goes to ground for the LAW attenuation. Correct?



I very much doubt it. I don't know which Alps pot you are referring to, but I cannot imagine any possibility of a manufacturer making any (mechanical, of course, as opposed to electronic) pot which you can choose to use in a linear or log fashion. As I said earlier, the conductive tracks are made differently.(post #35)

So you will most likely have either a linear, *or* a log pot, of which one lead will need to be grounded (at least) for it to function as an attenuator.
:)

Regards,
 
The 100k log ALPS I have are NOS and have 4th leg with 10k resistance between both ends of the "windings" and is open to the wiper. I wasn't sure if this was related to "law-faking" but yours and Kuei's discussions on this hit a note that reminded me of my measurements of the pots.
 
Philo said:
The 100k log ALPS I have are NOS and have 4th leg with 10k resistance between both ends of the "windings" and is open to the wiper. I wasn't sure if this was related to "law-faking" but yours and Kuei's discussions on this hit a note that reminded me of my measurements of the pots.

Hi,

Not absolutely certain what you have here from your description, but I would put money on the fact that it is an 'older' pot, with a tapping off the track for a 'loudness' control.

In case you are unfamiliar with this, it was usually a (switched) method of boosting the lower frequencies at low volumes to accord with the Fletcher Munsen (not 100% certain of spelling) curves which relate to the ears' ability, or rather lack of ability, to hear bass frequencies very well at lower levels.

Regards,
:)
 
Yeah, I think your right about that being a loudness pot. From what I remember about the purchase the guy said he "thought" they might have a loudness boost. I guess my Alzheimers is progressing. Now, what were we talking about....? So this leg still goes to ground if I want to use the boost feature? Leave it open for normal operation?
 
Hi,

Is it scandalous then, Kuei?;)

Some people prefer to wash their dirty underwear in public and sling mud at others. I prefer to try to resolve such things directly with the parties involved instead resorting to public character assasination, mudslinging etc. It leads to nothing. Hence no no public comments.

Sayonara
 
Die-Cast Clone

Ok, I have read this whole thread just now and there are a whole bunch of interestingly good observations and viewpoints, so I will work through them in turn. ;)

Hatasa,
You have made a perfectly fine pair of amps there.
Your choice of standard die-cast construction box is a good one for sonics, is economical and practical, and also is an easily reproducable example to others - I would have done pretty much the same. I;)
A 3 inch DC fan mounted ontop each amp and powered by a seperate plug-pack power supply running at lower speed would cheaply and quietly keep things nicely cool - running cooler is good for sonics ime.
You have chosen premium components in your construction - how much each per monoblock ?.

Some tips ? - I would make conductors non-directional by running two lengths of wire twisted together and in reversed directions, and I would connect two resistors wired in parallel, but in reversed directions in each resistor position, and I would shift the whole circuit sideways so that the ends of the supply capacitors are not hard up against the cabinet, and maybe spin the IC 90* to enable slightly better layout.

Well done,
regards,
Eric.
 
Kuei Yang Wang said:


.................
2) I have come across several claims that the FEEDBACK loop on the inverted gainclone includes the source and interconnect cable. It should not require any explanation as to why this is not so. The inverting input operates simply as near zero ohm currentsink, so the feedback loop is strictly limited to the feedback resistor.
.............

Sayanora San, tell me who dared to make those claims? :mad:
And yes, I require you to explain why this is not so! :mafioso:
We need to put an end to this rumor! :headshot:

Capito,

argo
 
Hi,



Kuei wrote:
“Transformer based Attenuation”

What is the price?

Realtively high, however it is one of the best possible solutions for a passive line control unit.


The problem with it is that lawfaking resistor should have small enough value (to be effective). That means at higher levels you’ll get relatively low input impedance. Further, that means tubes can not drive it.

Tubes (well implemented) can drive a 10k/100k Pot lawfaked combo JUST FINE. The Impedances are indeed relatively low, the highest value Pot's tend to be 1M, limiting us to an effective 50..100K nominal Impedance, sufficient in most cases I'd say.


Kuei wrote:
“Single Series Resistor Single Shunt Resistor Attenuator”

Is this volume control or what?

Yes. It is exactly what it says.


Kuei wrote:
“True "Ladder" Attenuator”

It needs one whole day of soldering to sort it and cost is not negligible. It may look nice you have only one resistor at signal path (and only one shunt resistor), but truth is also you have two switches at signal path.

Yup. It costs a lot, is a lot of work and sounds worse than the simpler S2R2A, but it offers a lower output impedance for a given minimum load on the source and a more or less constant input impedance (neither lawfaked pots nor S2R2A have a constant input impedance).


I’ve never compared it directly with something else (just listened with and without it, difference was very small), but some reports I can not ignore say it is not any better than Black APLS.

I have made extensive comparisons. I have build Transfomer, Ladder ,S2R2A and series chain attenuators using identical types of resistors and switches and have also compared wirewound pots, cermet pots, carbon and plastic pots to the above.



“Conductive Plastic Log Pot (linear w. Lawfaking not applicable due to distortion)”

??? :scratch:

Conductive plastic has some interesting behaviours if there is ANY current drawn or injected on the wiper. This results in interesting low level effects. Some people actually LIKE the resulting compression effect. BTW, the same effect also applies to a lesser degree to the current flowing through the plastic track.

Conductive plastic pot's have nothing to do in good audio gear, if you ask me. But then I dislike the obvious sound effetcs and deviations from neutrality present most of the specialist so called "Audio" parts (see my views on BG's, Oil Caps, Carbon Comp and Metal Foil resistors et al...). BWTFDIK.

Sayonara
 
Hi,


I very much doubt it. I don't know which Alps pot you are referring to, but I cannot imagine any possibility of a manufacturer making any (mechanical, of course, as opposed to electronic) pot which you can choose to use in a linear or log fashion.

You don't seem to have much experience of audio electronics then. In the years up to the 1990's when electronic attenuators took over in much of the low and mid-fi gear it was very common to find pots with a linear track and one or two or three "taps" along this track.

These taps would be used to "lawfake" a near logarythmic response and also for the loudness "feature". It is much easier to make a dual (or quad/sexta) pot that tracks well with a linear track than a log one, so it makes a multitapped MUCH, MUCH cheaper to make and just adding a resistor makes virtually no dent into the budget. Selecting one in ten pots for a better than 3db channel match at low volumes does.

These old "consumer" pots also tend to have a carbon track and sound surprisngly good, considering their usual sub $ price in quantity.

Sayonara
 
Re: Die-Cast Clone

mrfeedback said:
Some tips ? - I would make conductors non-directional by running two lengths of wire twisted together and in reversed directions, and I would connect two resistors wired in parallel, but in reversed directions in each resistor position...

Um... how exactly does a conductor or a resistor come to have any particular "direction"?

se
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.