Thoughts On LM1875 Gainclones

[BWRX]

Originally posted by danielwritesbac
Oh yes, I've tried dual mono. There's practically no difference.

You mentioned that changing the preamp improved imaging but dual mono amps did not, so that would point to a problem (possibly grounding, wiring, source/preamp impedance mismatch?) with either your preamp or source.

My intent is not to devalue your opinions, I'm just saying that these should be able to provide excellent imaging, which is contrary to what you've experienced.

[mravlcax]

Zero zone could you post schematic for the setup that you use?

Regard Aleš

[paulb]

For you LM1875-heads: ever considered a composite amp design?

[danielwritesbac]

Originally posted by BWRX


You mentioned that changing the preamp improved imaging but dual mono amps did not, so that would point to a problem (possibly grounding, wiring, source/preamp impedance mismatch?) with either your preamp or source.

My intent is not to devalue your opinions, I'm just saying that these should be able to provide excellent imaging, which is contrary to what you've experienced.

Imaging? No, I usually test an amplifier's abilities for soundstage depth and width, in mono, and then question: Does it draw pinpoint attention to 1 loudspeaker or is it a generous presentation? It fails to compete with its nearest peers, but, the gainclone's clarity is a godsend for hearing detail on TV/HT/Movies. That's were it makes a large and positive difference.

I am curious in this mention of preamp impedance mismatch.

My preamp originally fed a power amp that was 2.2k series to 4.7uf series to 100k parallel (to power amp).

Now it is spec for LM1875, 1k series to 1uf (now 2uf) series to 22k parallel (to lm1875).

This particular preamp terminates in a virtural A30k dual (which is what the leads measure with power off and the knob all the way up). So, I am somewhat worried that I might should have used 8k series (above) instead of National spec 1k series (above).
Mismatch? What do you think?

[BWRX]

Originally posted by danielwritesbac
Imaging?

Sorry, I should have said soundstaging and imaging.

No, I usually test an amplifier's abilities for soundstage depth and width, in mono, and then question: Does it draw pinpoint attention to 1 loudspeaker or is it a generous presentation?

When you say testing in mono it is safe to assume you mean testing both channels using the same signal, correct? If so, how could that possibly draw attention to 1 loudspeaker? The image should be precisely centered between the speakers.

In my experience (honestly not that much!) the speaker placement and the room have much more effect on soundstaging than amps do. Speakers placed close to the wall can have great imaging but can lose the soundstage depth. My monitor speakers have this effect when placed on my desk about 6 inches from the wall but regain the soundstage depth if placed a few feet away from the wall. This effect remains the same whether I'm using the LM3875 amps or some low power class d amps.

Originally posted by danielwritesbac
I am curious in this mention of preamp impedance mismatch.

By impedance mismatch I wasn't referring to the RF type of impedance mismatch. I was simply referring to an old rule of thumb where the input impedance of an amplifier should be about 10 times higher than the ouput impedance of the preamp stage to avoid possible bandwidth related problems.

You haven't said exactly how your LM1875s are wired up, but it sounds like your source is driving a 30k pot with the wiper connected to a 1k resistor in series with a 2uF cap with a 22k resistor to ground at the input of the LM1875. The non-inverting input is high impedance and shouldn't pose any problem, but the inverting input is a lower impedance and could change the frequency response of the amp as the position of the pot is changed.

[danielwritesbac]

Originally posted by BWRX


Sorry, I should have said soundstaging and imaging.

When you say testing in mono it is safe to assume you mean testing both channels using the same signal, correct? If so, how could that possibly draw attention to 1 loudspeaker? The image should be precisely centered between the speakers.

Believe it or not, the dispersion and forwards/rearwards projection can be judged in mono (with 1 speaker and a mono signal). . . I did comparison to Tripath and Thompson.
The gainclone had a smaller size presentation horizontally, and it didn't make any presentation rearwards of the speaker.

Thus the following recommend:
Pair LM1875 with oversize input cap (at least 2.5uf, if not 3.3uf polyproplene--see digikey Panasonic Polypro) and with power of 36vct and big heatsinks or 38vct and huge heatsinks, either heatsink with Artic Ceramic paste. LM1875 doing the stereo pair. An alpa A20k or a better quality A50k should be satisfactory.
And
One thompson amp (no diode-reliant models--no thru-cap output models) with undersize input filter caps in a simplisitic "Y" to input both left and right, summing to double what is similar--for a center channel speaker. This needs a firm load like A10k.

Whatsit do?
This combo uses the clean gainclone to reproduce stereo and one thompson to fill in the missing centerpoint information. Its absolutely stunning, like headphones and speakers at the same time.

In this example, the centerpoint speaker may set on top of a subwoofer if such is in use.

Subwoofer:
However, be sure to put a 2nd order crossover on a subwoofer even if a plate amp is in use (especially then), along with a 1st order (just a cap) feeding a fake tweeter (just an 8 ohm sandcast). That will make for expected results from a LF unit.
Thompson amps are very beautiful on bass and none of them are strong enough. Oh well. So, the sub deal is up to you. You can un-boom a powered sub with a DBX 120A unit (line level).

I hope this gives you some project ideas.

[Bob_Sled]

Originally posted by paulb
For you LM1875-heads: ever considered a composite amp design?

paulb,

Do you have a layout for that?
I would be very interested in trying it, if I may.

[paulb]

Originally posted by Bob_Sled

paulb,
Do you have a layout for that?
I would be very interested in trying it, if I may.

I don't, except maybe a paper copy of the pattern (I'll look). The schematic shows a bridged version with servo DC feedback; you could simplify it to just the composite amp itself. It was from a Popular Electronics article, I think, from long ago, I will check for this as well.

Are you living in the Snowbank they just auctioned off?

[paulb]

I have a paper copy of the PCB pattern, which would need to be retouched a bit. I also have the original article, Electronics Now, Nov. 1992. Email me if you'd like a scanned copy of either.

Paul

[danielwritesbac]

A paper copy of the pcb? That sounds very interesting!!
I'd sure like to be able to use 2 regular speakers to do the job of three regular speakers or 2 horns. (I'm saying that I'd like electronic means to un-cramp the amp rather than physical, because, hey, I could use the extra room).
Looking forward to seeing this "assisted" LM1875.
Maybe the circuit can work with LM3886 too?
Can you post the photo?