Beginner's Gainclone, HiFi LM1875, The Amplifier Board

Here's the "color" marked photo showing the RF blocker and the 1M resistor.

Green is ground
Yellow is input signal +
 

Attachments

  • lm1875 on veroboard rf blocker.jpg
    lm1875 on veroboard rf blocker.jpg
    67.6 KB · Views: 3,302
Optional Component, Output Zobel

Because a speaker wire is like a big antenna, and it leads directly into the NFB, then an amplifier could get some noise in through the speaker leads. A speaker output zobel helps block some of that. This is a noise reduction component.

In addition, the Speaker Output Zobel serves to enhance the stability of the amplifier.

This zobel shouldn't be within the audio band, but rather, just above.

For this, I've selected:
10R (10 ohms) and this one is brown, black, black,
plus
10nF, which is 0.01uF and probably has "103" printed on it.

Zobel. That's a resistor and capacitor, "RC," as a load.
This load is upon the speaker output.

See picture.
 

Attachments

  • lm1875 on veroboard zobel1.jpg
    lm1875 on veroboard zobel1.jpg
    89.9 KB · Views: 3,206
Edits said:
Hi Dan,
Can not wait the sequel! Compiled those parts, but still miss the 27K feedback resistor. Can I substitute with 20k, 22k? or 33k?
That is the one I'll try.

Okay! Now that its done, we can mod it if you like.

Its cross-compatible with the AudioSector resistor values, so 22k with 680R can work for the NFB. That's gain 32.

If your source/preamp is really strong or if you've used less than the recommended voltage, then you can use:
20k with 680R = Gain 30
22k with 820R = Gain 28
20k with 820R = Gain 26

*There may be slight differences in DC offset with different gain settings, and the chipamps do vary between individual samples.
Do check with your meter, at the amplifier's speaker output.

**I set a rather high gain of 35 (that's 27k with 820R) to help modern sources with digital volume controls that shouldn't be run "maxed out" because they are often op amp chips.

The National Semiconductor Overture Design Guide spreadsheet does work with with LM1875. Dial in LM1876, which is two LM1875's and an extra circuit crammed onto one LM1876 chip. The resulting designs will work for LM1875.
 
casiomax said:
hi Daniel,

personally i like my input cap to be poly film, i dont know about anybody else, but my guitar sounded much better.

Oh yes! You can use Polypropylene, Polyester, Ceramic, Electrolytic, and botique types. And, you can use them solo or in combination.

One combination, was recommended by "that very helpful guy who earth-grounds everything very thoroughly," and that suggestion was 4.7uF + 0.47uF.
I haven't tried it yet, because I was pretty happy with just 0.47uF part.

So, do what ever sounds good. ;)
 
eketehe said:
Wow, my favourite :)
Well, as said, max +/-30vdc, put big trafos... they're will not clipping.

Would you mind posting the schematic for your LM1875 amps?

I just got some very unusual results with mine. Its lovely, but the output power doesn't match up with the PDF. It wasn't the least bit reserved about pumping out a fierce amount of power. Does yours do that too?
 
danielwritesbac said:


Would you mind posting the schematic for your LM1875 amps?

Sure Dan,
Here belows..
the Input caps is optional ( and critical ), if the woofer can not go or bad sounding on 20Hz-100Hz, 0.68 - 1uF will be OK.

I just got some very unusual results with mine. Its lovely, but the output power doesn't match up with the PDF. It wasn't the least bit reserved about pumping out a fierce amount of power. Does yours do that too?

can you specify?
sorry i got a technical subtittle problem ( again) :xeye: ;)
 
eketehe said:
can you specify?
sorry i got a technical subtittle problem ( again) :xeye: ;)

Well, I think mine runs all the way to its theoretical 48 watt limit. It couldn't be doing more, because the thermal capacity won't allow.

Personally, I had used 4 ohm speakers; however, they're a 1st order crossover with BSC and lots of inductors, so "maybe" (and probably) it turned out to be more like a 6 ohm load.

On either 8 ohms or 4 ohms, it does 30 watts. However, its the 6 ohm load that can "coax" the LM1875 up to the 40's watts output. The LM1875 also appreciates the "difficult" 8 ohm speakers and will gladly hit them with a bit extra power, instead of faltering like LM3875. I'm not saying that one is more or less powerful than the other--only that chipamps are sensitive to speaker selection.

Oh thanks for posting the schematic! That's a "hi-ref" style, and a merry fight to find the "right" caps for it. I'll bet that you did exactly that, and that your amp is really wonderful.

Thanks again!
 
Re: MISSED

eketehe said:
the schematic

Oh, you'll notice that I left a big space in the top-left corner on the board of the amplifier that we designed here.

In case of DC offset (chips do vary), that space is for your 22uF cap (known as Ci on schematics). See how the ground from the NFB travels right over that space?
The + side of the cap faces the amplifier (and hooks to my 820R), while the - side hooks to input star ground. This "intercepting" the NFB ground with a 22uF cap does prevent the amplifier from being able to amplify DC.

However, selection of a "seemly sound" for a cap at position Ci, isn't easy, because it is used by the audio signal in direct proportion to the gain, so that any abberation in the cap is magnified/amplified.

There is the reason that I made the project without Ci, by using values similar to the Audiosector board (that doesn't use Ci).

There's a space for it if you need it.
 
Re: Re: MISSED

danielwritesbac said:
However, selection of a "seemly sound" for a cap at position Ci, isn't easy, because it is used by the audio signal in direct proportion to the gain, so that any abberation in the cap is magnified/amplified.

Error: That figure isn't 35 times, but rather a proportion of the voltage divider (two resistors, one hot, one ground, is a voltage divider, also called "fixed L-pad" by the speaker guys).

So, using a cap (at Ci) when you need to remove DC offset, isn't as frightening as it seems.

However, you do already have one capacitor to "select" for the input filter cap, and I just wanted to be clear that if you also use a cap at Ci, then that doubles the selection efforts.

Hey!! I think we covered it! ;)

P.S. Thanks eketehe, for your help and for the examples.
 
danielwritesbac said:
For example, here's an LM1875 that I built on veroboard (phenolic, no pads)

I should point out that Veroboard™ has copper strips running along the board (hence the 'stripboard' name applied to both the genuine Vero product and its clones); what you've used is perfboard, available plain or with a copper ring or square round each hole to aid component fixing.

I'd hate someone to order the wrong thing. ;)
 
cpemma said:


I should point out that Veroboard™ has copper strips running along the board (hence the 'stripboard' name applied to both the genuine Vero product and its clones); what you've used is perfboard, available plain or with a copper ring or square round each hole to aid component fixing.

I'd hate someone to order the wrong thing. ;)

Thank you so much! What I ordered was that brand, but with no copper at all. Anyway, "Perfboard" it is, and thanks again.

Say, how did you like the little amp?
 
Power supply schematic

I'm back home temporarily and received a few questions on email. These had much to do with electronic dampening properties.

So, for that, one can try a 0.47uF input filter cap (some sort of poly).

And/Or, one can try the attached speedy power supply schematic.
Its set up for stereo.
The LM1875 in dual mono doesn't require 4x 2200uF caps per each rail. Half that much (2x 2200uF caps per each rail) will do fine.

Edit: This schematic visually resembles the actual power supply as it would be built on Perfboard (no-copper phenolic board), and possibly using Mallory SEK 50v caps or something speedy from Nichicon.

This schematic (click attachment) represents a basic form of power supply, and although simple, has provided excellent performance for me, and it has continued to do so, even in comparison to more complex alternatives.
There are many power supply options available for this amplifier, and this schematic provided is just one of the many available.

The bridge rectifier unit pictured is a KBPC1004 or KBPC1005, with clearly marked connections and an all-plastic body (safer than usual). KBU1004 (looks like a fork) also works well and also features clearly marked connections and an all-plastic body.
Feel free to use the famous KBPC2504 rectifier, but be aware of its metal shell. And, do please use appropriate safety measures that are copesetic with your local regulations.

Click. ;)
 

Attachments

  • ladderback schematic.jpg
    ladderback schematic.jpg
    67.2 KB · Views: 3,351