Update
I have ordered a big selection of Nichion Muse (several types) and Elna caps.
Particularly interesting were these current production models of inexpensive audiophile ecaps that also claimed much clarity and volume at the bass and midbass notes, because. . .
I think that these can be paralleled (bypass cap practice) with a polyester film cap 0.01uf (10nf) or a mylar film cap 0.005uf (5nf), and that should add on the high-end "warmth and air" effect for under $2 total per amp (both NFB and input filter caps).
I'll find out Thursday (or so) if this works out.
If so, it should bring solutions:
A lower gain hifi amp for Lowther/Fostex owners and
A high gain hifi amp with a level frequency response according to "human ear" documentation.
1% matching stereo pairs.
What do you think?
Hey guys! Thanks for putting up with me. I'm finally making some progress.
I have ordered a big selection of Nichion Muse (several types) and Elna caps.
Particularly interesting were these current production models of inexpensive audiophile ecaps that also claimed much clarity and volume at the bass and midbass notes, because. . .
I think that these can be paralleled (bypass cap practice) with a polyester film cap 0.01uf (10nf) or a mylar film cap 0.005uf (5nf), and that should add on the high-end "warmth and air" effect for under $2 total per amp (both NFB and input filter caps).
I'll find out Thursday (or so) if this works out.
If so, it should bring solutions:
A lower gain hifi amp for Lowther/Fostex owners and
A high gain hifi amp with a level frequency response according to "human ear" documentation.
1% matching stereo pairs.
What do you think?
Hey guys! Thanks for putting up with me. I'm finally making some progress.
Hi Daniel,
Those sound like good experiments to perform.
You might also want to try other dielectric material types for the paralleled smaller values, such as polypropylene and polystyrene. Those both have better dielectric absorption ("memory/rebound-bounce") specs than polyester/mylar, and better other specs, too, in general. And, actually, NPO ceramic scores well, in that sense, too, and might be worth trying, even though it might seem counter-intuitive, to many.
You might want to keep in mind the warnings about how paralleling a film cap (which typically has VERY low ESR) with an electrolytic has the possibility of introducing unwanted high-frequency resonances, which would also be affected (and maybe effected) by wire/trace impedances. There is what I consider to be a great thread about that, here: http://www.diyaudio.com/forums/showthread.php?s=&threadid=106648 .
Those sound like good experiments to perform.
You might also want to try other dielectric material types for the paralleled smaller values, such as polypropylene and polystyrene. Those both have better dielectric absorption ("memory/rebound-bounce") specs than polyester/mylar, and better other specs, too, in general. And, actually, NPO ceramic scores well, in that sense, too, and might be worth trying, even though it might seem counter-intuitive, to many.
You might want to keep in mind the warnings about how paralleling a film cap (which typically has VERY low ESR) with an electrolytic has the possibility of introducing unwanted high-frequency resonances, which would also be affected (and maybe effected) by wire/trace impedances. There is what I consider to be a great thread about that, here: http://www.diyaudio.com/forums/showthread.php?s=&threadid=106648 .
Egads!
So, my pillow, those darned little yellow capacitors, a nasa space suit, and the shiny packaging for my favorite ice cream sandwich are all related. That's why those pillows get so hard after a while. Egads! Obviously, one shouldn't put any of those things into a clothes dryer. LMAO!! Ice cream sandwich in the dryer!!
Its nicely surprising to find out that this is only remotely related to those pants from the 70's that make your shoe stuck if you try to run, the lovely little green capacitors, and my car tires. And, all of these are potentially safe in the clothes dryer. Aha! That's the difference.
I laughed so hard that it made me snort! Thanks!
So, my pillow, those darned little yellow capacitors, a nasa space suit, and the shiny packaging for my favorite ice cream sandwich are all related. That's why those pillows get so hard after a while. Egads! Obviously, one shouldn't put any of those things into a clothes dryer.
LMAO!! Ice cream sandwich in the dryer!! Its nicely surprising to find out that this is only remotely related to those pants from the 70's that make your shoe stuck if you try to run, the lovely little green capacitors, and my car tires. And, all of these are potentially safe in the clothes dryer. Aha! That's the difference.
I laughed so hard that it made me snort! Thanks!
Well, to be more specific I believe Mylar only refers to the film form of polyester.
Possibly your perceived difference between caps called Mylar and polyester is merely the difference between capacitor brands and manufacture, as in all types of cap? I haven't looked into it as it's not important to me a the moment.
Cheers
Stuey
Possibly your perceived difference between caps called Mylar and polyester is merely the difference between capacitor brands and manufacture, as in all types of cap? I haven't looked into it as it's not important to me a the moment.
Cheers
Stuey
The clarification
'........Mylar is often used to generically refer to polyester film or plastic sheet. However, it is a registered trademark owned by Dupont Tejjin Films for a specific family of plastic sheet products made from the resin Polyethylene Terephthalate (PET). The true generic term for this material is Polyester Film or Plastic Sheet.
..........'
from : http://www.grafixplastics.com/mylar_what.asp
'........Mylar is often used to generically refer to polyester film or plastic sheet. However, it is a registered trademark owned by Dupont Tejjin Films for a specific family of plastic sheet products made from the resin Polyethylene Terephthalate (PET). The true generic term for this material is Polyester Film or Plastic Sheet.
..........'
from : http://www.grafixplastics.com/mylar_what.asp
My 1875 up and running.........
Last night after weeks of delay I switched on my LM1875 amp. basically I used exactly the same board I used for the TDA2050.
All parts are identical except the power amp chip. So I can now compare both chips with exactly the same parts.
The 1875 is VERY good ! IMHO it sounds much better than the TDA2050 which by itself I had found to be very good earlier.
I think the LM1875 sounds better than my LM3886 amp in my system! I drive a Mission 701 which is a decent 8 ohm rated speaker. By that I mean it's impedance does not dive below 6 ohms anywhere and so is an easy load.
I use no preamp and the signal comes directly from an Alps blue Velvet pot from a Zaholu DAC.
Anyone who wants to know what the LM1875 is like MUST listen to one on his/her own system. How does this teeny weeny chip compare with Krells or others like that "at low volume" ?
Nothing wrong in trying it that way I guess. We aren't trying to say $1 = $10,000/- but HOW different do they sound at low levels ? I want to know , just out of curosity .
I can do a comparison with a Bryston4B on Totem Mani2 speakers . But I think the Mani2 is a low impedance speaker and needs lots of drive. Maybe I'll take my Missions across.
Can anyone make some comaprisons with other amps and report on the outcome. We would expect the megabuck amps to sound better , but what is it that sounds better ? Anyone ?
Edit: I might add that I get very good deep bass from this chip. No lack of power at all at the levels it can provide. That will not be stomach pounding bass but "full" sound at reasonable listening levels.
Last night after weeks of delay I switched on my LM1875 amp. basically I used exactly the same board I used for the TDA2050.
All parts are identical except the power amp chip. So I can now compare both chips with exactly the same parts.
The 1875 is VERY good ! IMHO it sounds much better than the TDA2050 which by itself I had found to be very good earlier.
I think the LM1875 sounds better than my LM3886 amp in my system! I drive a Mission 701 which is a decent 8 ohm rated speaker. By that I mean it's impedance does not dive below 6 ohms anywhere and so is an easy load.
I use no preamp and the signal comes directly from an Alps blue Velvet pot from a Zaholu DAC.
Anyone who wants to know what the LM1875 is like MUST listen to one on his/her own system. How does this teeny weeny chip compare with Krells or others like that "at low volume" ?
Nothing wrong in trying it that way I guess. We aren't trying to say $1 = $10,000/- but HOW different do they sound at low levels ? I want to know , just out of curosity .
I can do a comparison with a Bryston4B on Totem Mani2 speakers . But I think the Mani2 is a low impedance speaker and needs lots of drive. Maybe I'll take my Missions across.
Can anyone make some comaprisons with other amps and report on the outcome. We would expect the megabuck amps to sound better , but what is it that sounds better ? Anyone ?
Edit: I might add that I get very good deep bass from this chip. No lack of power at all at the levels it can provide. That will not be stomach pounding bass but "full" sound at reasonable listening levels.
Re: My 1875 up and running.........
Love to make some comparisons! The LM1875 is capable of quite a variety, most of which is good, and "could" include stomach pounding bass. However, I need your help to make comparisons. To do that, I need to make an LM1875 that sounds like yours.
So, what does your circuit diagram look like? And, what specific (extremely specific) capacitor is used for Ci? What's the voltage for power?
Thank you!!
ashok said:Last night after weeks of delay I switched on my LM1875 amp. . . . The 1875 is VERY good!. . .
. . .I use no preamp and the signal comes directly from an Alps blue Velvet pot from a Zaholu DAC. . . .
Anyone who wants to know what the LM1875 is like MUST listen to one on his/her own system. . . Can anyone make some comaprisons with other amps and report on the outcome. We would expect the megabuck amps to sound better , but what is it that sounds better ? Anyone ?
Edit: I might add that I get very good deep bass from this chip. No lack of power at all at the levels it can provide. That will not be stomach pounding bass but "full" sound at reasonable listening levels.
Love to make some comparisons!
The LM1875 is capable of quite a variety, most of which is good, and "could" include stomach pounding bass. However, I need your help to make comparisons. To do that, I need to make an LM1875 that sounds like yours. So, what does your circuit diagram look like? And, what specific (extremely specific) capacitor is used for Ci? What's the voltage for power?
Thank you!!
Re: Re: My 1875 up and running.........
At first I owned an Audiolab 8000 and when that died, I build an amp using LM3886. It did actually sound more detailed and correct. Now I’m using one LM3886 pr. unit in my speaker (with active filter) and I have just build an amp using LM1875 for an other speaker.
My thoughts are jumping around building an non feedback amp using discreet components, and I’m already sure that it will sound better than the LM3886’s. Just because of your argument
BWRX said:
Often, the unmeasurable euphoria that comes from building an amp yourself can make it sound better than just about anything else, except for the next amp you build
At first I owned an Audiolab 8000 and when that died, I build an amp using LM3886. It did actually sound more detailed and correct. Now I’m using one LM3886 pr. unit in my speaker (with active filter) and I have just build an amp using LM1875 for an other speaker.
My thoughts are jumping around building an non feedback amp using discreet components, and I’m already sure that it will sound better than the LM3886’s. Just because of your argument
Re: Re: My 1875 up and running.........
Wise!!
I actually do find the lm1875 charming (depending on Ci). It also runs on $28 worth of power parts, so that's nice too. Perhaps LM1875 is so interesting because of the wide variety of ways that it can be surprisingly effective?
BWRX said:Often, the unmeasurable euphoria that comes from building an amp yourself can make it sound better than just about anything else, except for the next amp you build
Wise!!
I actually do find the lm1875 charming (depending on Ci). It also runs on $28 worth of power parts, so that's nice too.
Perhaps LM1875 is so interesting because of the wide variety of ways that it can be surprisingly effective?
Hey , my post generated some interesting response !
BWRX : Thanks for your observation and I think it's quite valid. But in my case I've been building amplifiers for over 35 years ! So it certainly isn't the first and I couldn't even count how many I made. But then you couldn't have known that.
BDEF : Yup. I found my LM3886 sounds better than my friends Audiolab 8000A which is still working ( and now hardly used !).
Danielwritesbac : You are right , you need to build it exactly the same way to determine what my build is all about. Send me a PM and your postal address and I'll send you a pcb and some parts . I have some spare.
( ashokm(AT)sify(DOT)com ) Change the AT and DOT ... you know that already.
This will be interesting !
LOHK : Great suggestion . I will try this as soon as I can pick up some batteries and determine what to do about the capacitors.
I will need to build a battery charger also. Pity, that will slow things down.
Cheers.
BWRX : Thanks for your observation and I think it's quite valid. But in my case I've been building amplifiers for over 35 years ! So it certainly isn't the first and I couldn't even count how many I made. But then you couldn't have known that.
BDEF : Yup. I found my LM3886 sounds better than my friends Audiolab 8000A which is still working ( and now hardly used !).
Danielwritesbac : You are right , you need to build it exactly the same way to determine what my build is all about. Send me a PM and your postal address and I'll send you a pcb and some parts . I have some spare.
( ashokm(AT)sify(DOT)com ) Change the AT and DOT ... you know that already.
This will be interesting !
LOHK : Great suggestion . I will try this as soon as I can pick up some batteries and determine what to do about the capacitors.
I will need to build a battery charger also. Pity, that will slow things down.
Cheers.
A quick observation.
I took off the input capacitor making it dc coupled. Due to the Elco on the negative feedback loop the dc gain is now 1. I get the offset at the speaker as 7 and 6 mV .
The mid range improved and was a bit smoother but the bass end lost "punch" ! Still very good but the input capacitor seemed to increase the transient bite
A note on input capacitors. I compared a few types with this board and ranked them in order of what I thought sounded best.
When I did this with another amp ( different chip ) the capacitor order was different. So what's good in one place sounds different in another application !! I haven't seen anyone explain that .
So plenty of things to explore!
I took off the input capacitor making it dc coupled. Due to the Elco on the negative feedback loop the dc gain is now 1. I get the offset at the speaker as 7 and 6 mV .
The mid range improved and was a bit smoother but the bass end lost "punch" ! Still very good but the input capacitor seemed to increase the transient bite
A note on input capacitors. I compared a few types with this board and ranked them in order of what I thought sounded best.
When I did this with another amp ( different chip ) the capacitor order was different. So what's good in one place sounds different in another application !! I haven't seen anyone explain that .
So plenty of things to explore!
Lohk, currently I am looking at experimenting with a D cell battery set-up like the final labs, using a pair of my microclones. They sound brilliant on 12V SLAs but I know from experiments that the optimum sound wise for the LM1875 chips seems to be around 16-18V which is easy to get using D cells. I have already bought the bits.
The life of the cells should be OK as when driving 94db efficient speakers you really are just idling along at normal listening levels. I suspect that it would be good to put a small non electrolytic cap across the +/- supply rails at the battery pack and use a bank of low value/low ESR caps between the power pack and amp module with small caps non electros at the amp module. From what I can tell the noise produced by the battery pack is of high frequency but really low amplitude so it should be easy to get a very very clean supply.
All of this is easy and cheap to try, I can get d cells for just 50 each now so using 12 per rail only cost $12.00. Compared to a really good mains set-up this is actually not dear at all, though a little bit wasteful of course.
The thing I really like about off the grid supplies is the consistency, I have tried all sorts of mains supply set-ups, but the system always sounded very different depending on the time of day etc.
I suspect the biggest improvement however will likely be realised by running my phono and pre-amp on the d cells, the one thing I am totally convinced about is LM1875 based amps need really good source signal to shine as they reveal everything!
I will certainly post my findings when done.
Cheers
The life of the cells should be OK as when driving 94db efficient speakers you really are just idling along at normal listening levels. I suspect that it would be good to put a small non electrolytic cap across the +/- supply rails at the battery pack and use a bank of low value/low ESR caps between the power pack and amp module with small caps non electros at the amp module. From what I can tell the noise produced by the battery pack is of high frequency but really low amplitude so it should be easy to get a very very clean supply.
All of this is easy and cheap to try, I can get d cells for just 50 each now so using 12 per rail only cost $12.00. Compared to a really good mains set-up this is actually not dear at all, though a little bit wasteful of course.
The thing I really like about off the grid supplies is the consistency, I have tried all sorts of mains supply set-ups, but the system always sounded very different depending on the time of day etc.
I suspect the biggest improvement however will likely be realised by running my phono and pre-amp on the d cells, the one thing I am totally convinced about is LM1875 based amps need really good source signal to shine as they reveal everything!
I will certainly post my findings when done.
Cheers
Zero One said:
Hi Zero One!! Thanks for starting such an interesting discussion! I like your battery idea, but I have yet to apply it in a seemly fashion. For sure, some serious rocking out with the little suckers giving a PA rig the what for, powered just on the car battery. . . is great fun, but then the car won't start. I didn't want to buy the "D" cells, so I thought I'd use something rechargable.
Well, it IS rechargable. They (LM1875) are wonderfully adapatable. For instance I've been using them for rear channel (prosound style) with reverb/delay/expander/aphex. . . because their clarity still shines through.
I think its amazing that our applications are perhaps opposite. . . while the LM1875 does both so nicely.
ashok said:A quick observation.
I took off the input capacitor making it dc coupled. Due to the Elco on the negative feedback loop the dc gain is now 1. I get the offset at the speaker as 7 and 6 mV .
The mid range improved and was a bit smoother but the bass end lost "punch" ! Still very good but the input capacitor seemed to increase the transient bite
A note on input capacitors. I compared a few types with this board and ranked them in order of what I thought sounded best.
When I did this with another amp ( different chip ) the capacitor order was different. So what's good in one place sounds different in another application !! I haven't seen anyone explain that .
So plenty of things to explore!
The physical size of the capacitor can really vary its effects. Generally, I like the 100v and larger caps at Ci. Perhaps the worst is Xicon, but maybe it isn't the worst because they "undersize" their caps, so their 100v is probably "really" a 50v cap. Its the cool "new and improved" modern thing to do--make the caps small sized by just swapping the label. Well. . .
Anyway, I'd hazard a guess, that if the bass end loses punch, then the cap is physically too small (applies to electrolytic caps) or it is more efficient than intended (explained in a bit).
It is rumored that Panasonic (National Semiconductor) makes specific caps for C and for Ci, just for the LM chips. I have not found out more.
Older amplifiers had 100k layouts, partly because the in-series resistance also goes up to about 2.2k. This additional in-series resistance would occur at C and Ci in the Gainclones, and it can reduce the variety of ESR effects in capacitors, C and Ci.
One way to reduce capacitor variety is to purposefully increase its resistance (with a resistor) to decrease the variety of "average" ESR (by dropping the capacitor's efficiency a bit).
That's useful when interviewing several capacitors.
Another reason that 100k layouts were previously so popular, is that you could have a 25k load at the input (pot or resistor), as a more seemly place for DC from the source, to find a ground.
The topic of resistors can be quite confusing because ESR in capacitors isn't a level frequency response. So, different values for in-series resistance (the R in RC), will also effectively "aim" the capacitor's frequency response curve. . . at a different point in the audio band. That's such a nice trick for a tweeter or even a zobel, but not much fun in the amplifier's input and NFB.
Did that help?
Edit: See the bass boost circuit here: http://www.amb.org/audio/mmm/mmm_sch.png for the left channel. Well, its possible that a capacitor could do that whole little circuit (or any sort of weird boost) all by itself.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.