Silk Sense gain stage using 2 X LME49710
Here's the gain stage schematic. I have been using it for nearly a year. The main source is an E-MU TrackerPre USB DAC (up to 192 KHz / 24 bit). The power amp is LM4702 with TIP142/7 output stage.
With the volume pot at minimum the output noise is about 2 uV RMS, which is -114 dB referred to 1 V. Frequency response is 0 Hz (ie. DC) to -3 dB at about 100 KHz.
Without the RC filter, slew rate limiting kicks in at about 1 MHz with 2 V RMS output,
at 0.16 V RMS output the amp is 3dB down at 7.5 MHz, but that's all pretty meaningless for audio.
Here's the gain stage schematic. I have been using it for nearly a year. The main source is an E-MU TrackerPre USB DAC (up to 192 KHz / 24 bit). The power amp is LM4702 with TIP142/7 output stage.
With the volume pot at minimum the output noise is about 2 uV RMS, which is -114 dB referred to 1 V. Frequency response is 0 Hz (ie. DC) to -3 dB at about 100 KHz.
Without the RC filter, slew rate limiting kicks in at about 1 MHz with 2 V RMS output,
at 0.16 V RMS output the amp is 3dB down at 7.5 MHz, but that's all pretty meaningless for audio.
Attachments
Hi Glenn
thanks for the schematic!
What is the audible effect of the RC filter. Sorry, but I can't imagine that you will hear a difference (negative) if you omit the RC filter because I can hear just about 17kHz or less. It's a LPF with 100kHz.
@audioman54
Hi Mark
can you share some of your National line preamp test circuits for the LME497x0 family? Would be interesting how pros did it!
thanks for the schematic!
What is the audible effect of the RC filter. Sorry, but I can't imagine that you will hear a difference (negative) if you omit the RC filter because I can hear just about 17kHz or less. It's a LPF with 100kHz.
@audioman54
Hi Mark
can you share some of your National line preamp test circuits for the LME497x0 family? Would be interesting how pros did it!
Welcome to the club, amigo. National did something uncommonly stupid: their massive panic layoffs included some very taleneted design engineers as well as talented others like me and you. Good designers are as valuable as gold around here, and they are gone forever from NSC. This absolutely idiotic strategy just solidifies my belief that the retard Brian Halla is still on his path to destroy NSC as an analog company. He is clueless about analog and hates being in the "commodity" market where they sell low ASP parts in high volume. I hope I get to cash in my stock options before he finally runs the company into the toilet.
Because once they fire you and take away your badge, you really have no choice but to leave the company....
I don't hear anything. I have tried silver mica and polystyrene. Both are inaudible. I wouldn't use polyester or ceramic in this position.
True, but its always a good idea to low-pass limit amplifiers to make them less susceptible to Electro Magnetic, Radio Frequency or Magnetic Field interference, and instability caused by out-of-band signals entering the amplifier.
I chose 100 KHz so the full high-frequency resolution of my 192 KHz sampling USB and SACD player would pass through the preamp. (Pity the power amp and the speakers don't go quite that high, but that's a completely different issue). Its been demonstrated again and again that hires audio formats operating at much higher sample rates and bit depths than CD actually sound much better than CD.
Buy some puts on NSM -- fwiw, the street still thinks that National is an "analog" play. when you purchase puts on a stock, it "stops the clock" -- so if you are long term, you're long term, but if you were aiming to be long term and buy the puts early you get hit by the higher marginal tax. If you've got a half-way decent accountant they can really help you out here.
FYI
Brian Halla "steps down" as CEO at National Semi.
See:
National Semiconductor’s CEO Brian Halla to Retire (Update2) - Bloomberg.com
I don't know if this changes things at National Semi in any significant way.
ZAP
Brian Halla "steps down" as CEO at National Semi.
See:
National Semiconductor’s CEO Brian Halla to Retire (Update2) - Bloomberg.com
I don't know if this changes things at National Semi in any significant way.
ZAP
From the Bloomberg story:
"During his tenure, Halla refocused National on analog chips"
Halla Gone!!! Good!
Hi Everyone,
Sorry not to have reported in recently but I have been busy...babysitting the grandchildren! Oh and I also did a demo of my LME based D/A preamp and LME49811 power amp at a very, very high-end store here on the SF Peninsula. They were "Shocked" at the sound quality. (Actually said the best they have ever hear!) Talking about 0.0005% THD+N for the entire signal path did not impress them as much as listening to the 2 units together did. As several of you now know a complete LME signal path is something to behold.
Anyway I was just turned down for a “building maintenance job” yesterday and a bit depressed. (Some AV work would have been involved but some plumbing also and my lack of plumbing references cost me the job!) Going to Burning Amp on Sunday though and that should be fun. We will be showing the smaller 49811 amps and my personal D/A with several special mods (1 from Bob Pease) and a set of speakers. Still want to shoehorn in the new 4399 DAC in place of the 4396 though. Not sure if 24 vs 32 bits will matter much with a 16 bit source though but my apps manager buddy at AKM claims it will! We shall see sometime in the next week or two. Hopefully before Sunday! Is anyone going to Burning Amp? I have not been before but I am going with on of my best buddies (he designed the speakers we will be using) who has been working with me on all my projects and companies for the past 20 years as a senior mechanical engineer. He says it will be fun.
Glad to see that #$%&$^% Halla gone from National...in a couple of months. He will get to keep his options though because he will be able to retire. I however had to give up my 8k of vested options up after 3 months because I missed the 55 year old deadline when then booted me out the door by 3 months. When they decided who was going to be part of the 3500 person layoff they selected "Highly Compensated Individuals" which of course meant everyone over 50 except for anyone in management!!! #%^%@^%#@
Homemodder what job were you talking about in Europe a month ago?
Mark
audioman54
Hi Everyone,
Sorry not to have reported in recently but I have been busy...babysitting the grandchildren! Oh and I also did a demo of my LME based D/A preamp and LME49811 power amp at a very, very high-end store here on the SF Peninsula. They were "Shocked" at the sound quality. (Actually said the best they have ever hear!) Talking about 0.0005% THD+N for the entire signal path did not impress them as much as listening to the 2 units together did. As several of you now know a complete LME signal path is something to behold.
Anyway I was just turned down for a “building maintenance job” yesterday and a bit depressed. (Some AV work would have been involved but some plumbing also and my lack of plumbing references cost me the job!) Going to Burning Amp on Sunday though and that should be fun. We will be showing the smaller 49811 amps and my personal D/A with several special mods (1 from Bob Pease) and a set of speakers. Still want to shoehorn in the new 4399 DAC in place of the 4396 though. Not sure if 24 vs 32 bits will matter much with a 16 bit source though but my apps manager buddy at AKM claims it will! We shall see sometime in the next week or two. Hopefully before Sunday! Is anyone going to Burning Amp? I have not been before but I am going with on of my best buddies (he designed the speakers we will be using) who has been working with me on all my projects and companies for the past 20 years as a senior mechanical engineer. He says it will be fun.
Glad to see that #$%&$^% Halla gone from National...in a couple of months. He will get to keep his options though because he will be able to retire. I however had to give up my 8k of vested options up after 3 months because I missed the 55 year old deadline when then booted me out the door by 3 months. When they decided who was going to be part of the 3500 person layoff they selected "Highly Compensated Individuals" which of course meant everyone over 50 except for anyone in management!!! #%^%@^%#@
Homemodder what job were you talking about in Europe a month ago?
Mark
audioman54
Metal Can HA's
Jack,
Were they the metal can parts? If you get a chance and can go with a 1.2k feedback resistor in inverting mode you might try the metal can LME49713 current feedback opamps. My favorites! I am a bit biased though as I specified those parts based on the specs from my previous favorite signal path current feedback opam the AD811/AD9610. Funny that the part number for my favorite power amp driver part is now the LME49811. Still have 500 AD811's that I never used!
Mark / audioman54
Jack,
Were they the metal can parts? If you get a chance and can go with a 1.2k feedback resistor in inverting mode you might try the metal can LME49713 current feedback opamps. My favorites! I am a bit biased though as I specified those parts based on the specs from my previous favorite signal path current feedback opam the AD811/AD9610. Funny that the part number for my favorite power amp driver part is now the LME49811. Still have 500 AD811's that I never used!
Mark / audioman54
Hi Tolu and Andrea
Hi Tolu and Andrea,
Pricey yes but to get that last little bit of extra sound quality is sometimes worth 10X in price to get a very small gain. I only use the metal cans in all my designs and in my power supplies! LME49710HA's and LME49713HA's are my favorite parts.
...and Andrea I am not at liberty to tell you about the design of the LME49723MA (even though I don't work at National anymore I would like to return someday however unlikely that may be) but suffice to say it is not in the same class as the 710's or the 713's. Just listening is not enough!!!!! You need to measure first then listen. the 710's are 10X better in THD than the 723's 2X in noise! Those are critical measurements for audio and really do matter. You are fooling yourself if you think the 723 sounds better than the 710 but I will say the 723 is an ok cheaper part. \
Also Andrea, as I have said many times before on this forum, the LME4720 and the LM4562 "Are The Exact Same Parts" with different numbers on the plastic package!!! The LM4562 was produced second, after the discovery of the great sound qualities of the LM4702 power amp preamp part, on the new VIP National audio parts process. Then the number sequence was changed to the LME49XXX 5 digit series and we renamed the 4562 the LME49720 to match the 49710 and 49740. However, that said, different runs of parts can have process variations that sometimes can have a very, very minor effect on signal/sound quality and so parts with the same numbers c-a-n sound slightly different. It took me a while to learn about how these process variations affected sound quality until I worked at National for many years and others there will dispute this but there were certain process tests (one in particular) that could be used to pick out better parts in a lot! However without matching levels in listening tests to 1/2 a dB and doing your best to do at least single blind (if not double blind which is really hard to do right!) just listening is not enough. Believe me, after 35 years at this, I know quite a bit about this "Stuff"!!! So please take my word for it and stay with the LME49710/20 over the 723... unless you need to save a few cents on a 1,000,000 piece order!
Best "Out of Work" Audio Engineer Regards,
Mark / audioman54
Hi Tolu and Andrea,
Pricey yes but to get that last little bit of extra sound quality is sometimes worth 10X in price to get a very small gain. I only use the metal cans in all my designs and in my power supplies! LME49710HA's and LME49713HA's are my favorite parts.
...and Andrea I am not at liberty to tell you about the design of the LME49723MA (even though I don't work at National anymore I would like to return someday however unlikely that may be) but suffice to say it is not in the same class as the 710's or the 713's. Just listening is not enough!!!!! You need to measure first then listen. the 710's are 10X better in THD than the 723's 2X in noise! Those are critical measurements for audio and really do matter. You are fooling yourself if you think the 723 sounds better than the 710 but I will say the 723 is an ok cheaper part. \
Also Andrea, as I have said many times before on this forum, the LME4720 and the LM4562 "Are The Exact Same Parts" with different numbers on the plastic package!!! The LM4562 was produced second, after the discovery of the great sound qualities of the LM4702 power amp preamp part, on the new VIP National audio parts process. Then the number sequence was changed to the LME49XXX 5 digit series and we renamed the 4562 the LME49720 to match the 49710 and 49740. However, that said, different runs of parts can have process variations that sometimes can have a very, very minor effect on signal/sound quality and so parts with the same numbers c-a-n sound slightly different. It took me a while to learn about how these process variations affected sound quality until I worked at National for many years and others there will dispute this but there were certain process tests (one in particular) that could be used to pick out better parts in a lot! However without matching levels in listening tests to 1/2 a dB and doing your best to do at least single blind (if not double blind which is really hard to do right!) just listening is not enough. Believe me, after 35 years at this, I know quite a bit about this "Stuff"!!! So please take my word for it and stay with the LME49710/20 over the 723... unless you need to save a few cents on a 1,000,000 piece order!
Best "Out of Work" Audio Engineer Regards,
Mark / audioman54
Thanks, your post was quite interesting to read!
Yes... the LME49723 is noisier... I did realize it. It's just that I wasn't entirely happy with the LME49710NA or 49720NA, so I had to get to try something else in the same family...
I don't know, maybe being slower and higher distortion the LME49723 got the balance & the tonal color right more than the LM4562NA, for instance. I guess I should try the LME49725 (which I have) to see if it has that last bit of body & sonic color that the LME49720NA & LM4562NA lacked for my ears & taste.
Anyway, to the LME49710NA I prefer the LT1028ACN8 (my favorite plastic package opamp). It's a harder choice between LT1028ACN8 & the "metal can" LME's, which are indeed excellent parts.
Yes... the LME49723 is noisier... I did realize it. It's just that I wasn't entirely happy with the LME49710NA or 49720NA, so I had to get to try something else in the same family...
I don't know, maybe being slower and higher distortion the LME49723 got the balance & the tonal color right more than the LM4562NA, for instance. I guess I should try the LME49725 (which I have) to see if it has that last bit of body & sonic color that the LME49720NA & LM4562NA lacked for my ears & taste.
Anyway, to the LME49710NA I prefer the LT1028ACN8 (my favorite plastic package opamp). It's a harder choice between LT1028ACN8 & the "metal can" LME's, which are indeed excellent parts.
Last edited:
BTW I don't think there's such a small difference in sonic character between the DIP/SOIC and the TO99 LME's. There really is a significant difference, for my ears, although the basic sonic signature remains unchanged - and in terms of basic sonic signature, I tend to prefer that of an LT1028 as it feels more musical/pleasant in the long term, without giving up the degree of transparency that I desire.
The preamp to which I refer used LF356 in cans. Measured THD% was pretty good for one of its vintage. Putting in the LME49710's really improved the sound, even though the THD% was higher (and I measured it several times). It turns out that in all measurements, the LME49710 is much, much quieter. The CCIR2K measured noise for the LME49710 is 1/3 that of the LF356. (These are surface mount devices on Aries carriers).
This is pretty much a stock Crown IC-150 preamp, no fiddling with the power supply and associated switching and control circuitry. I hope to document the POOGE to demonstrate the simple things you can do to reduce noise. The other Crown IC-150A is extremely quiet with the LME49710's and the Jung-Didden super regulator with LT1028's. Of course, in the latter the 25u non-polar electrolytic has been removed and the controls "de-oxitized".
This is pretty much a stock Crown IC-150 preamp, no fiddling with the power supply and associated switching and control circuitry. I hope to document the POOGE to demonstrate the simple things you can do to reduce noise. The other Crown IC-150A is extremely quiet with the LME49710's and the Jung-Didden super regulator with LT1028's. Of course, in the latter the 25u non-polar electrolytic has been removed and the controls "de-oxitized".
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.