Around a month ago I have had very bad experience with these chips:
https://www.ti.com/store/ti/en/p/pr...arch=Search-EN-everything&usecase=part-number
Only one part from ten was usable regarding the noise.
One would think if TI was responsible enough and acknowledge the issues, they would screen the parts so the customers would not have to be burden with the cost of screening themselves, shame on TI.
I searched TI's web site regarding this issue, there are two threads both are locked,
[Resolved] LME49720: popcorn noise - Audio forum - Audio - TI E2E support forums
LME49720: Pop-cone noise - Audio forum - Audio - TI E2E support forums
I take it as they know it is a problem and chose to do nothing about it or there answer is thatthe customer is responsible to test/grade and they will offer replacements as their solution.
I searched TI's web site regarding this issue, there are two threads both are locked,
[Resolved] LME49720: popcorn noise - Audio forum - Audio - TI E2E support forums
LME49720: Pop-cone noise - Audio forum - Audio - TI E2E support forums
I take it as they know it is a problem and chose to do nothing about it or there answer is thatthe customer is responsible to test/grade and they will offer replacements as their solution.
Last edited:
I would have thought the 4562 issue were solved and settled silenty, by now, 10 years after... but apparantly it's still there, you say, what, a 1 out of 10 yield rate? OTOH, there's tons of LM4562-based products out there, so for many circuits the occasional unit that misses specs (or, rather, passes them) doesn't seem to be a major problem....
Anyway, full agree on the OPA2156/1656, a giant killer opamp, especially at that price point. And plenty of other excellent low noise & low distortion opamps available today.
Anyway, full agree on the OPA2156/1656, a giant killer opamp, especially at that price point. And plenty of other excellent low noise & low distortion opamps available today.
Bingo, do you replaced the used TL072 also with this new op-amp and measured the benefit of the lower noise?
Hp
A bit on dBc and dBr
From the legacy part of my day job, dBc is more of an RF term. I'm not sure how it found it way into audio. dBr is a convenience term used to take dBc measurements. But dBr does have a firm place in any measurement it can be applied to, in audio or anything else for that matter. Like I said it's more for aconvenience.
I'll use the legacy NTSC TV channel standard as an example because it's analog.
The standard specifies the sound carrier in an NTSC channel shall be -15dBrBc below the picture carrier where the picture carrier level is reference or simply -15dBc. dBc state which is the reference level. From any measurement of the NTSC channel signal strength the picture or video level can be anything depending on where the level is measured. To be sure the sound or audio carrier is -15dB below the video carrier we may have to do a bit of math. The dBr allows us to add an offset to the video carrier reference level to bring it to a more convenient, say 0dB level. This way we can get our measurement at a glance -15dB, rather than having to calculate it out.
dBr can be used to add or subtract to bring it to +/- 10dB to make the math easier, It's doesn't have to be a reference of 0dB. it's just a means of making measurement life easier. Of itself it has no meaning. We express it, in RF, as dBc. But in audio I think simply dBr is more appropriate.
From the legacy part of my day job, dBc is more of an RF term. I'm not sure how it found it way into audio. dBr is a convenience term used to take dBc measurements. But dBr does have a firm place in any measurement it can be applied to, in audio or anything else for that matter. Like I said it's more for aconvenience.
I'll use the legacy NTSC TV channel standard as an example because it's analog.
The standard specifies the sound carrier in an NTSC channel shall be -15dBrBc below the picture carrier where the picture carrier level is reference or simply -15dBc. dBc state which is the reference level. From any measurement of the NTSC channel signal strength the picture or video level can be anything depending on where the level is measured. To be sure the sound or audio carrier is -15dB below the video carrier we may have to do a bit of math. The dBr allows us to add an offset to the video carrier reference level to bring it to a more convenient, say 0dB level. This way we can get our measurement at a glance -15dB, rather than having to calculate it out.
dBr can be used to add or subtract to bring it to +/- 10dB to make the math easier, It's doesn't have to be a reference of 0dB. it's just a means of making measurement life easier. Of itself it has no meaning. We express it, in RF, as dBc. But in audio I think simply dBr is more appropriate.
I like dBc because it's already used for distortion measurements. Yes, RF amplifier distortion, but what's the big difference? Most of the audio world seemed stuck in distortion percentages for too long anyway, and that gets really confusing once you get below 1 ppm distortion, so I suspect that this is why dBc isn't used as much. Further, the world of percentages seemed to be a holdover from THD+N measurements, whereas dBc has almost always been used for a direct measurement of one spurious tone, not the aggregate level of an analyzer residual, which is more of an indication of noise floor than linearity these days.
I've never seen dBr used for spurious tone levels anywhere but here, so do we really need to dig a moat around ourselves and make a parallel universe?
I've never seen dBr used for spurious tone levels anywhere but here, so do we really need to dig a moat around ourselves and make a parallel universe?
As long as everyone is on the same page with it. I just wanted to clear up some definitions. But by all means use it.
I hate to suggest this as there's too many dBthis and dBthats floating around, but if it's the fundamental, call it dBf. Or would that mean a harmonic that's so many dB's down FROM the fundamental level?
I recall dBV, dBm, dBu and dBFS as used in audio and RF from decades ago (and probably well defined in a RANE note), but the dBc and dBr are new to me. I may have to use Google, or even Wikipedia ...
David, an AP distortion analyzer has the option to have the Y scale in dBr.
There is an entry on the analyzer panel where you can set the ref level. Then the reported distortion is scaled referred to that setting.
So let us say that your signal (carrier, fundamental) is 1.8V, you set the dBr ref to 1.8V and you get dBr referred to 1.8V.
It sound much more complicated than it is to use it.
Jan
Attachments
Last edited:
Jan when we scale a signal to 0dBV to make a measurement easier to read we have used a form dBr to get it there. What's complicated about this? If we're going through a number of gain changes in a chain of measurement such as we might to get to -150dBV on more that one instrument, say sound card and analyzer readout, it makes sense to do this.
Supposing you are calibrating a batch of your autoranger and the value you want has a lot of decimal points. Whouldn't it make sense to use the dBr function in your AP to scale the set value to an even number. This would make it easier to read at a glance. This is what dBr is used for.
My point is that dBr is scaling function usually only found on instruments. If it doesn't work for your particular measurement than it doesn't make sense to use it.
Supposing you are calibrating a batch of your autoranger and the value you want has a lot of decimal points. Whouldn't it make sense to use the dBr function in your AP to scale the set value to an even number. This would make it easier to read at a glance. This is what dBr is used for.
My point is that dBr is scaling function usually only found on instruments. If it doesn't work for your particular measurement than it doesn't make sense to use it.
Last edited:
What are the datecodes? I'm not doubting your experience, but I want to nail this down. In my experience, the recent datecode (914V) LM4562 that I have used have been really nice compared to the best of what I'd like to think of as "old stock" from the 6" line.
The only theory I have is that the old 6" line was used to make a range of parts like gold doped logic or discretes, and that process contaminated the furnaces, making subsequent linear process wafers 'contaminated'. Note - this is completely wild theory, based on less than nothing, except that if a furnace runs a process, that process can contaminate subsequent wafers made with a different process.
Fully agree. If you have a toolbox, use the best tool for the job at hand. Tools have no intrinsic value, the value is in the way they help you get the job done
Jan
Well, mai you also consider to lower the LDO resistors to 1K ohms direction? The values should be large enough so noTHD is product to drive the R/C filter.
Also both parts of the filter uses none equal R/C values.. Why?
Hp