The NE5532 is quite nice sounding as long as the output swings are kept low (say below 0.5V peak). There are applications where it is superior to much better spec'ed op-amps - for instance, in a nested chipamp topology using an LM1875 as the power stage, and the NE5532 for an outer feedback loop. The relatively slower and lower-gain NE5532 is much easier to compensate for in this kind of topology.
Apologies in advance for waking this old dead thread up. 
Anyone try and succeed at a simple LME49713 pre, either inverted or non-inverted configuration?
The one I tried (attached) is far from working without a lot of noise. I have also tried the non-inverting config by swapping pins 2-3 input and it is even worse with about -6 volts at the outputs.
I'd hate to waste these parts, if you have a working schema, by all means...kind of been in the "try this try that" road and need a proven working circuit.
Anyone try and succeed at a simple LME49713 pre, either inverted or non-inverted configuration?
The one I tried (attached) is far from working without a lot of noise. I have also tried the non-inverting config by swapping pins 2-3 input and it is even worse with about -6 volts at the outputs.
I'd hate to waste these parts, if you have a working schema, by all means...kind of been in the "try this try that" road and need a proven working circuit.
Attachments
Last edited:
Now to schematic.
The inverting topology has a gain of 1k2/100 = 12times (+21.6dB)
That is a lot for a pre-amp. Ends up adding noise when the excess gain is not required.
Have you read
http://www.diyaudio.com/forums/diyaudio-com-articles/186018-what-gain-structure.html
Zin, the impedance seen by the Source is 100ohms. That is bad for the sources.
The bandwidth is unlimited at RF except by the limitations of the LME. That is bad for interference.
The LF bandwidth is set by the 2u2F and the 100ohms. It has a roll off F-3dB @ ~700Hz. That is bad for mid and bass reproduction.
did you download these two technical documants?
http://www.ti.com/product/LME49713/technicaldocuments#doctype10
Do you know that CFA opamps require different feedback from VFA opamps?
The inverting topology has a gain of 1k2/100 = 12times (+21.6dB)
That is a lot for a pre-amp. Ends up adding noise when the excess gain is not required.
Have you read
http://www.diyaudio.com/forums/diyaudio-com-articles/186018-what-gain-structure.html
Zin, the impedance seen by the Source is 100ohms. That is bad for the sources.
The bandwidth is unlimited at RF except by the limitations of the LME. That is bad for interference.
The LF bandwidth is set by the 2u2F and the 100ohms. It has a roll off F-3dB @ ~700Hz. That is bad for mid and bass reproduction.
did you download these two technical documants?
http://www.ti.com/product/LME49713/technicaldocuments#doctype10
Do you know that CFA opamps require different feedback from VFA opamps?
Last edited:
The current feedback opamps are better suited to driving low impedance loads and fed from low impedance sources. A voltage feedback opamp driving a current feedback opamp would be more successful in this application. I would use an LM49880 for the input and the LM47613 for the output with feedback from the output. Set the '613 for unity gain non-inverting and the '880 for about 6 dB gain using low value resistors, e.g.: 475 Ohms feedback and 475 Ohms to ground. The bandwidth of the '610 won't add phase to destabilize the '880 in this configuration. You should be able to drive something like 50 Ohms with unmeasurable distortion to possibly 1 MHz at 3V out. Input Z can be as high as 1 Meg so it won't load the pot and add distortion there.
You might get some clues from this,
http://www.diyaudio.com/forums/anal...rent-feedback-opamps-simulation-practice.html
Thanks Mooly, that was some needle in the haystack information and helped me solve this problem. You are right, not much practical information on CFB op amps out there. Lots of theory though.
I matched the closest resistors I had to your schematic and got rid of the input caps that were too near the inputs and breaking one of the golden rules. Not sure even the output caps are necessary. The schematic attached show a real life working LME49713 pre. The circuit appears to work well in real life, sounded great and crystal clear--too tired to scope now.
Only mishap is one of the LME49713 blew up in trials or was deffective, it hums and is cool to the touch and not warm like the other. Swap channels and problem follows it. $12 and change each!
Only change I will make to the below schematic is to go with a 150 ohm input resistor for 8X gain and the vol pot will be 100K ohms.
Thanks folks for the help!
Attachments
No, much appreciated. This is by no means the final pre, but the scratchpad as the piece of wood foundation witnesses. Now that I have the circuit somewhat figured out, (after Mooly shared his experience), I can proceed to fine tune. Apparently, he is one of the few that has actually tried it other than in Spice model.
All prior comments were absorved; but this is totally different animal and I am stuck on making an inverting configuration work, perhaps stubbornly.
One of the guiding principles I have is to keep Rf at around 1.2K (I think mentioned earlier), hence if I want gain, I must go low on Ri.
Thanks again for the input.
Last edited:
Whaleman,
Do not use a non-inverting stage for a CFB opamp config for audio. Remember they are not a typical diff pair input stage opamps internally. Also "please get rid of all the DC coupling caps"as they wreck the sound quality!!! 1.5k to 1.2k for the feedback R is the mandatory range and ultra high quality power supply regulators are also required. I have been running my D/A preamp with LME49713HA's for...what 7-8 years now at +/- 18VDC.
Thanks for using my CFB opamp! (I specified that part but did not design the internal circuits. I had been using the AD811's in all my audio designs for years but wanted a CFB opamp designed with audio specs in mind not video specs like the AD811 was designed for. It is a great story how we got those parts made before TI took over National and eliminated the audio group in Santa Clara. Still job hunting by the way. Both LME49713 and AD811 are great CFB opamps but the LME part is better and also the AD811 does not come in a metal can and double blind tests proved the metal cans sound better. Bob Pease even picked cans over dips in a double blind test in the my sound room at National. However the first ADI CFB opamp, the AD9610 that the AD811 came from, only came in a can and had an internal feedback R. Super expensive though, $300 each, and were MIL spec parts. That is also a great sounding CFB opamp and I still have a bunch of those.)
Enjoy your LME49713HA preamp! Next try the LME49811 in a power amplifier!
"audioman54" / Mark
Do not use a non-inverting stage for a CFB opamp config for audio. Remember they are not a typical diff pair input stage opamps internally. Also "please get rid of all the DC coupling caps"as they wreck the sound quality!!! 1.5k to 1.2k for the feedback R is the mandatory range and ultra high quality power supply regulators are also required. I have been running my D/A preamp with LME49713HA's for...what 7-8 years now at +/- 18VDC.
Thanks for using my CFB opamp! (I specified that part but did not design the internal circuits. I had been using the AD811's in all my audio designs for years but wanted a CFB opamp designed with audio specs in mind not video specs like the AD811 was designed for. It is a great story how we got those parts made before TI took over National and eliminated the audio group in Santa Clara. Still job hunting by the way. Both LME49713 and AD811 are great CFB opamps but the LME part is better and also the AD811 does not come in a metal can and double blind tests proved the metal cans sound better. Bob Pease even picked cans over dips in a double blind test in the my sound room at National. However the first ADI CFB opamp, the AD9610 that the AD811 came from, only came in a can and had an internal feedback R. Super expensive though, $300 each, and were MIL spec parts. That is also a great sounding CFB opamp and I still have a bunch of those.)
Enjoy your LME49713HA preamp! Next try the LME49811 in a power amplifier!
"audioman54" / Mark
Hi Mark,
Thanks for chiming in! Appreciate your "feedback" and your diligent work at National with RAP. You are responsible for my small journey into this LME49713 preamp trial!
I was convinced that the LME49720 was the holy grail after the awards years back, but the 13 has bumped it off.With some limited time and the only working channel (since I blew up one of the LME49713HA--yes, I am a hack and nowhere near an engineer obviously), I compared it to my other LME49720HA pre, and I must say it easily edged it out. So I am working towards building another chassis like the one that wraps the LME49720HA pre (attached), or just replace it altogether. I do plan to use regulated power supply for it and was aiming for 15VDC, but I think you said they will handle 18 VDC comfortably.
Sorry to hear you are still looking for a position, a guy like you would be snapped up right away in a different economy. Hang in there and hope you kept the Mustang. These LME products speak for themselves.
Believe it or not, the LME49811 power amp is next, I have been looking at a kit here and there. It will match my Altec 811 horn speakers well.
Thanks.
Attachments
Whaleman,
One of the other posters mentioned correctly that driving the LME49713 in inverting mode directly with a pot will be a problem. I would add the variable gain stage after the LME49713 if possible which is what I do in my preamp. I end up with three LME opamps in the entire signal path that then drives the LME49811 power amps.
Download the ap-note, AN-1490, that I wrote for the LM4702 power amp. (it was too big to attach to this file and when TI took over they removed all our names from our AP Notes!) The circuit for the mono amp is just 1/2 of the LM4702 circuit. You can also purchase boards as I know that there are several companies that make power amplifier boards using the LME49811. I assisted a couple of them with their designs before Texas Instruments got rid of us all.
15 to 20VDC is fine but I would suggest a snap on heatsink on the metal can parts.
As for your non-audio related comment...Yes I still have my 2007 Shelby GT500 SVT#12 which is one nice thing I got from working at National for 12 years. It now has 137K miles on it but since I have not been working the last year an a half I have only added about 1K miles! I miss not driving it 100 miles every day...but it will last longer now. I am told the first 20 to 50 pre-production SVT vehicles are all built with hand selected parts and the production is done in a much more "personally supervised manor" than the normal production line. Since I could never replace it now I intend to keep it for as long as I can and then pass it on to one of sons. My wife does not want it as she does not like driving a stick shift but she has driven it a few times.
Sorry to bother everyone with this car talk but it is interesting that whenever I visited high end companies, for National, without exception the owners of the companies had a high performance car of some kind. Wilson had 2 Ferrari's out front when I visited them 5 years or so ago. Of course Noel Lee of Monster is famous for his large high performance car collection that he posts on the Monster Web Page. There are many others I saw during my company travels that are too numerous to mention at this time. Fast cars and high end audio...toys for boys? Count me in!...even in forced retirement!
Best LME Audio Regards Everyone,
audioman54 / Mark
One of the other posters mentioned correctly that driving the LME49713 in inverting mode directly with a pot will be a problem. I would add the variable gain stage after the LME49713 if possible which is what I do in my preamp. I end up with three LME opamps in the entire signal path that then drives the LME49811 power amps.
Download the ap-note, AN-1490, that I wrote for the LM4702 power amp. (it was too big to attach to this file and when TI took over they removed all our names from our AP Notes!) The circuit for the mono amp is just 1/2 of the LM4702 circuit. You can also purchase boards as I know that there are several companies that make power amplifier boards using the LME49811. I assisted a couple of them with their designs before Texas Instruments got rid of us all.
15 to 20VDC is fine but I would suggest a snap on heatsink on the metal can parts.
As for your non-audio related comment...Yes I still have my 2007 Shelby GT500 SVT#12 which is one nice thing I got from working at National for 12 years. It now has 137K miles on it but since I have not been working the last year an a half I have only added about 1K miles! I miss not driving it 100 miles every day...but it will last longer now. I am told the first 20 to 50 pre-production SVT vehicles are all built with hand selected parts and the production is done in a much more "personally supervised manor" than the normal production line. Since I could never replace it now I intend to keep it for as long as I can and then pass it on to one of sons. My wife does not want it as she does not like driving a stick shift but she has driven it a few times.
Sorry to bother everyone with this car talk but it is interesting that whenever I visited high end companies, for National, without exception the owners of the companies had a high performance car of some kind. Wilson had 2 Ferrari's out front when I visited them 5 years or so ago. Of course Noel Lee of Monster is famous for his large high performance car collection that he posts on the Monster Web Page. There are many others I saw during my company travels that are too numerous to mention at this time. Fast cars and high end audio...toys for boys? Count me in!...even in forced retirement!
Best LME Audio Regards Everyone,
audioman54 / Mark
thanks for the suggestion Mirlo, I think I will add an op amp buffer stage there and see how that works. I have some extra BB op amps laying around.
The LME49713 op amps arrived from Mouser yesterday. I quickly replaced the damaged ones and tried the circuit attached. If the volume control was say over 8 o'clock, things seem to work just fine. Turn the vol pot all the way down and pink noise appeared that seem to be getting louder and louder. Also the vol pot was very scratchy and popped with movement with no signal. The op amps were getting warm to the touch as well. I did add the buffer as suggested by Mirlo. I then replaced the 13's with some LME49710 instead and the circuit works wonderfully, no noise of any kind, op amps not even warm to the touch, and may just leave it there--unless someone has the clue. I thought about placing the vol pot at the end of the circuit and maybe try that, but blowing these op amps again is another $30 proposition. Anyhow, that is the update...
Attachments
Some ideas to think about.
Reduce the bandwidth very slightly by increasing the 1k2 feedback to 1k3, or maybe 1k5
Add a 360r to the +IN to SG of the 713
What does the 24r do for the 713 output?
What is the output offset of the 604?
Reduce the bandwidth very slightly by increasing the 1k2 feedback to 1k3, or maybe 1k5
Does the 100r on the -IN add to the 1k2 to give an effective 1k3?
Add a 360r to the +IN to SG of the 713
What does the 24r do for the 713 output?
What is the output offset of the 604?
Last edited:
In 1983 this guy IPO's his tech company and buys a Ferrari and immediately takes it off a cliff and kills himself and his broker.
(the early 1980's was the era in which Apple and Genentech went public, United Tech bought MOStek etc.)
I think that the original datasheet for the LME49713 had some errors. TI gets a bit huffy over this -- there is still an error in the LM317 datasheet.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.