Nice, this thing is worth a try. LME49810 driving mosfets was really hard to get working. I have done a "mini-amp" based on LM4562 + IRFP since i failed too much times with the LME.
rikkitikkitavi said:
I think that the LME49811 is a new version of the LM4702 for a single output stage.
This is possible, the specs you posted for the LME49810 where anything but impressive. You said it was without output stage, but at AN1645 THD was better with a LM4702 direct driving a mosfet output stage than yours with LME49810 standalone.
The specs posted on National website for LME49811 is quite similar to that of LM4702. so is it the single channel version of LM4702 assembled in the LME49810 package?
Trust the Llama. . . if you want to use FETs, I'd suggest waiting for the FET part. Bench performance with the LME49810 can achieve/exceed the same levels as the LM4702 and the LME49811 is indeed a LM4702 based mono part with a little more drive current.
If you want to 'play' with the LME49810 driving FETs then add a resistor from the gate of the NFET directly to the gate of the PFET. Try 10K - 20K and play with the values to see what works best for your circuit.
-SL
If you want to 'play' with the LME49810 driving FETs then add a resistor from the gate of the NFET directly to the gate of the PFET. Try 10K - 20K and play with the values to see what works best for your circuit.
-SL
Hi,
I can't see what voltage you plan to run.
3pair output stage looks OK for 4ohm @ +-55Vdc.
You have used output devices as drivers and used drivers as pre-drivers. This seems like overkill. An 8pair output stage might need that level of current drive capability.
The Sziklai pairs have their resistor missing.
What are the values of the pre-driver emitter resistors?
Show a leading zero on values below 1r0, to avoid confusion.
you have chosen a very low gain due to selecting 2k6? for R2.
and used mixed DC and AC blocking (at input and NFB) which will screw up the output offset.
Fit an RF attenuating cap at the input, or even on the input RCA terminal.
What does National say about a Thiele or Zobel on the output?
I can't see what voltage you plan to run.
3pair output stage looks OK for 4ohm @ +-55Vdc.
You have used output devices as drivers and used drivers as pre-drivers. This seems like overkill. An 8pair output stage might need that level of current drive capability.
The Sziklai pairs have their resistor missing.
What are the values of the pre-driver emitter resistors?
Show a leading zero on values below 1r0, to avoid confusion.
you have chosen a very low gain due to selecting 2k6? for R2.
and used mixed DC and AC blocking (at input and NFB) which will screw up the output offset.
Fit an RF attenuating cap at the input, or even on the input RCA terminal.
What does National say about a Thiele or Zobel on the output?
Thanks for the input 🙂
DC resistance is 2.6ohm, is this what you mean? Or the impedance v. freq curve?
+-70vAndrewT said:
Yes I may need another pair? I am planning to use this for sub out, which to match the rest of the system spl I will not need to push it at max.
The resistor value is to pair a reasonable sized capacitor. The Fb Res was an oversight - thanks!
NS recommends not to do this as it will degrade the signal.
I do not think I need this for a sub application?
BWRX said:
DC resistance is 2.6ohm, is this what you mean? Or the impedance v. freq curve?
I have just run 4pair through Bensen's spreadsheet.
4ohm and 45degree phase angle takes the devices to 100mS SOAR limit at Tc=65degC when on +-70Vdc.
The model is predicting 460W into 4r0 and 27.9Apk into 2r0.
These devices are struggling with supply rails at that high a voltage.
4ohm and 45degree phase angle takes the devices to 100mS SOAR limit at Tc=65degC when on +-70Vdc.
The model is predicting 460W into 4r0 and 27.9Apk into 2r0.
These devices are struggling with supply rails at that high a voltage.
InlineTwin said:
Doesn't matter the speaker -- these high gain opamps are prone to oscillation -- I have had them oscillate in the low MegaHertz region when I didn't include the little LPF.
You only have gain of 10.7dB -- is this purposeful? Usually gain of 28 to 30dB is shown.
Increase the value of C1 to at least 22uF.
The cap you have in series with the feedback resistor is most likely unnecessary --
AndrewT said:
I found these which have better SOA @150C: MJL3281A/MJL1302A.
Would this be enough or is more pairs is necessary?
Thanks for your input!
Scheme of my device:
http://ivanaudiomaster.narod.ru/TORRESlme49810.gif
Current protect:
http://ivanaudiomaster.narod.ru/shema.gif
My current protect:
http://ivanaudiomaster.narod.ru/myPROTECTlme49810.GIF
http://ivanaudiomaster.narod.ru/TORRESlme49810.gif
Current protect:
http://ivanaudiomaster.narod.ru/shema.gif
My current protect:
http://ivanaudiomaster.narod.ru/myPROTECTlme49810.GIF
IVANSOUND said:
Hi Ivansound
I changed your links to URL rather than embedded images as they weren't showing. If I were to hazzard a guess as to why that is I'd say the server your linking to doesn't allow redirects of images.
At least this way folks can click the link to see the images.
Parameters of LME49830 chip doesn't look better than LME49810 - same output power, lower PSRR and slew rate. They are pin compatible so I will probably try it in near future without changing layout of my board 😉
An externally hosted image should be here but it was not working when we last tested it.
If you like the LM4702 then try out the LME49811. This is a mono version of the LM4702 with more drive current. Since the LM4702 can be less than ideal to layout a PCB the LME49811 should help in this area. It is also pin for pin compatible with the LME49810 and LME49830. One board can be made for all three parts and then with some forethought and planning, the same board can be used for Darlingtons, BJTs, or FETs. Nice way to go, can do sound comparisons between parts and type of output stage device.
-SL
-SL
fwiw, the datasheet on the LME49830 was just issued last week:SpittinLLama said:
http://www.national.com/ds/LM/LME49830.pdf
Guys,
What is the major difference between this three chips
1) LME49810
2) LME49811
3) LME49830
1 and 3 having similar specification but 3 is made specifically to drive MOSFET output. Question : What is the feature of 3 that made it a mosfet driver? Is 1 not suitable for MOSFET?
The price of 1 and 2 are the same. 1 seems to be much better than 2 interms of current drive etc. Why would anyone use 2 instead of 1? Maybe I miss something important here.
Thanks.
What is the major difference between this three chips
1) LME49810
2) LME49811
3) LME49830
1 and 3 having similar specification but 3 is made specifically to drive MOSFET output. Question : What is the feature of 3 that made it a mosfet driver? Is 1 not suitable for MOSFET?
The price of 1 and 2 are the same. 1 seems to be much better than 2 interms of current drive etc. Why would anyone use 2 instead of 1? Maybe I miss something important here.
Thanks.
LME49810 has an internal Baker Clamp and a clip flag indication pin. The LME49811 is a mono version of the LM4702 with a little more drive current. LME49830 is designed for FETs in that it has the highest bias voltage (16V) possible. I believe the LME49810 has 11V bias possible. The LM4702 only has about 7V according to AN-1645. I am not sure how much bias the LME49811 has but maybe it is the same as the LM4702. So you have to look at if you want the extra features of the 810 and then what sort of bias voltage is needed for your output stage design. There are also those who believe each part has a different sound. Those that like the LM4702 sound are probably most interested in the LME49811. Since the pin out is such that you can make one board you can try them out and see what sounds the best to you.
-SL
-SL