EMRFD Message Archive 956
Message Date From Subject 956 2007-08-16 10:09:23 Sam Morgan LTSpice and ELR audio amp circuit help needed.
Sad part is I have no clue what I'm doing.
The circuit is the audio amp used in the electroluminescent receiver.
Circuit can be found here:
http://www.pan-tex.net/usr/r/receivers/elraudioamps.htm
I am focusing on area from the input at C95 to the output of C98.
I wanted to get a visual of the circuit with the 2.2Uf cap at C99
and then play with the value of C99 and see the results.
I spoke with WN5Y and he said to replaced the 2.2Uf with a 10uf and it did
increase the gain, I did, but it was to large an signal increase.
Now my speaker distorts when volume is over 10-15%.
So I thought I might could learn Spice using this simple 1 transistor circuit,
and be able to find a value I needed for C99.
Here you can find my attempt on doing this:
http://www.pan-tex.net/usr/r/receivers/elraudioamps.htm
I just guessed that the output resistor would be mid range
between the 70k and 150k input listed in the TDA2003 specs.
http://home.eunet.cz/rysanek/pdf/tda2003.pdf
Any and all help would be appreciated.
Maybe even a meeting on echolink later this evening?
--
GB & 73's
KA5OAI
Sam Morgan958 2007-08-16 10:30:26 Michael Neverdosk... Re: LTSpice and ELR audio amp circuit help needed.
http://tech.groups.yahoo.com/group/LTspice/?yguid=226809086
A very good resource for help with the simulator.
michael N6CHVOn 8/16/07, Sam Morgan <ka5oai@jass-ltd.org> wrote:I am trying to analyze a circuit in LTSpice.
Sad part is I have no clue what I'm doing.
The circuit is the audio amp used in the electroluminescent receiver.
Circuit can be found here:
http://www.pan-tex.net/usr/r/receivers/elraudioamps.htm
I am focusing on area from the input at C95 to the output of C98.
I wanted to get a visual of the circuit with the 2.2Uf cap at C99
and then play with the value of C99 and see the results.
I spoke with WN5Y and he said to replaced the 2.2Uf with a 10uf and it did
increase the gain, I did, but it was to large an signal increase.
Now my speaker distorts when volume is over 10-15%.
So I thought I might could learn Spice using this simple 1 transistor circuit,
and be able to find a value I needed for C99.
Here you can find my attempt on doing this:
http://www.pan-tex.net/usr/r/receivers/elraudioamps.htm
I just guessed that the output resistor would be mid range
between the 70k and 150k input listed in the TDA2003 specs.
http://home.eunet.cz/rysanek/pdf/tda2003.pdf
Any and all help would be appreciated.
Maybe even a meeting on echolink later this evening?
--
GB & 73's
KA5OAI
Sam Morgan
959 2007-08-16 11:26:33 Leon Re: LTSpice and ELR audio amp circuit help needed.
960 2007-08-16 11:32:37 Sam Morgan Re: LTSpice and ELR audio amp circuit help needed.
> Are you aware of the LTSpice group on Yahoo?no I wasn't but I'm now a member, thanks for the pointer.
> http://tech.groups.yahoo.com/group/LTspice/?yguid=226809086
>
> A very good resource for help with the simulator.
>
I see I have lots of reading and head scratching ahead.
--
GB & 73's
KA5OAI
Sam Morgan961 2007-08-16 11:37:33 Sam Morgan Re: LTSpice and ELR audio amp circuit help needed.
> I just simulated it with Pulsonix SPICE, I think I've got all the valuesyou might should refer to the original diagram from the elr pages:
> correct. The output is very distorted, unless I reduce C99 to 100n, which
> only gives a gain of about 2x. The frequency response is wrong, as well. I
> haven't checked the design, but there is obviously something seriously
> wrong
> with it.
>
http://www.pan-tex.net/usr/r/receivers/elraudioamps.htm
I'm sure I may have messed up my attempt to put it in LTSpice.
--
GB & 73's
KA5OAI
Sam Morgan963 2007-08-16 13:37:54 Leon Re: LTSpice and ELR audio amp circuit help needed.
964 2007-08-17 08:21:25 Alan Yates Re: LTSpice and ELR audio amp circuit help needed.
Even with SPICE simulation I'd suggest you put a small value resistor in
series with C99 and Q19's emitter. This will let you control the gain
which much more precision. Increase C99 until it is 'large' enough to
shunt the lowest frequency of interest, then set the gain with the
series resistance.
The gain of Q19 is approximately its collector resistance over its
emitter resistance. The fact that its load varies with the volume
control V3's position is a poor design feature IMO, I would have
connected the pot around the other way so the collector sees a more
constant load. I didn't check the datasheet, but I assume the TDA2002's
-ve input is a relatively high impedance at AF in this configuration?
The 100n from the collector to ground is a hack. I suspect it is really
there because of poor filtering in the product detector. It would be
better and neater to put a Sallen-Key low-pass filter between the AF
pre-amp and the power amp, probably using a simple 2N3904 emitter
follower. You could build it into the preamp stage with gain and save a
stage, or heck use an opamp if you wanted, but I prefer discrete
transistors.
Regards,
Alan VK2ZAY
On Thu, 16 Aug 2007, Sam Morgan wrote:
> To: "emrfd@yahoogroups.com" <emrfd@yahoogroups.com>
> From: Sam Morgan <ka5oai@jass-ltd.org>
> Date: Thu, 16 Aug 2007 12:08:53 -0500
> Subject: [emrfd] LTSpice and ELR audio amp circuit help needed.
>
> I am trying to analyze a circuit in LTSpice.
> Sad part is I have no clue what I'm doing.
> The circuit is the audio amp used in the electroluminescent receiver.
> Circuit can be found here:
>
> http://www.pan-tex.net/usr/r/receivers/elraudioamps.htm
>
> I am focusing on area from the input at C95 to the output of C98.
> I wanted to get a visual of the circuit with the 2.2Uf cap at C99
> and then play with the value of C99 and see the results.
>
> I spoke with WN5Y and he said to replaced the 2.2Uf with a 10uf and it did
> increase the gain, I did, but it was to large an signal increase.
> Now my speaker distorts when volume is over 10-15%.
> So I thought I might could learn Spice using this simple 1 transistor circuit,
> and be able to find a value I needed for C99.
>
> Here you can find my attempt on doing this:
> http://www.pan-tex.net/usr/r/receivers/elraudioamps.htm
>
> I just guessed that the output resistor would be mid range
> between the 70k and 150k input listed in the TDA2003 specs.
> http://home.eunet.cz/rysanek/pdf/tda2003.pdf
>
> Any and all help would be appreciated.
> Maybe even a meeting on echolink later this evening?
> --
> GB & 73's
> KA5OAI
> Sam Morgan
>
>
>
> Yahoo! Groups Links
>
>
>
>
--
Alan Yates Office: +61 2 9905 2883
System Administrator Fax: +61 2 9938 5952
AY Communications Mobile: +61 419 424 165965 2007-08-17 20:05:35 Sam Morgan Re: LTSpice and ELR audio amp circuit help needed.
> Even with SPICE simulation I'd suggest you put a small value resistor inthat all sounds good Alan,
> series with C99 and Q19's emitter. This will let you control the gain
> which much more precision. Increase C99 until it is 'large' enough to
> shunt the lowest frequency of interest, then set the gain with the
> series resistance.
>
> The gain of Q19 is approximately its collector resistance over its
> emitter resistance. The fact that its load varies with the volume
> control V3's position is a poor design feature IMO, I would have
> connected the pot around the other way so the collector sees a more
> constant load. I didn't check the datasheet, but I assume the TDA2002's
> -ve input is a relatively high impedance at AF in this configuration?
>
> The 100n from the collector to ground is a hack. I suspect it is really
> there because of poor filtering in the product detector. It would be
> better and neater to put a Sallen-Key low-pass filter between the AF
> pre-amp and the power amp, probably using a simple 2N3904 emitter
> follower. You could build it into the preamp stage with gain and save a
> stage, or heck use an opamp if you wanted, but I prefer discrete
> transistors.
>
I particular like the idea of just reworking the existing stage.
Since it is already on the circuit board for the elr.
I uploaded a redrawn schematic by Nick, WA5BDU
if you would redraw it to show what you think would be the best application of
what you said, I would appreciate it very much. I don't mean to impose, I just
don't know how to do it!
the re-draw is here:
http://pages.suddenlink.net/ka5oai/ltspice/elr-audioamp.asc
the complete circuit as is can be seen here:
http://www.pan-tex.net/usr/r/receivers/elraudioamps.htm
and you can also see the board that I need to work this on.
the data sheet for the input of the TDA2003 says it wants to see 70k-150k
http://home.eunet.cz/rysanek/pdf/tda2003.pdf
--
GB & 73's
KA5OAI
Sam Morgan966 2007-08-19 18:27:55 Alan Yates Re: LTSpice and ELR audio amp circuit help needed.
http://nexus.cable.nu/junk/elafamp.jpg
967 2007-08-21 15:31:34 jr_dakota Re: LTSpice and ELR audio amp circuit help needed.
essentially wired bass ackwards, the wiper should be on the TDA2002
side not the transistor side .... I use a similar circuit with the
LM386 when I need a voltage gain larger than 20, if you try to squeeze
more gain from the LM386 than that you increase the noise quite a bit
so I use a transistor in front of the LM386 with the volume pot
BEFORE the transistor (I also use feedback from the output to pin 8 to
roll off the highs, they call it bass boost but it's really a low pass
/high cut and you don't need 20Khz bandwidth with 2.7Khz bandwidth SSB
signals) ... Try not to drive the TDA2002 too hard because it will
suck power as it's efficiency drops quite a bit at over a 1 watt
output (8 ohm load, 2 watts with a 4 ohm load)) ... I like the
TDA2002/LM383 IC's a lot because if you drive them for an output of 1
watt or less they are as efficient as any of the 1 Watt chips but
since you have plenty of headroom you'll have considerably less
distortion and more dynamic range ... plus it's almost impossible to
blow one up (almost) they will usually survive shorted outputs
(Although your power source might not) and overvoltages
Note also that C101 on the output of the TDA2002 is the wrong value,
it should be .1 uF and it could very well be in the kit ... They call
it stablization in the data sheets but what it really is is a Zobel
network which is designed to keep the impedance of the inductive load
(speaker coil) constant with frequency (In this case it's keeping the
output impedance of the power amp constant with frequency to prevent
oscillation) ... as the frequency rises the impedance of the speaker
rises and can cause the amp to break out into high frequency
oscillation .... It's also handy in speaker crossovers to ensure the
crossover works at the desired frequency because chances are your 8
ohm nominal speaker won't be 8 ohms at the crossover point and the
calculations you use assume an 8 ohm load at the crossover point and
the Zobel network ensures this .... It's also one of my favorite
circuits, it's just a resistor and capacitor, it doesn't get much
simpler than that, yet it can make the difference between a great
sounding speaker system and a mediocre one and it can make the
difference between a stable power amp and one that breaks out into
inaudible high frequency oscillati