I need to add a circuit in my amplifier to convert the balanced input to unbalanced for the LM3886 Gainclone amplifier circuit.
Can I just use an audio line level transformer? Do I need any associated circuitry?
If so, anyone have a model or circuit that they recommend?
Also, I know this can be done with op amps, does anyone have particular op amps and/or circuits that they like with this as well?
Can I just use an audio line level transformer? Do I need any associated circuitry?
If so, anyone have a model or circuit that they recommend?
Also, I know this can be done with op amps, does anyone have particular op amps and/or circuits that they like with this as well?
lgreen said:
I don't quite follow.
What exactly is your "amplifier" as it relates to the "Gainclone amplifier"? The way it's written, it sounds like you're going to use a power amplifier to feed your Gainclone power amplifier.
Do you mean that the source you're going to feed your Gainclone amplifier has a balanced output and you want to convert to unbalanced for the Gainclone?
se
Konnichiwa,
Sure, look up Jensen transformers on the Web. Or you could simply provide a "pseudo balanced" input on the Gainclone. If (for arguments sake) you use 10k/220k Feedback Resistors in the negative input (inverting connection) on XLR Pin 3 you add a 10K series resistor from the positive input to XLR pin 2 and a 220k Resistor to ground from the positive input. To balance the impedances simply connect a 11k Resistor to ground from XLR pin 2. If you use coupling Capacitors they belong into each input directly after the XLR Pin, same value. Voila, 20K balanced Input or close enough anyway.
Sayonara
lgreen said:
Sure, look up Jensen transformers on the Web. Or you could simply provide a "pseudo balanced" input on the Gainclone. If (for arguments sake) you use 10k/220k Feedback Resistors in the negative input (inverting connection) on XLR Pin 3 you add a 10K series resistor from the positive input to XLR pin 2 and a 220k Resistor to ground from the positive input. To balance the impedances simply connect a 11k Resistor to ground from XLR pin 2. If you use coupling Capacitors they belong into each input directly after the XLR Pin, same value. Voila, 20K balanced Input or close enough anyway.
Sayonara
Not so fast
I don't know if I would put a transformer secondary into the 10K resistor for the inverting input of the amp. The contribution from the leakage inductance in series with the 10K input resistor will cause the close loop gain to drop below the minimum gain of 10 required for stability when the frequency gets high enough.
Inverting amplifier designs need a low and preferably resistive source impedance since it becomes part of the feedback network. Jam was telling me not long ago that in his experience that many inverting amplifiers work better with a buffer between the amplifier input and the input resistor to the inverting input. AT THE VERY LEAST, a snubber or R C input filter at the inverting input resistor is a real good idea. You need to think carefully about adding an input transformer to a circuit (even non inverting circuits) as they often don't just "drop in" and get optimal performance. Understanding the transformer parasitics and how they effect the circuit is a good design practice.
I don't know if I would put a transformer secondary into the 10K resistor for the inverting input of the amp. The contribution from the leakage inductance in series with the 10K input resistor will cause the close loop gain to drop below the minimum gain of 10 required for stability when the frequency gets high enough.
Inverting amplifier designs need a low and preferably resistive source impedance since it becomes part of the feedback network. Jam was telling me not long ago that in his experience that many inverting amplifiers work better with a buffer between the amplifier input and the input resistor to the inverting input. AT THE VERY LEAST, a snubber or R C input filter at the inverting input resistor is a real good idea. You need to think carefully about adding an input transformer to a circuit (even non inverting circuits) as they often don't just "drop in" and get optimal performance. Understanding the transformer parasitics and how they effect the circuit is a good design practice.
A while back I worked at a place where we were adding balanced inputs to some industrial VCR's. One of my coworkers contacted the Jensen Transformer people who wanted to know all about our application. A few faxes were exchanged. They recommended a capacitor and a couple of resisters be placed between the output of the transformer and the input to the VCR. I think it was Dean himself who did the calculations.
I did the final fabrication and stuffed the transformers and other parts into an LMB that got bolted to a shelf next to the VCR's. First capacitor melted when I tried soldering onto it.
The application also required -14 dB of attenuation. I don’t remember if we put an H pad in front of the transformer or if they calculated it into the RC network.
They have that application note on the web site now that has an op-amp on the secondary side of the transformer. I don’t know if the company still has the time to do engineering like I got for small orders these days. Back in the mid 80’s when this happened I think Mr. Jensen found it somewhat necessary to evangelize the transformer. Popular wisdom at the time seemed to be that they were evil. He put a lot of time and effort into showing people that his were actually built (very) well and that performance was dependent on getting the engineering correct.
Jensen was a big evangelist for the perfect square wave,
and adjusted his loading accordingly, and was not thinking
in terms of going into a low impedance input, such as the
inverting junction of an op amp.
In any case, if you need to throw away some open loop gain
to stabilize an inverting amp, a resistor to ground from the
virtual ground always does the trick.
and adjusted his loading accordingly, and was not thinking
in terms of going into a low impedance input, such as the
inverting junction of an op amp.
In any case, if you need to throw away some open loop gain
to stabilize an inverting amp, a resistor to ground from the
virtual ground always does the trick.
Nelson Pass said:
With an input transformer such as the JT-11P-1, it's ideally resistively loaded by 10k so as long as the series resistor going into the inverting junction is 10k, he's all set.
Similarly, the JT-11P4-1 is ideally resistively loaded by 20k.
So with either of these transformers, at worst all he'd have to do is change the values of his feedback resistors.
se
Ideally.........
"With an input transformer such as the JT-11P-1, it's ideally resistively loaded by 10k so as long as the series resistor going into the inverting junction is 10k, he's all set."
No........ as the open loop gain of the amplifier falls of the load is not a resistive input impedance at very high frequencies.
A resistor or RC circuit to ground from the inverting input to ground will increase the open loop gain at high frequencies restoring to minimum gain requirements.
Beware of vague advice and especially "advice" containing the word "ideally." It is usually from someone who doesn't really understand the pitfalls in amplifier design.
"With an input transformer such as the JT-11P-1, it's ideally resistively loaded by 10k so as long as the series resistor going into the inverting junction is 10k, he's all set."
No........ as the open loop gain of the amplifier falls of the load is not a resistive input impedance at very high frequencies.
A resistor or RC circuit to ground from the inverting input to ground will increase the open loop gain at high frequencies restoring to minimum gain requirements.
Beware of vague advice and especially "advice" containing the word "ideally." It is usually from someone who doesn't really understand the pitfalls in amplifier design.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.