Winston OBoogie
Well-known member
ruffrecords said:
Gareth would be the best to explain it but, if he doesn't see this thread, I'll copy over some of the blurb that goes with it tomorrow.
Makeup Gain Stage for Passive Summing Mixer Using ETI 430 Line Amplifier
- Thread starter cpsmusic
- Start date
Help Support GroupDIY Audio Forum:
Gareth would be the best to explain it but, if he doesn't see this thread, I'll copy over some of the blurb that goes with it tomorrow.
ruffrecords
Well-known member
Winston O'Boogie said:
OK, cool. Much appreciated.
Cheers
Ian
Interesting, the front end looks a little like the Transamp sum amp application that Paul Buff published back when his Transamp was popular decades ago (of course he didn't publish the innards of the potted Transamps)
The mixed feedback to add opposite polarity signal to the negative feedback path looks to me like like it will increase bus gain, not reduce it. As the master level pot increases some of the negative feedback is subtracted so it appears that the bus amp gain increases too (I'm still on my first cup of coffee so check my work).
Curious to know what value bus feed resistors are used, to evaluate this wrt headroom.
JR
PS: I'm not a huge fan of the 100uF cap in series with the bus but it was necessary in the Transamp application too, for the same reason. Nice to see bus spelled properly.
The mixed feedback to add opposite polarity signal to the negative feedback path looks to me like like it will increase bus gain, not reduce it. As the master level pot increases some of the negative feedback is subtracted so it appears that the bus amp gain increases too (I'm still on my first cup of coffee so check my work).
Curious to know what value bus feed resistors are used, to evaluate this wrt headroom.
JR
PS: I'm not a huge fan of the 100uF cap in series with the bus but it was necessary in the Transamp application too, for the same reason. Nice to see bus spelled properly.
I don't understand the benefit of such a circuit, at least in the context of increasing headroom. U2b still has to do the weightlifting against bus currents.Winston O'Boogie said:
In addition it expands the fader taper, which may not be desirable.
Indeed, I'd be interested in GC's comments.
Winston OBoogie
Well-known member
OK, these are my words, interpreted from the larger text.
First the standard bits:
This is a VE amp as used on aux sends, where turning down a bunch of knobs to avoid overload would be a hassle.
We have a hybrid summing amp using discrete transistors on the front end to give lower noise. Using two transistors in a balanced config. (as done here) will also help with power rail noise without resorting to elaborate de-coupling. The biasing is also much simplified over using just a single transistor.
The feedback for the whole amp must return to the VE transistor to prevent large current swings (which would result in distortion) so we need to invert it first, which is what U1B does.
What’s not shown on the pic I uploaded is the bus feed resistor which, in this case, is 22K.
So at lower settings of RV1, the gain of the amp as set by FB resistor R3 is -6.8dB which should avoid headroom overload.
When RV1 is turned up, a positive feedback signal is sent to U1B’s non inverting input via R11 & R12, and the partial cancellation decreases U1B’s output so we have less feedback and mo’ gain. We also eliminate the usual post summing amp ‘fixed gain’ amplifier so noise is eliminated there too.
To put it into perspective, this particular design by Doug Self and Gareth Connor was used in the Soundcraft Delta which won a British Design award in 1991. I daresay there would be the usual changes and improvements if done today in 2020.
First the standard bits:
This is a VE amp as used on aux sends, where turning down a bunch of knobs to avoid overload would be a hassle.
We have a hybrid summing amp using discrete transistors on the front end to give lower noise. Using two transistors in a balanced config. (as done here) will also help with power rail noise without resorting to elaborate de-coupling. The biasing is also much simplified over using just a single transistor.
The feedback for the whole amp must return to the VE transistor to prevent large current swings (which would result in distortion) so we need to invert it first, which is what U1B does.
What’s not shown on the pic I uploaded is the bus feed resistor which, in this case, is 22K.
So at lower settings of RV1, the gain of the amp as set by FB resistor R3 is -6.8dB which should avoid headroom overload.
When RV1 is turned up, a positive feedback signal is sent to U1B’s non inverting input via R11 & R12, and the partial cancellation decreases U1B’s output so we have less feedback and mo’ gain. We also eliminate the usual post summing amp ‘fixed gain’ amplifier so noise is eliminated there too.
To put it into perspective, this particular design by Doug Self and Gareth Connor was used in the Soundcraft Delta which won a British Design award in 1991. I daresay there would be the usual changes and improvements if done today in 2020.
OK, I got it. A standard inverting opamp (with or without discrete front end) would have the same headroom, but with the help of some PFB, it is capable of providing some additional gain. So it saves a post-fader stage. A standard VE with the fader in the NFB could also work, but would require a higher value than the usual 10k.Winston O'Boogie said:
Once the goals are understood it makes more sense.
![Electronics Soldering Iron Kit, [Upgraded] Soldering Iron 110V 90W LCD Digital Portable Soldering Kit 180-480℃(356-896℉), Welding Tool with ON/OFF Switch, Auto-sleep, Thermostatic Design](https://m.media-amazon.com/images/I/41gRDnlyfJS._SL500_.jpg)
Winston OBoogie
Well-known member
abbey road d enfer said:
Yep I think so. Me forgetting to include the bus feed resistor in the screen capture didn't help make it easy to immediately see what was going on so I apologize, but we got there in the end.
I'm also sure there are other, equally valid, ways to do this, I just wanted to give one example since the question was raised by Ian.
cpsmusic
Well-known member
This amp relies on an input xfmr for delivering good noise performance, because, being battery powered, it runs at very low current.cpsmusic said:
The xfmr will always be the weakest link in the chain. You have to spend big to get performance that could be achieved efficiently with a properly optimized circuit.
