THAT1646 Current Booster and Transformer Driver

Where we discuss new analog design ideas for Pro Audio and modern spins on vintage ones.
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: THAT1646 Current Booster and Transformer Driver

Post by JR. »

Connecting a few dots from different threads (and life), One comment made in one of the Bateman capacitor article stuck with me.. the distortion in caps was reduced when driven from low impedance and increased at higher impedance (i.e. distortion was a current so ohms law applies for translating the distortion current into a distortion voltage.

Now stepping into the way-back machine to a phone conversation I had with (the) Deane Jensen back in the late '70s about driving output transformers and he made a similar comment, namely driving transfirmers from a lower impedance, resulted in lower distortion. Just imagine a transformer guy who actually wanted low distortion!

But connecting the dots, apparently some distortion mechanisms in transformers look like currents, so driving from a low Z is apparently a good thing.

I still live by the mantra that such things can usually be avoided, but different strokes for different folks. The customer is still always right.

JR
Cancel the "cancel culture", do not support mob hatred.
User avatar
mediatechnology
Posts: 5442
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: THAT1646 Current Booster and Transformer Driver

Post by mediatechnology »

OK, so the internal build-out with both 1646 outputs in parallel is 12.5R.
Connecting a few dots from different threads (and life), One comment made in one of the Bateman capacitor article stuck with me.. the distortion in caps was reduced when driven from low impedance and increased at higher impedance (i.e. distortion was a current so ohms law applies for translating the distortion current into a distortion voltage.

Now stepping into the way-back machine to a phone conversation I had with (the) Deane Jensen back in the late '70s about driving output transformers and he made a similar comment, namely driving transfirmers from a lower impedance, resulted in lower distortion. Just imagine a transformer guy who actually wanted low distortion!

But connecting the dots, apparently some distortion mechanisms in transformers look like currents, so driving from a low Z is apparently a good thing.
JR: Hmm. Seems to make sense to me. The lower the source impedance the less the non-linearities of the load modulate the source to decrease distortion. Isn't this like the delta-C thing with FET input op amps and high impedance sources? Didn't SamG see something like this in his op amp tests?
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: THAT1646 Current Booster and Transformer Driver

Post by JR. »

Not to veer off on a tangent... but I did have an exchange with Sam about a related issue. The new uber opamps are so clean that the only way they can make meaningful bench measurements is to run then at elevated closed loop gains of 100x or something like that and extrapolate the distortion by dividing it down by that gain. My concern is that the 100x closed loop gain termination, typically involves a low resistance resistor from the - input to ground. The + input is likewise typically driven from a low impedance for low noise.

FWIW I never did find a smoking gun suggesting that the uber opamps are not as good as they claim (-140dB), while I remain a little suspicions of that test methodology. Likewise it is probably a good idea to terminate both inputs with similar impedance, assuming internal symmetry.


JR
Cancel the "cancel culture", do not support mob hatred.
ricardo
Posts: 262
Joined: Sun Apr 10, 2011 7:24 am

Re: THAT1646 Current Booster and Transformer Driver

Post by ricardo »

mediatechnology wrote:The lower the source impedance the less the non-linearities of the load modulate the source to decrease distortion.
You can do better than that. At LF, tranformer distortion is purely 3rd harm and goes away if you have zero resistance in the windings. If yours aren't wound with the tresses of Swedish virgins, you can introduce -ve resistance to cancel.

http://www.lundahl.se/pdfs/papers/feedbck.pdf

We used an improved version with both parts of a NE5532 in Calrec M-series. Not sure if Ken Farrar (Tech. Dir. at Calrec and my electronics mentor) saw the German patent. It's all around the same time. We were going to bed with Lundahl from our previous affair with Sowters.

Transformer outputs with LF distortion, overload & response of transformerless.
Last edited by ricardo on Wed Apr 13, 2011 6:01 pm, edited 1 time in total.
User avatar
mediatechnology
Posts: 5442
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: THAT1646 Current Booster and Transformer Driver

Post by mediatechnology »

Thanks for the link. I saved it.

I think Audio Precision got a patent on something similar as well which they used in the original System One. It is now expired. I may have a copy of that somewhere...

Hofer, "Low-Distortion Transformer-Coupled CIrcuit," US Patent 4614914.

http://www.proaudiodesignforum.com/Imag ... 614914.pdf
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: THAT1646 Current Booster and Transformer Driver

Post by JR. »

Once again I am shocked... Low distortion transformers seem counter productive for all the folks who use them solely to add coloration. :lol:

I recall Jack Sondermeyer (at Peavey) making a power amp that was capable of negative output impedance, to deliver numerically high damping factor, by canceling out the small real resistance in output jack and internal wiring. This was mostly a cosmetic benefit that appealed to people who read spec sheets (not typical Peavey customers) so this was only used in one model and abandoned. (Coincidentally he used a transformer in his design so the sense was floating and worked even when the amp output was bridged.)

Negative impedance has other benefits when driving loudspeakers, but a little too hip for Peavey to pursue 20 years ago (or ever?). I didn't even try to explain to customers that we could compensate for resistive wiring losses.

JR
Cancel the "cancel culture", do not support mob hatred.
User avatar
mediatechnology
Posts: 5442
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: THAT1646 Current Booster and Transformer Driver

Post by mediatechnology »

Once again I am shocked... Low distortion transformers seem counter productive for all the folks who use them solely to add coloration.
Yeah, well. I suppose one could put a knob on the negative impedance and label it "color." Crusty, you readin' this?

There is a lot to be said for a dab of compressive THD-3.
I recall Jack Sondermeyer (at Peavey) making a power amp that was capable of negative output impedance, to deliver numerically high damping factor, by canceling out the small real resistance in output jack and internal wiring. This was mostly a cosmetic benefit that appealed to people who read spec sheets (not typical Peavey customers) so this was only used in one model and abandoned. (Coincidentally he used a transformer in his design so the sense was floating and worked even when the amp output was bridged.)

Negative impedance has other benefits when driving loudspeakers, but a little too hip for Peavey to pursue 20 years ago (or ever?). I didn't even try to explain to customers that we could compensate for resistive wiring losses.
Wasn't that what Crown's "Delta-Omega" was about? I remember us trying to use that at Dallas Sound Lab with UREI monitors. There was this calibration trim on the front panel to set the negative impedance. IIRC you set it to where this low-level oscillation kicked in and then backed off. This caused the woofer to be highly damped - so much in fact that we had issues with the aluminum voice-coil fracturing off the cone.

I think it may have been a little too hip for Crown too.
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: THAT1646 Current Booster and Transformer Driver

Post by JR. »

I am not very familiar with the delta omega, but their literature claims it is compensating for both resistive and reactive components in the woofer.

Sondermeyer actually got a patent for his, so clearly not exactly the same, but I am not confident in the patent offices vetting of inventions. I am also aware of other patents by other guitar amp engineers at Peavey to intentionally degrade amplifier damping factor. With guitar amps one of the desirable characteristics of old school tube guitar amps, is how their relatively high output impedance, interacts with speaker/cabinet resonances to actually hype or enhance these cabinet resonances. This phenomenon was covered electronically in cheap solid state guitar amps with a variable output impedance that was adjustable and limited to either HF or LF. I recall seeing them messing with both HF and LF resonance knobs while that seems like more than would be useful.

While I am not a speaker expert, I know enough to know that negative output impedance alters some driver characteristics and therefore box tuning. I suspect Crown engineers were a little optimistic to expect their circuit to work with all speakers, and probably dialed it in to do something useful for a few popular models. You can probably dial it off or down when it wasn't helpful, while I bet this was a nightmare for their service department.

I have long been a proponent of powered loudspeakers with DSP and tricks on board so electronic magic can be performed inside the box.

JR
Cancel the "cancel culture", do not support mob hatred.
ricardo
Posts: 262
Joined: Sun Apr 10, 2011 7:24 am

Re: THAT1646 Current Booster and Transformer Driver

Post by ricardo »

JR. wrote:I know enough to know that negative output impedance alters some driver characteristics and therefore box tuning.
The most sophisticated version of this was ACE by AudioPro, Sweden in the early 80's who made some very nice subs. But these methods are seriously affected by speaker voice coil heating. David Birt (ex BBC) had the speaker & amp in a DC bridge to compensate for this on the fly.
I need to dig up some AES & IOA papers but this was in some of the zanier implementations of my patented Powered Integrated Super Sub technology. Wharfedale marketing never liked that name for some reason.
My specialty is integrating electronics, DSP, linear & non-linear into speakers. eg to me, a surround system is a speaker. You put electrical signals in at one end and get a noise at the other. Today, the sensible place to put digital room & speaker EQ is your surround system.
User avatar
mediatechnology
Posts: 5442
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: THAT1646 Current Booster and Transformer Driver

Post by mediatechnology »

David Birt (ex BBC) had the speaker & amp in a DC bridge to compensate for this on the fly.
I've followed David's work on audio interfaces and did a fair amount of work adapting his line reciver to the 1570 mic preamp. Small world. viewtopic.php?f=6&t=347&start=7
Post Reply