[ppl]

Dyn-O-Bias II for your PPA2.0

I was comparing two similar PPA Amps today, One a PPA 2.0 and the other a PPA1.0 with A Dyn-O-Bias Module installed. Both were fully equipped with Black Gate capacitors thought the power supply in the Black Gate supper E configuration. Pith amps were simultaneously powered by the Lab power No 9 Audio Grade power supply and the gains were the same at 20 dB. The big surprise was I was not aware that these two amps were configured any differently and that each was the full build and ether a PPA V 2.0 or v 1.0 with the discrete buffer modules that were available for the PPA. At first I was not intending to do any listening comparisons how ever having two similar amps both operating at the same time and at first with separate sources so I could just unplug and plug in headphones and not have any input switches degrading the signal. (I am a fanatic about the minimum number of non soldered contacts in the signal path) The input and output interconnection only topology keeps with the KISS philosophy popular now days.

At first I was happily listening to different program material from the two and did not notice any difference in tonality or extension bass slam and clarity but for some reasoned one was captivating me more than the other, I was getting ready to switch systems because a Harry James song was coming around and I am really not a Jazz fan. So I simply unplugged from one system and immediately into the other for non stop music while I changed the CD in the system I just got done listening to. Well I found myself listening and thinking maybe the jazz aficionado hare on to something I previously was not paying attention to. Now I have listened to some of the best sound systems in the world and jazz never really interested me with the exception of Dixieland jazz. Interestingly enough I found myself intently listening to this Harry James version of take the A train and noticing how musical the horns sounded and I was listening for and could identify all kinds of clues that allowed me to see quite well into the recording studio way back when in the day that was recorded I do love some of that, otherwise I have quite diverse musical tastes and one of them is electronic and Ambient music and I was listening to. After that song I switched to the other system where Rodger waters magnificent atmosphere music. This is totally magnificent in creating a 3D holographic space. I use it often to check micro dynamics on system Auditions because it’s just loaded with quickly disappearing morsels of fading conversations with heavy spatial processing. Well in comparison to the previous Harry James recording I was listening to on the other system Amused to deaths normally magnificent first track is now sounding somewhat less than magnificent in comparison. I thought

Ok it’s the source I know this, but since its all set up lets compare the two amps using only one source. Yes I know what a hassle interconnecting and un-Interconnecting (is that a word) in put and headphone plugs complicated by the locking nutrick headphone jack on the PPA. It did not take me long to conclude that one of these PPA’s must be defective as they sound substantially different yet the inferior sounding Amp did not act like any technical problems were going on, no clicks in the headphones indicating instability or high DC offset and no distortions that I could pinpoint yet something surly was amiss somewhere. Well since I had no test equipment or for that matter a soldering iron available so I just continued casually comparing the two to hopefully ascertain what the actual differences were each has similar bass and treble extension and equally clear and even tonality (Hope that’s a word) was real close so much so that I assumed both had the same op amps. After hours compared on a verity of musical styles I concluded the difference was defiantly in the midrange and it was in spatial imaging and in concert with what appeared a hyper since of Air and space almost like the music was surrounding your head as opposed to a flat almost two dimensional soundscape. At this point I decided to remove the rear cover on the less than stellar PPA and take a look because I might get lucky a lose wire on a jack or something could cause that.

After I got the amp apart ya nothing appeared out of order the buffer module was sitting in the sockets well and everything appeared ok then while poking around I noticed the words on the buffer board Dyn-O-Bias II. What is this I thought I remember doing a rework of the Dyno bias to address stability issues Dyno One was having in some Amps? Well this dyno biased amp sure appeared way more stable than normally I remembered the dyno bias being. My quick stability test is touching the non inverting input of the op amp at min volume and headphones around your neck. On your ears could get annoying.

Since this circuit can be hacked into a PPA V2 simply connect the Diode and emitter resistor in series prior to installing them or if working on a completed Board unsolder one side of the emitter resistor. Then all that’s required is soldering the CRD diodes from the proper supply rail to the emitter of there respective output transistor.

 

[Epicurean]

Hmm, that was a bit confusing. So, the Dyn-O-Bias II sounded better? Or the other way around?

I've been having some stability issues with my PPAv2 with the regular buffers, mainly with high impedance headphones it seems, and it seems it depends on which power outlet I use in my house

After listening for a few minutes there is a pop and DC at output shoots from 3mV to over 1.5V! I recently started a thread about that and haven't found the problem yet. Don't mean to hijack this thread

 

[ppl]

the Midrange is alot better on the Dyn-O-Bias and is as stable as the PPA V2. As for your stability issues i have found most involve the ground channel and with the opa-627 this is resolved by a larger CC in the Ground Channel a secure grounding of the Input jacks and a Cleen PC Board left over solder flux is real bad and most DIY'ers do not cleen this off with solvent

 

[thedoctor]

finished building this yesterday :

and though i would say that it indeed has better midrange compared to this

which is built with the same components (minus the Dyno Bias of course), i did found that my new Dyno Biased PPA have a bit peculiar behavior regarding the bias adjustment.

adjusting the bias also affects the DC offset which i presume not supposed to happen, right? the dynamic bias on the other hand does work, and the bias changes automatically when given a load.
anyone care to shed some light?

^ edit : i did traced the circuit a few times to make sure that i didn't wire something wrongly.

 

[ppl]

Quote:

Originally Posted by thedoctor finished building this yesterday : and though i would say that it indeed has better midrange compared to this which is built with the same components (minus the Dyno Bias of course), i did found that my new Dyno Biased PPA have a bit peculiar behavior regarding the bias adjustment. adjusting the bias also affects the DC offset which i presume not supposed to happen, right? the dynamic bias on the other hand does work, and the bias changes automatically when given a load. anyone care to shed some light? ^ edit : i did traced the circuit a few times to make sure that i didn't wire something wrongly.

How much dose the DC offset change with Bias adjustment a few mV is norm. I will look at your Photos and see if everything looks correct. Thanks for trying this out

 

[thedoctor]

Quote:

Originally Posted by ppl How much dose the DC offset change with Bias adjustment a few mV is norm. I will look at your Photos and see if everything looks correct. Thanks for trying this out

most probably it's correct. i'm sure i'm doing things right since the other PPAv2
works without a hitch.
i did tested all the transistors and JFETs and TLE2426 before soldering them. but just to be sure, i'll build me another one on a new board and see if i get to the same effect.

offset increases from less than 1mV to about 30mV after going pass 90% of the bias adjustment. the same thing happens in all 3 channel. and thus i left the bias below this point and the amp outputs less than 10mV offset while being unloaded.

without load, the bias idles at 5-8mA, but once i plug in my 32Ohm cans, the bias increases to 35-40mA. and thus i presume the dynamic bias is working as it should.

and many thanks to you for sharing this mod to us all.

 

[ppl]

Yes from what you say the dynamic biasing is working correctly. That changing DC offset at high Bias is something I never had an issue with even with the Dyn-O-Bias I that had stability issues if not properly set up.

Your DC offset issue could be caused by mismatched Diodes the In4XXX see that the diode drop in these are the same on all (4) Diodes. In addition you could change the 2R2 ohm resistors with 4R7 ohms and the 10 ohms with 22-47 ohms and get rid of some of that overlap at High bias settings as explained below

One thing that could cause this is what is called Beta Doubling. This happens in Class A and dynamically Biased Complementary output stages when the crossover point is overlapped between the PNP and NPN output transistors. at and beyond this point Both topologies may go into oscillation because of the increased Current gain caused from the operation of both transistors at the crossover point and at the same time.

 

[Epicurean]

I finally got a 'real' case for my PPAv2 and I recently upgraded the buffers to the dyno-bias-II. I didn't follow ppl's dyno-bias-II schematic exactly and left the q23/33s connected directly to the power rails since moving them would've been too much work for me. Now, after a lot of listening I can pretty safely say there is an improvement in the sound. The midrange is a little smoother and more natural to my ears, while the bass and the treble remains quite the same.

Maybe the dyno-bias-II should be added to tangents build instructions? At least if we get even more confirmations of the improvements.

 

[Vladco]

PPL could you please make some comment on schematic. Is 1N5298 in the same directions as 1N4935? Could I consider the 1N4935 as an extra in PPA V2?
Vlad.

 

[ppl]

If you are not reconnecting the driver transistors collectors to the output transistors emitter then you don’t need D1-D4, the propose of these diodes is to develop a voltage drop so as to keep the driver transistors from turning off. CRD1 and CRD2 are then all that need be installed. This is no longer a Dyn-O-Bias circuit but is a non-switching circuit. The bias thus remains fixed as it is in the stock configuration; however the current sources keep the non-conducting transistor from turning off and thus eliminates the dead zone or crossover region gap known to induce crossover distortion. The bands on the CRD go towards the most negative supply potential so the bands will be the same as the (-) symbol on the schematic. I’m glad some one is likening this mod

 

[Clutz]

Hi ppl;

To perform this upgrade on a PPA it would be optimal to include D1-D4 as well as the CRDs? Is the point of D1-D4 to prevent cross over distortion from occuring in the driver stage, and the point of the CRDs is to prevent crossover distortion in the output stage?

 

[ppl]

the only reason to install D1-D4 is to devlop enought operating voltage for the drivers when the colectors of the driver transistors are conected to the emiters otherwise the driver transistors will have only 20-100 mV Vce and that is not enought to turn them on.