Bern,
Those look good.
Can't wait to see the rest
Those look good.
Can't wait to see the rest
Zen -> Cen -> Sen, evolution of a minimalistic IV Converter
- Thread starter EUVL
- Start date
I have built CEN for AD1862. Overall chain: AD1862 -> CEN - >Potentiometer -> B1 buffer ->out. Settled on floating supplies from 18650 batteries (5x3.7V), everything else on shunt PSU's. Overall I am very happy
There are still two questions I would like to ask:
1. In the first couple of listening tests I had the sound deteriorate after maybe 20-30 minutes of listening. I would start getting noisy cracking background and overall sound "quality" would drop a lot. Batteries powering the CEN are not even close to being discharged. Now after a few days this effect is gone. Same everything, same batteries (not recharged). What gives? I do get a bit better ventilation as compared to early tests, did something overheat?
2. Is it just me, or does this thing somehow still work with floating supplies not powered on?
There are still two questions I would like to ask:
1. In the first couple of listening tests I had the sound deteriorate after maybe 20-30 minutes of listening. I would start getting noisy cracking background and overall sound "quality" would drop a lot. Batteries powering the CEN are not even close to being discharged. Now after a few days this effect is gone. Same everything, same batteries (not recharged). What gives? I do get a bit better ventilation as compared to early tests, did something overheat?
2. Is it just me, or does this thing somehow still work with floating supplies not powered on?
I have built almost the same configuration : AD1862->CEN->SJ74/SK170 Buffer->Tribute AVC.
The circuit will still work with the floating supply OFF, but will have a much higher distortion, because the dac will now see a Riv resistor, that is much too high in value
The circuit will still work with the floating supply OFF, but will have a much higher distortion, because the dac will now see a Riv resistor, that is much too high in value
Look for oscillations with a scope.
Put your batteries and the CEN IV inside a grounded Faraday's cage.
(Connect cage to DAC Gnd, not CEN)
Patrick
Thanks. Batteries I can do, but CEN is on integral all-in PCB. I mean nothing is set in stone and I can redo the whole thing, but will go step by step for now
In a latest version of SEN with very low capacitances, I found that it was picking up HF easily without shield.
For example from LED lamps.
Patrick
For example from LED lamps.
Patrick
This has a peak ouput current of 11mApp.
The original CEN was designed for 4mApp.
So you need a lot of JFETs in parallel.
And then the gate capacitance increases.
Not impossible, but not straight forward.
Patrick
The original CEN was designed for 4mApp.
So you need a lot of JFETs in parallel.
And then the gate capacitance increases.
Not impossible, but not straight forward.
Patrick
Just found this thread. Haven't gone thru all pages, so I don't know how it evolved.
Looking at the orig schematic:
https://www.diyaudio.com/community/attachments/sen-png.237594/
How does this I/V compare to the one Jocko posted in 2002?
https://www.diyaudio.com/community/threads/easy-to-build-i-v-stage.6121/
https://www.diyaudio.com/community/attachments/i_v-gif.2701/
Looking at the orig schematic:
https://www.diyaudio.com/community/attachments/sen-png.237594/
How does this I/V compare to the one Jocko posted in 2002?
https://www.diyaudio.com/community/threads/easy-to-build-i-v-stage.6121/
https://www.diyaudio.com/community/attachments/i_v-gif.2701/
Totally different design and even working principles.
I don't see how they should / could be compared.
Maybe you should at least read the article on the Linear Audio website first to understand how it works.
Cheers,
Patrick
I don't see how they should / could be compared.
Maybe you should at least read the article on the Linear Audio website first to understand how it works.
Cheers,
Patrick
Were you asking for a technical comparison of circuit topologies? About reviews or opinions on perceived SQ? Suitability for use with particular type of dacs? Cost to build? FFT measurements? Input Z? Something else?
Yes, all of the above! I've only just glanced at the Sen, but it "looks" similar to Jocko. Similar number of transistors , etc. They are both cathode, correct?
IAC, which -- Sen or Jocko -- would would be better for one I/V vs another I/V? I'm guessing Jocko's may be a bit better because of the Sen's single-rail design ???
I did build the Jocko I/V over a decade ago with VERY good results in my TDA1541-based CDPs. The Jocko's are used essentially as discrete drop-in op-amps. I'm only using the default +/- 15v rails (instead of Jocko's suggested +/- 18 v).
Yes, apologies -- my bad for having only a cursory glance at your thread. Yes, totally diff. Yours uses FETs, etc.
IAC, my above query about which -- Jocko or Sen -- which would be more ideal in a classic CDP (see images I posted), such as TDA1540 or 1541 , still stand
SEN IV must use floating power supply.
So no use for you.
Cheers,
Patrick
So no use for you.
Cheers,
Patrick
Hi,
the Cen/Sen suites well DACs with a symmetrical current swing wo. offset current., while the Jocko circuit suites well DACs with an offset current (sinking in the above linked circuit).
For current sourcing DACs with offset current like the popular TI PCM179x series a 'inverted' Jocko utilizing PNPs fits better.
In that case one could not only omit with the biasing current source Q3, but overall current consumption, resp. heat power could be reduced also.
The DAC output would only be connected to the PNP-Emitter (Q2).
Since the offset current is always larger than the peak output current -never reaching, let alone crossing 0mA- the circuit remains biased under all circumstances ... hence the extra current source (Q3) is not required.
Also not required would be the positive supply line, making things even easier.
jauu
Calvin
the Cen/Sen suites well DACs with a symmetrical current swing wo. offset current., while the Jocko circuit suites well DACs with an offset current (sinking in the above linked circuit).
For current sourcing DACs with offset current like the popular TI PCM179x series a 'inverted' Jocko utilizing PNPs fits better.
In that case one could not only omit with the biasing current source Q3, but overall current consumption, resp. heat power could be reduced also.
The DAC output would only be connected to the PNP-Emitter (Q2).
Since the offset current is always larger than the peak output current -never reaching, let alone crossing 0mA- the circuit remains biased under all circumstances ... hence the extra current source (Q3) is not required.
Also not required would be the positive supply line, making things even easier.
jauu
Calvin
It would be good if someone could create TABLE of the I/V topologies you hint at above, in concordance with their "ideal" DAC topologies .
I have not extensively experimented with discrete FET i/v's, other than the Prometheus DIY D1 I/V Stage from Pass Dac (this one works very well with TDA1541).
For single-rail, I have used the Single rail, active I/V for TDA1543, TDA1545A, as noted here:
https://www.diyaudio.com/community/threads/single-rail-active-i-v-for-tda1543-tda1545a.28144/
https://diyparadise.com/w/rudolf-broertjes-ss-iv-gain-stage/
Interesting that rbroer notes in the 2004 OP: "IMHO it sounds better than the active I/V posted in the "simple I/V" threads, but not as smooth as the folded cascode."
The Prometheus i/v is especially good --- not surprising given those +/-30 volt rails!
Back to CEN/SEN ... the Group Buy economics of that PCB and parts -- offered elsewhere on DIYA -- are reasonable. Given rbroer comment about folded cascode, I may be convinced to implement the project (either on Veroboard or the GB pcb) --- but more of you will have to convince me that CEN/SEN IV is superior for Philips TDA dacs than the Jocko or Prometheus or Rudolf-Broertjes kits I already have in my kit
Hi,
the RBroer stage is imho only good for the TDA1543, but not the TDA1541A, due to the voltage levels the I/V input resp. DAC-output should settle on.
2.2V is the level for the TDA1543 and 0V for the TDA1541A (see#16 of the RBroer thread for voltages).
Both beeing current sinking DACs the Jocko and RBroer NPN based topologies should generally suit them well.
And supplying from a Dual supply the RBroer could suit the TDA1541A also ... in the end its a Jocko with cascodes, including its limitations.
I also don´t share the optimizm about the folded cascode design.
In simulation already it never worked as well as the current mirror.
Never got any really good THD, PSRR etc. sims, while the CM stuctures simmed well and proved their qualities in praxis as well.
jauu
Calvin
the RBroer stage is imho only good for the TDA1543, but not the TDA1541A, due to the voltage levels the I/V input resp. DAC-output should settle on.
2.2V is the level for the TDA1543 and 0V for the TDA1541A (see#16 of the RBroer thread for voltages).
Both beeing current sinking DACs the Jocko and RBroer NPN based topologies should generally suit them well.
And supplying from a Dual supply the RBroer could suit the TDA1541A also ... in the end its a Jocko with cascodes, including its limitations.
I also don´t share the optimizm about the folded cascode design.
In simulation already it never worked as well as the current mirror.
Never got any really good THD, PSRR etc. sims, while the CM stuctures simmed well and proved their qualities in praxis as well.
jauu
Calvin
Yes, as the RBroer thread title clearly states: it is only intended for TDA1543, 1545A [the 8-pin Philips TDAs]. It should be noted that diyparadise.com was able to iterate that orig. design into further refinement in their myriad "mojo" PCBs:
https://diyparadise.com/w/give-me-mojo/
https://diyparadise.com/w/mojo-fy-your-mojo/
https://diyparadise.com/w/whats-better-than-mojo-mojo2/
This product has been discontinued for over a decade. But the schematics are still avail. I built the orig. (SS I/V, not Mojo) with output coupling caps shorted.
====================
Apologies ... Back to CEN/SEN i/v... playing catch-up here ... I've been absent from diy electronics for over a decade and have forgotten much .... so one is Single-Ended and and the other Complimentary ... which dac (make, model, topology) is best for CEN and which SEN? Or is the application more nuanced? Thanks!!
BTW: A small niggle. When referring to a current-to-voltage circuit, use a slash "/" between the I and V. Such as I/V. Other wise the search engines will mistake it for a Roman Numeral "IV" (four). Or newbies will mistake it for "mark four". The CEN/SEN threads don't use slash.
I'm not I quite get why one would want the added complexity of cascode in an audio i/v stage? TTBOMK, Miller affects very high freqs., and not audio (human ear) bandwidth. Or does cascodting improve buffering, S/N or dynamic range?