Re: Oscillation??
Which circuit?
Terry
Zodiac said:Hi guys,
I'm having some problems with this circuit...
When adjusting pot R2 to get 0V DC on the output, the voltage jumps between +12V and -12V and it won't stay anywhere near 0. I am using SA1085 and SC2546. Everything else is per schematic/notes, except the 100r resistors are not surface mount.
Could this be the dark horse of oscillation rearing its ugly head?
Any help/ideas/solutions greatly appreciated
Which circuit?
Terry
Thanks Rudolf! Finally finished my implementation of Rudolf's common base circuit in a separate DAC. Details:Welwyn RC55 resistors
Panasonic FC caps
SA1085 & SC2546 transistors
Hard wired on perf board. I love hard wiring, but someday I will have to make some pcbs especially for the digital stuff....
No DC servo as yet - I am planning on putting a volume control in the DAC followed by a diamond buffer.
And the sound - large increase in clarity, everything seems "purer" with real separation to instruments, vocals, etc. Detail retrieval seems to have improved also and soundstage opened up a bit. But the biggest shock to me was the bass, a great improvement in tightness and extension and real body
Thanks Rudolf for sharing your circuit
Re: Bax
Zodiac, did you try different currents ?
. I created the pcb with smd base stopper resistors, but I ended up "jumpering" these with solder drop.
Great Zodiac. You might want to consider adding a jfet cascode on the current sink for the PNP transistor. IMHO it does improve sound more than the Baxendall super pairs.Zodiac said:
Zodiac, did you try different currents ?
The double Baxandall pair I used in the "Super Pair I/V" thread didn't oscillate as far as I could noticepeufeu said:
. I created the pcb with smd base stopper resistors, but I ended up "jumpering" these with solder drop.Hi!
I did try the circuit with and without the jfet cascode. It sounded better with the cascode in, so that's where it stayed!
With regard to bias levels, I have not tried different values for the current sources...any tips here?
The whole circuit has to burn in properly, I have to admit I was so impressed with the result that I'm not that inclined to start fiddling
I did try the circuit with and without the jfet cascode. It sounded better with the cascode in, so that's where it stayed!
With regard to bias levels, I have not tried different values for the current sources...any tips here?
The whole circuit has to burn in properly, I have to admit I was so impressed with the result that I'm not that inclined to start fiddling
With or without an input signal...
With an input signal, I gen the signal with oscillations superimposed ; without an input signal I just get the oscilations.
Layout is very tight, ground plane, good decoupling, etc...
Note that the super pair also showed signs of oscillations in the simulator !
With an input signal, I gen the signal with oscillations superimposed ; without an input signal I just get the oscilations.
Layout is very tight, ground plane, good decoupling, etc...
Note that the super pair also showed signs of oscillations in the simulator !
rbroer's stage in a Rotel RCD-855
Sorry to raise this old of thread from the dead
Finished up the prototype for a discrete I/V stage in my Rotel RCD-855 CD. This player uses the standard Philips late '80s/early '90s chipset (SAA7210 / SAA7220B / TDA1541a). It sounds really good, much better than any opamps I have rolled through this player. I really wanted to compare it to a G08 or something at a head-fi meet last weekend, but hadn't gotten in enough listening/testing yet, so wound up not going.
Mods include:
Thanks to Rudolf for designing and sharing this!
Sorry to raise this old of thread from the dead
Finished up the prototype for a discrete I/V stage in my Rotel RCD-855 CD. This player uses the standard Philips late '80s/early '90s chipset (SAA7210 / SAA7220B / TDA1541a). It sounds really good, much better than any opamps I have rolled through this player. I really wanted to compare it to a G08 or something at a head-fi meet last weekend, but hadn't gotten in enough listening/testing yet, so wound up not going.
Mods include:
- Shottky PSU diodes (11DQ10 player, 31DQ10 o22)
- Tentlabs XO on a PFM Flea board (modified design with 2 PSU sections, one for the XO, one for the 74HC logic), with /2 direct DAC clocking
- Tread board taking ~25Vdc down to 18-20Vdc for the Flea
- home layout of an amb o22 PSU for the discrete stage (I needed a smaller board than the o22 at 3.5" x 6.5")
- rbroer's less simple I/V for TDA1541a, modified to use 2SA970/2SC2240 transistors, FET buffer per cetoole's I/V stage (I removed this for the time being, works but needs redesign), and muting relay (not put in yet)
- home-etched double-sided PCBs (I used tinnit for the first time on these, works pretty well)
Thanks to Rudolf for designing and sharing this!
An externally hosted image should be here but it was not working when we last tested it.
Thanks for the comments!
From what Rudolf told me via email, he is still using this one... not sure if he ever built any of the others in that thread or not. In mine I'm currently using 1.5K I/V resistors and 1nf caps... I have not measured the output but it might be a touch hot; I may try 1.3k instead. As far as Zout, ??? not sure.
Chris
From what Rudolf told me via email, he is still using this one... not sure if he ever built any of the others in that thread or not. In mine I'm currently using 1.5K I/V resistors and 1nf caps... I have not measured the output but it might be a touch hot; I may try 1.3k instead. As far as Zout, ??? not sure.
Chris
Hi Chris,
What I was inquiring about was whether the "less simple" I/V stage is a gain stage such that one can get say 2V output like a standard digital source...or do you need a gain stage in addition to the discrete I/V? I am only familiar with the traditional 33Rish passive I/V + tube gain stage for the 1541a.
Also, by Zout I mean the output impedance. I'm currently using a TVC volume control and would like to keep the impedance below 1k ohm if possible. I imagine with an SS stage like this it probably isn't an issue.
Thomas - I interpret Chris's response to mean that transformer I/V would not be his #1 choice.
BTW Thomas, I am enjoying your v2 twin 1541a dac but have completely bypassed the cs8414 and am feeding the dac signal pins directly with a synchronous reclocking board that receives left justified signal from squeezebox transport and then converts them to i2s and resynchs them with tent 11.2896 clock. The master clock is in the dac. The resolution and soundstage have improved dramatically.
What I was inquiring about was whether the "less simple" I/V stage is a gain stage such that one can get say 2V output like a standard digital source...or do you need a gain stage in addition to the discrete I/V? I am only familiar with the traditional 33Rish passive I/V + tube gain stage for the 1541a.
Also, by Zout I mean the output impedance. I'm currently using a TVC volume control and would like to keep the impedance below 1k ohm if possible. I imagine with an SS stage like this it probably isn't an issue.
Thomas - I interpret Chris's response to mean that transformer I/V would not be his #1 choice.
BTW Thomas, I am enjoying your v2 twin 1541a dac but have completely bypassed the cs8414 and am feeding the dac signal pins directly with a synchronous reclocking board that receives left justified signal from squeezebox transport and then converts them to i2s and resynchs them with tent 11.2896 clock. The master clock is in the dac. The resolution and soundstage have improved dramatically.
Riotubes: I guess it is a gain stage in the sense that it takes the 2mA current from the DAC, moves it while trying to keep the input impedance that the DAC sees very low, and runs it through the I/V resistor (in my case 1.5K ohms) resulting in approximately 2Vpp (I guess) output. No additional gain stage is required, although a buffer stage might be if you were to run long ICs. The design is based on Jocko's I/V stage, and is a transimpedance amplifier (I guess that is the term normally used).
As far as its output impedance (Zout), would this in effect be the I/V resistor (1.5K?). Or is it the I/V resistor in parallel with the impedance of Q6, etc.? The schematic is in the 2nd or 3rd post of this thread. The only resistances that I adjusted were for the current sources in order to run the 2SA970/2SC2240s below 10mA (actually, more like 9mA max).
thomas: I guess I'm not sure what you are asking? Are you asking if this stage could be used with an output transformer? I suppose it could be... you would probably want to omit the DC servo, etc. in that case, and possibly put an output capacitor in if the transformer is sensitve to DC.
You are lucky you don't have to try to understand my (nonexistant) Chinese... your English is quite good compared to that.
As far as its output impedance (Zout), would this in effect be the I/V resistor (1.5K?). Or is it the I/V resistor in parallel with the impedance of Q6, etc.? The schematic is in the 2nd or 3rd post of this thread. The only resistances that I adjusted were for the current sources in order to run the 2SA970/2SC2240s below 10mA (actually, more like 9mA max).
thomas: I guess I'm not sure what you are asking? Are you asking if this stage could be used with an output transformer? I suppose it could be... you would probably want to omit the DC servo, etc. in that case, and possibly put an output capacitor in if the transformer is sensitve to DC.
You are lucky you don't have to try to understand my (nonexistant) Chinese... your English is quite good compared to that.
hi riotubes,
do U hd interest to try the final version of DAC. they use project D-1 to test & was only 5% different between project D-.
if u like to try. give me your address & I will free to provide one set to to try.
pars,
I understand wha U mean. Yes, it must rating several ma.
this is my prepare 1:10 nano amorphous IV step up transformers.
can use in 1541a DAC.
take a look first. it still under testing!
thomas: I guess I'm not sure what you are asking?
Do U see Audionote DAC 5?, it se transformers couple for IV. most people use sowter DAC transformers( IV ) & some said was very god & some was said not good. I thought reason was cause by sowter IV trans can onl ratin very low ma or cannot rating an DC. I cause saturate of the core & performance was not good. but I am not sure.
Are you asking if this stage could be used with an output transformer?
yes, 1541 use trans couple for IV. I suppose it could be... you would probably want to omit the DC servo, etc.
yes, 1541a was quite hot & rating higher ma than nowsdays chips, t must desig higher ma passing in OPT.
in that case, and possibly put an output capacitor in if the transformer is sensitve to DC.
yes, U can use this method, but I was worry performance was block or said Bottle neck was capactor.
You are lucky you don't have to try to understand my (nonexistant) Chinese... your English is quite good compared to that. many thanks, nice to discuss to U too.
thx
thomas
do U hd interest to try the final version of DAC. they use project D-1 to test & was only 5% different between project D-.
if u like to try. give me your address & I will free to provide one set to to try.
pars,
I understand wha U mean. Yes, it must rating several ma.
this is my prepare 1:10 nano amorphous IV step up transformers.
can use in 1541a DAC.
take a look first. it still under testing!
thomas: I guess I'm not sure what you are asking?
Do U see Audionote DAC 5?, it se transformers couple for IV. most people use sowter DAC transformers( IV ) & some said was very god & some was said not good. I thought reason was cause by sowter IV trans can onl ratin very low ma or cannot rating an DC. I cause saturate of the core & performance was not good. but I am not sure.
Are you asking if this stage could be used with an output transformer?
yes, 1541 use trans couple for IV. I suppose it could be... you would probably want to omit the DC servo, etc.
yes, 1541a was quite hot & rating higher ma than nowsdays chips, t must desig higher ma passing in OPT.
in that case, and possibly put an output capacitor in if the transformer is sensitve to DC.
yes, U can use this method, but I was worry performance was block or said Bottle neck was capactor.
You are lucky you don't have to try to understand my (nonexistant) Chinese... your English is quite good compared to that. many thanks, nice to discuss to U too.
thx
thomas
An externally hosted image should be here but it was not working when we last tested it.
hi pars,
I trust Sowter 's quality. probably your method ( add one small cap in front of the transformer)
pls remember this transformers must shield to lower noise!.
U can considerate use this, this sowter was my reference & compare material.
thx
thomas
I trust Sowter 's quality. probably your method ( add one small cap in front of the transformer)
pls remember this transformers must shield to lower noise!.
U can considerate use this, this sowter was my reference & compare material.
thx
thomas
An externally hosted image should be here but it was not working when we last tested it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.