We are finding that having a decent pi filter with COG caps instead of X7R on the input to the clocks on our boards helps reduce EMC noise! Still testing and looking at what is happening, but it seems that the COG's are better at handling RF frequencies.
When I get chance I'll dig all my links for ferittes as we have been adding them to everything recently, mainly to stop leads connected to the equipement becoming antennas.
When I get chance I'll dig all my links for ferittes as we have been adding them to everything recently, mainly to stop leads connected to the equipement becoming antennas.
Transformer upgrade
After months of procrastination, I finally took your advice and upgraded the audio transformer. Wow! Who would have guessed. Thank you
In my case I left the old one so I could do A/B tests of the two. No contest. The new transformer did several things. Everything became distinct and natural. Often you find a vail removed. Its a little of that, but where there was a muddling of everything into one soup of a sound, with the new transformer, everything became separate, distinct and yet in a relaxed natural way. Also brought our both ends of the spectrum. More of that nice tight bass and much more treble. Symbols, percussion become very natural and distinct. The sound stage more 3d. More emotion. Less hifi, more being there feeling the emotion. In well recorded Jazz, a lot of soft symbols just appeared out of the mush that had been there before.
The transformer I used was a Plitron. Luckily, my local electronics surplus place has a big box of custom order Plitron torids that are being sold off for $7. They have several taps, but one was made to measure, a 42vct. Thats 21-0-21 to replace the 23-0-23.
The only place I had room left in the box was a spot that I had an EI transformer for the 120v anode of my tube stage. I pulled out the EI and put in its place 2x the plitron. I connected an 18v output tap from the first one that powers the audio board and connected it to the same 18v tap of the second Plitron. The primarys of the second one now pumped out 125v which was perfect for my 6n2p triodes. Not the same jump in sound I found on the audio board, but not insignificant! So out came the EI and in went two Plitron toroids stacked on top of each other.
I can imagine further improvements following your suggestions on diode and regulator upgrades. That will have to wait. I am listening to Patricia Barber's A Distortion of Love as I write. I will need to rediscover my entire music collection before I go back in to the Arcam.
Thanks again Simon for your lead.
After months of procrastination, I finally took your advice and upgraded the audio transformer. Wow! Who would have guessed. Thank you
In my case I left the old one so I could do A/B tests of the two. No contest. The new transformer did several things. Everything became distinct and natural. Often you find a vail removed. Its a little of that, but where there was a muddling of everything into one soup of a sound, with the new transformer, everything became separate, distinct and yet in a relaxed natural way. Also brought our both ends of the spectrum. More of that nice tight bass and much more treble. Symbols, percussion become very natural and distinct. The sound stage more 3d. More emotion. Less hifi, more being there feeling the emotion. In well recorded Jazz, a lot of soft symbols just appeared out of the mush that had been there before.
The transformer I used was a Plitron. Luckily, my local electronics surplus place has a big box of custom order Plitron torids that are being sold off for $7. They have several taps, but one was made to measure, a 42vct. Thats 21-0-21 to replace the 23-0-23.
The only place I had room left in the box was a spot that I had an EI transformer for the 120v anode of my tube stage. I pulled out the EI and put in its place 2x the plitron. I connected an 18v output tap from the first one that powers the audio board and connected it to the same 18v tap of the second Plitron. The primarys of the second one now pumped out 125v which was perfect for my 6n2p triodes. Not the same jump in sound I found on the audio board, but not insignificant! So out came the EI and in went two Plitron toroids stacked on top of each other.
I can imagine further improvements following your suggestions on diode and regulator upgrades. That will have to wait. I am listening to Patricia Barber's A Distortion of Love as I write. I will need to rediscover my entire music collection before I go back in to the Arcam.
Thanks again Simon for your lead.
Hi Simon
Since we first spoke on this thread my Alpha has become ever so slightly Simonesque ( what ? ) and has evolved slowly into a major contributor to my now quite nice system.
So much so that the only major steps I can now take will be either very dangerous to the machines health ( or mine ) or.... be very costly.
In an attempt to avoid the latter ( or the former for that matter ) I'd like to discuss with you the merits of some small changes that I've either stupidly overlooked or just never noticed until recently :
470 pf polystyrene on the DEM as you did ( C223 ) to replace the 680 pf already installed - good idea ?
Infact - can I stuff the board with polystyrene caps to replace most of the standard issue EVOX pfr's fitted ?
The lytics in the output stage and before rca's - you linked them out - if I remove them and link out can I just put 3.3 uf or higher film caps across RCA out +/v and -v ( there's no room to put them anywhere else as you know )
On your DAC board - there a small piece of black wire connected to the -v point of a removed cap ( C225 ) and connected to the other side of what appears to be a ( now removed ) tubby inductor ( L201 ) This is all linked to the LM337 on the -5 and -15 rail. Safe to do this ? Or was it part of something related to your D1's install ?
The plan is to keep on with it culminating in a direct DAC connection to a 6N6P buffr ( that I have already ) and additional separate ps to 7310 and RAM - instead of the separate regs you initially instructed me to do.
I have not even touched the servo side and right from the start you were not sure if this reaped big enough gains to dabble with it.
Is this still your opinion ?
Or eventually do I need to do it for small gains ?
In the pic, the Aerovox large- ish black cap ( 10,000 uf / 63 ) sticking out the back replaces the on board 6800 and is bypassed with a green Vishay 10 uf )
The bit sticking out next to it is a fixed alu case with a traff / bridge and
large Phillips blue cap all dedicated to the SAA7220.
The aluminium heatsink holds ( crap ) std 7805's which in turn feed 7310, ram and 7220.
Have some teddy regs for 7310 and RAM to go in and a possible discreet ps board for the 7220.
The clock is a mediocre ( but surprisingly very flexible and GOOD ) Valab which I thoroughly endorse for budget modders like me.
( 11.289 - 5.6 and 2.8 )
As is - it stuns me - effortless, liquid and all over the wall and into the room.
As you reported with your more advanced Alpha female vocals and close mic'd stuff is hair raising at times.
So, what happens when I start with super regs ( one at a time
) from Rowmeister on the DAC board ?
It's clearly a ' no brainer ' from reports read so I'll need to eventually do it.
Just need a bit of convincing I guess.
I'm hoping it'll lift it quite significantly - like the full clock job ( 7220,7310 and pin 2/4 of the dac - which was staggering incidentally - thank you )
Any other secrets to offer with HD pics ?
In fact.... just ALL HD pics would do !!
As a close, and thanks to you, I have a very nice CD Player.
I also know my way around the service manual and recognise tracks and components like a true engineer ( all complete boll...s )
Many thanks
Andrew
Since we first spoke on this thread my Alpha has become ever so slightly Simonesque ( what ? ) and has evolved slowly into a major contributor to my now quite nice system.
So much so that the only major steps I can now take will be either very dangerous to the machines health ( or mine ) or.... be very costly.
In an attempt to avoid the latter ( or the former for that matter ) I'd like to discuss with you the merits of some small changes that I've either stupidly overlooked or just never noticed until recently :
470 pf polystyrene on the DEM as you did ( C223 ) to replace the 680 pf already installed - good idea ?
Infact - can I stuff the board with polystyrene caps to replace most of the standard issue EVOX pfr's fitted ?
The lytics in the output stage and before rca's - you linked them out - if I remove them and link out can I just put 3.3 uf or higher film caps across RCA out +/v and -v ( there's no room to put them anywhere else as you know )
On your DAC board - there a small piece of black wire connected to the -v point of a removed cap ( C225 ) and connected to the other side of what appears to be a ( now removed ) tubby inductor ( L201 ) This is all linked to the LM337 on the -5 and -15 rail. Safe to do this ? Or was it part of something related to your D1's install ?
The plan is to keep on with it culminating in a direct DAC connection to a 6N6P buffr ( that I have already ) and additional separate ps to 7310 and RAM - instead of the separate regs you initially instructed me to do.
I have not even touched the servo side and right from the start you were not sure if this reaped big enough gains to dabble with it.
Is this still your opinion ?
Or eventually do I need to do it for small gains ?
In the pic, the Aerovox large- ish black cap ( 10,000 uf / 63 ) sticking out the back replaces the on board 6800 and is bypassed with a green Vishay 10 uf )
The bit sticking out next to it is a fixed alu case with a traff / bridge and
large Phillips blue cap all dedicated to the SAA7220.
The aluminium heatsink holds ( crap ) std 7805's which in turn feed 7310, ram and 7220.
Have some teddy regs for 7310 and RAM to go in and a possible discreet ps board for the 7220.
The clock is a mediocre ( but surprisingly very flexible and GOOD ) Valab which I thoroughly endorse for budget modders like me.
( 11.289 - 5.6 and 2.8 )
As is - it stuns me - effortless, liquid and all over the wall and into the room.
As you reported with your more advanced Alpha female vocals and close mic'd stuff is hair raising at times.
So, what happens when I start with super regs ( one at a time
) from Rowmeister on the DAC board ?It's clearly a ' no brainer ' from reports read so I'll need to eventually do it.
Just need a bit of convincing I guess.
I'm hoping it'll lift it quite significantly - like the full clock job ( 7220,7310 and pin 2/4 of the dac - which was staggering incidentally - thank you )
Any other secrets to offer with HD pics ?
In fact.... just ALL HD pics would do !!
As a close, and thanks to you, I have a very nice CD Player.
I also know my way around the service manual and recognise tracks and components like a true engineer ( all complete boll...s )
Many thanks
Andrew
Attachments
Hi Andrew,
Yes, go for it, that won't hurt and probably gives a small gain.
Given polystyrenes usually only come in small values or are very chunky I'm not sure they're the best decoupling cap around. PPS may be a good substitute for the decoupling caps. If you're talking about replacing the filter caps in the original output stage then you can't go wrong with polystyrene.
I have to say I'm not 100% sure what you mean by that. If you mean film caps as DC blockers then yes, 3.3uF films should do the trick. Only issue is maybe some bass roll-off due to the low value.
Ah, that. I've had to get my photos and the datasheet open to see what's goin' on! What I've done is put a larger cap than wanted to fit decoupling the -15V rail. That meant removing the series coil (L201) and using the space of both C224 and C225. As I'm no longer sharing the -15V rail with the op-amps (and the D1 has its own +-30V supplies) the coil is not needed for isolation purposes.
After huge success regulating the servo driver op-amps in the CD63 (others report the same success) I thought the same would work well in the Alpha. Apparently not, I heard no real changes in doing this mod, so I don't recommend doing it until you run out of obviously-good changes.
In the pic, the Aerovox large- ish black cap ( 10,000 uf / 63 ) sticking out the back replaces the on board 6800 and is bypassed with a green Vishay 10 uf )
The bit sticking out next to it is a fixed alu case with a traff / bridge and
large Phillips blue cap all dedicated to the SAA7220.
The aluminium heatsink holds ( crap ) std 7805's which in turn feed 7310, ram and 7220.
Have some teddy regs for 7310 and RAM to go in and a possible discreet ps board for the 7220.
The clock is a mediocre ( but surprisingly very flexible and GOOD ) Valab which I thoroughly endorse for budget modders like me.
( 11.289 - 5.6 and 2.8 )
As is - it stuns me - effortless, liquid and all over the wall and into the room.
As you reported with your more advanced Alpha female vocals and close mic'd stuff is hair raising at times.
So, what happens when I start with super regs ( one at a time
It's clearly a ' no brainer ' from reports read so I'll need to eventually do it.
Just need a bit of convincing I guess.
I'm hoping it'll lift it quite significantly - like the full clock job ( 7220,7310 and pin 2/4 of the dac - which was staggering incidentally - thank you )
Any other secrets to offer with HD pics ?
In fact.... just ALL HD pics would do !!
As a close, and thanks to you, I have a very nice CD Player.
I also know my way around the service manual and recognise tracks and components like a true engineer ( all complete boll...s )
Many thanks
Andrew
Your mods look good and are obviously doing the job! Hanging stuff off the back sure makes more space, just don't drop it.
S Powers (or similar) will bring out more detail, smoothness, sweetness and openness. Eventually you'll start to hear diminishing returns but it's amazing what is available from CD if you put in the effort.
There are probably more caps you can upgrade, including the MSB and that tantalum cap next to it on the DAC filter. Small tweaks but might as well as it's easy to do.
Copper foil shielding on chips may make it sound a tad smoother but leave it till the last mod as they can get in your way and make it take longer to realise what chip is what!
Simon
Are your DAC supply smoothing caps standard? If so they'll want renewing. I'm actually about to do this on mine. I put 5 naff 1000u caps in as well as one Black Gate. I'm going to replace them with 2 x 22,000uF Fidelity Audio caps, which are the same as Mundorf M-lytic. Fitting will be interesting, may involve a drill. The DAC deserves a good supply though, and it already has that nice transformer and schottky diodes. I may at some point change those for HEXfreds as I think they may sound better, not sure. Schottkies may be good for digital, probably less so (bright and sterile) for analogue applications.
RE 6N6P OUTPUT
I have had good results by eliminating the entire stock output. I take left an right channel directly from the tda.. 24guage twisted pair straight to the pins of a 6n2p output section. The output from the tubes coupled via 4uf Obligatto PIO caps to new RCA. I tried 1uf Mundorf film caps, the PIO was a big step up low end and rich clear texture. Most recently I have made another big step up by bypassing the PIO with .1uf Russian teflon. To summarize, the big jumps in order on output were:
bypass all stock output with a 6n1p tube output,
switch to 6n2p tubes
remove all bits and pieces (caps, resitors and opamps) from the PC between the tda and rca. (75% of the board is now empty)
upgrade film to PIO
add teflon
I follow the Lampizator approach to output. He would claim you 6np6 will not have enough gain for the tda. Not sure if he is right. I have a sony cdp that I use for experimenting with a current out dac as well. Tube rolling on it recently between 6n1p, 6n6p and 6n2p all worked well. It is no contest that the 6n1p is the loser. The 6n6p is lovely (takes a ton of power for heaters). Its really a matter of preference between it and the 6n2p.
The russian military caps really are a deal for output coupling.
If in the mean time you are talking about the filter caps between the + and - on the stock output RCA, just snip them off and throw them away. You will find a wonderful improvement with no cap. Then run toward getting rid of all the stock ouput IMHO
I have had good results by eliminating the entire stock output. I take left an right channel directly from the tda.. 24guage twisted pair straight to the pins of a 6n2p output section. The output from the tubes coupled via 4uf Obligatto PIO caps to new RCA. I tried 1uf Mundorf film caps, the PIO was a big step up low end and rich clear texture. Most recently I have made another big step up by bypassing the PIO with .1uf Russian teflon. To summarize, the big jumps in order on output were:
bypass all stock output with a 6n1p tube output,
switch to 6n2p tubes
remove all bits and pieces (caps, resitors and opamps) from the PC between the tda and rca. (75% of the board is now empty)
upgrade film to PIO
add teflon
I follow the Lampizator approach to output. He would claim you 6np6 will not have enough gain for the tda. Not sure if he is right. I have a sony cdp that I use for experimenting with a current out dac as well. Tube rolling on it recently between 6n1p, 6n6p and 6n2p all worked well. It is no contest that the 6n1p is the loser. The 6n6p is lovely (takes a ton of power for heaters). Its really a matter of preference between it and the 6n2p.
The russian military caps really are a deal for output coupling.
If in the mean time you are talking about the filter caps between the + and - on the stock output RCA, just snip them off and throw them away. You will find a wonderful improvement with no cap. Then run toward getting rid of all the stock ouput IMHO
Simon Hi
Thank you for taking the time to reply to all that - I'll try and be more precise / shorter for future - apologies.
I've changed the MSB's to .47uf's which improved things a bit - maybe a 1 uf in the future.
I did change the 6 DAC supply caps to Elna's but your idea sounds MUCH better.
Wire links connecting pos and neg's together on the DAC board with flying leads to the new big caps - is that how you'll do it ?
Mounting them outside adjacent to the RCA sockets would be OK with short wires to the DAC board - my largish cap for the main ps works fine this way...well - it works !
Get your drill in there right away and report back on the gains asap
470 pf Polystyrene will be in tonight - and also thanks for saving me lots of time re Servo mods
The output DC blocking caps issue could be irrelevant if I can find the confidence to apply the valve buffer - still toying over passive i/v and which value resistor to use - this whole valve thing makes me a bit nervous tbh
Thanks again
Andrew
Thank you for taking the time to reply to all that - I'll try and be more precise / shorter for future - apologies.
I've changed the MSB's to .47uf's which improved things a bit - maybe a 1 uf in the future.
I did change the 6 DAC supply caps to Elna's but your idea sounds MUCH better.
Wire links connecting pos and neg's together on the DAC board with flying leads to the new big caps - is that how you'll do it ?
Mounting them outside adjacent to the RCA sockets would be OK with short wires to the DAC board - my largish cap for the main ps works fine this way...well - it works !
Get your drill in there right away and report back on the gains asap
470 pf Polystyrene will be in tonight - and also thanks for saving me lots of time re Servo mods
The output DC blocking caps issue could be irrelevant if I can find the confidence to apply the valve buffer - still toying over passive i/v and which value resistor to use - this whole valve thing makes me a bit nervous tbh
Thanks again
Andrew
Attachments
wlowes hi
Thanks for this.
Do you mean 24 gauge twisted wire straight to the pins of the 6n2p ' INPUT ' section ? ( is ' output ' a typo ? )
NO passive i/v in between DAC and buffer at all ?
Andrew
Ok I've just had a look and the "Si" caps are way too big to fit the space available. One of them will have to be mounted more elaborately with long wires and the other going in where the old caps were. I've taken a pic of how that might look - I needed a careful test-fit to be sure the board could still go in and out and the lid fit on. It should all work out ok - just!
Attachments
Yes, that is a typo. The Lampizator guide is here:
teachings of Prometheus
He describes the circuit as:
"Lampizator is an anode follower single ended triode in pure class A with a dynamic active load. It is called SRPP circuit or totem pole. "
Attachments
Removing stock output
I thought a picture would help describe what I mean by removing the stock output.
The first pic is with the output gone. I pick up the signal just before the former location of the first opamp. I have blackgates on the power supply and oscons everywhere else.
Second pic added more blackgates where the output usesd to reside.
I thought a picture would help describe what I mean by removing the stock output.
The first pic is with the output gone. I pick up the signal just before the former location of the first opamp. I have blackgates on the power supply and oscons everywhere else.
Second pic added more blackgates where the output usesd to reside.
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.