You will not hear any difference before you have sorted out the analogue output stage.
Replace the opamps and capacitors or introduce DOS-output stage...
Guys, it is mentioned several times that one should remove dc blocking caps and short the spaces with a piece of wire but there are no before and after photos for members that don't read schematics. Does anyone that has done this mod have photos that they can post? I know this is an old thread but I think it would be helpful.
You know what? That is probably the very worst advice you can give. I have held my piece, but understand the following.
1. A manufacturer will never, ever install a part that is not required. Please consider this.
2. Coupling capacitors are always sized well below the lowest frequency that matters. Translation: You don't hear them unless they are defective or horrible quality.
Before recommending modifications, why not actually study capacitors (or whatever) and audio. Most advice is well intentioned but not based in fact at all. Listening tests are horribly flawed and more influenced by what is between your ears than in fact. In other words, you will hear whatever it is you are presupposed to hear.
I actually do upgrades that make real differences. I measure and listen before and after, and my gear will easily measure a -135dB noise floor below -10 dBV. A human is hard pressed to hear much below -80dB, proved fact. I'm looking at a spectrum, not a meter pointer with an instrument, not a sound card. My experience exceeds 45 years. I'm just trying to help and interject some sanity here.
Carefully consider the fact that your soldering iron, depending on type, can inject currents into he circuit that can damage some components even powered off due to AC leakage, or static discharge from you. The components may still operate, but you may compromise noise performance. It is also extremely easy to end up with reduced performance with changes.
Upgrading equipment is an engineering thing. "Better" parts may not be stable in that particular circuit and may oscillate well above your hearing, or even above 1 MHz without your knowledge. Easy to see if you have the right instruments, experience and knowledge. You should understand the characteristics of different component types before sticking the the "best" (best for what job??). I do see a steady stream of hacked equipment (upgraded) where people have put in significant effort or paid a lot of money to a "golden ears expert" where performance is below that of a stock unit. The internet is typically rammed full of incorrect information and it often takes an expert to sort that stuff out. Really you are choosing on what the best story is to you. Sorry, it is a science and measurements do fully describe how something will sound, and we routinely and easily measure well past human limits of perception. It ain't the 70's or 80's anymore, we really do know what we are doing. If you or any other human can hear it, it is easily measured with a few different tests with the right equipment and skilled technical person.
Technical things always look easy to people who don't know the subject, just like anything else. But you know what? That professional (in any field) is worth their weight in gold. Why is it that in audio, people with a desire only are thought to know more than a trained person with the equipment and experience? Does that make any sense to anyone if you think about it?
-Chris
1. A manufacturer will never, ever install a part that is not required. Please consider this.
2. Coupling capacitors are always sized well below the lowest frequency that matters. Translation: You don't hear them unless they are defective or horrible quality.
Before recommending modifications, why not actually study capacitors (or whatever) and audio. Most advice is well intentioned but not based in fact at all. Listening tests are horribly flawed and more influenced by what is between your ears than in fact. In other words, you will hear whatever it is you are presupposed to hear.
I actually do upgrades that make real differences. I measure and listen before and after, and my gear will easily measure a -135dB noise floor below -10 dBV. A human is hard pressed to hear much below -80dB, proved fact. I'm looking at a spectrum, not a meter pointer with an instrument, not a sound card. My experience exceeds 45 years. I'm just trying to help and interject some sanity here.
Carefully consider the fact that your soldering iron, depending on type, can inject currents into he circuit that can damage some components even powered off due to AC leakage, or static discharge from you. The components may still operate, but you may compromise noise performance. It is also extremely easy to end up with reduced performance with changes.
Upgrading equipment is an engineering thing. "Better" parts may not be stable in that particular circuit and may oscillate well above your hearing, or even above 1 MHz without your knowledge. Easy to see if you have the right instruments, experience and knowledge. You should understand the characteristics of different component types before sticking the the "best" (best for what job??). I do see a steady stream of hacked equipment (upgraded) where people have put in significant effort or paid a lot of money to a "golden ears expert" where performance is below that of a stock unit. The internet is typically rammed full of incorrect information and it often takes an expert to sort that stuff out. Really you are choosing on what the best story is to you. Sorry, it is a science and measurements do fully describe how something will sound, and we routinely and easily measure well past human limits of perception. It ain't the 70's or 80's anymore, we really do know what we are doing. If you or any other human can hear it, it is easily measured with a few different tests with the right equipment and skilled technical person.
Technical things always look easy to people who don't know the subject, just like anything else. But you know what? That professional (in any field) is worth their weight in gold. Why is it that in audio, people with a desire only are thought to know more than a trained person with the equipment and experience? Does that make any sense to anyone if you think about it?
-Chris
I’d just got some details together on how to do it when this post popped up - an interesting read and got me into some further reading from there.
I’d always just taken it as a given that “better” caps improve sound (or rather, degrade sound less) especially in the signal path.
Actually, not always true. If there is no voltage developed across the capacitor, and by definition a coupling capacitor does not have any significant signal voltage across it, it doesn't matter much. If you take an extreme, like a ceramic Z5U type and replace it with an electrolytic , it might sound better. Maybe, and that is an extreme case.
Also, what cap is "better"? Each capacitor type is designed for various jobs and conditions of operation. The most important thing is that the new part fits correctly. There is no difference between various good quality brands of capacitors, sorry but this is factually true.
People "improving" equipment aren't even looking at the correct parts most of the time. Then they do not have the experience / knowledge to even select an appropriate part. This is true of resistors, capacitors and op amps. Even PCB layout matters. When you have discrete transistors, other factors come into play. Then consider workmanship. What is the very most expensive component? How about the circuit board, and this is what I see damaged most often.
I will say that anyone who actually does know exactly what they are doing are not writing about it (very few anyway). They are too busy for one. For two that knowledge is hard won and they charge for it (like me). They also constantly upgrade their knowledge and instrumentation. You do not want to know how much $$ is on my bench, but I will say much is current and under warranty, made by Keysight. Other equipment is specialised, made by various manufacturers, plus really specialised test jigs made by each person.
I return to the fact that most "upgrade" folks are not even talking about the correct parts. If you look at the levels of performance you are actually trying to achieve, don't you think that you really could use a trained professional? Then your equipment isn't damaged by the work if nothing else.
Just throwing this out there. Take it or leave it, but at the minimum, think about it clearly and come to a decision.
Also, what cap is "better"? Each capacitor type is designed for various jobs and conditions of operation. The most important thing is that the new part fits correctly. There is no difference between various good quality brands of capacitors, sorry but this is factually true.
People "improving" equipment aren't even looking at the correct parts most of the time. Then they do not have the experience / knowledge to even select an appropriate part. This is true of resistors, capacitors and op amps. Even PCB layout matters. When you have discrete transistors, other factors come into play. Then consider workmanship. What is the very most expensive component? How about the circuit board, and this is what I see damaged most often.
I will say that anyone who actually does know exactly what they are doing are not writing about it (very few anyway). They are too busy for one. For two that knowledge is hard won and they charge for it (like me). They also constantly upgrade their knowledge and instrumentation. You do not want to know how much $$ is on my bench, but I will say much is current and under warranty, made by Keysight. Other equipment is specialised, made by various manufacturers, plus really specialised test jigs made by each person.
I return to the fact that most "upgrade" folks are not even talking about the correct parts. If you look at the levels of performance you are actually trying to achieve, don't you think that you really could use a trained professional? Then your equipment isn't damaged by the work if nothing else.
Just throwing this out there. Take it or leave it, but at the minimum, think about it clearly and come to a decision.
That’s what I mean - I had always just made assumptions based on “listening tests” and will be more conscious of the engineering going forward.
Sure a more suitable cap in the right place can make a difference and all the ones in my gear are carefully chosen based on application so coupling caps are an interesting one
Required by who? Design engineers, legal department, accounting department?
It seems to me that you are completely in a wrong thread - these 21k posts are "all about" chopping, slicing and dicing and these posters cannot be in corpore wrong.
You should look at Tony Gee's "Humble Homemade HiFi" at his cap comparisons... I have tried many, many of them and found his site to be pretty much spot on. The Vishay 1837 fit under the board and you can bypass the output caps. To my ear the Silmic lack a little in the high register and these do the trick!
You are missing the point - the best capacitor is missing (coupling) capacitor
In case you intend to start arguing - no use in my case, I have done my mods on CD-67 (some posts also in this thread but mostly thankful reading) and would never listen to a stock unit ever again...
Last edited:
Just my 2 Cents: I have done a lot of CD67 Mods in the Last 10 Years. Also removing the blocking caps.
Correct me if I'm wrong - but AFAIK those are intended to prevent having DC on the Audio Signal. If it would be passed through the whole chain the Coil of the Speaker might get burned (among other issues with the Amp itself). Does it increase the soundquality in combination with all the other tweaks? mostlikely yes.
First of all. Companies are lazy because sometimes adding one part can save money because you might be able to choose 10 cheaper parts at other locations instead e.g. less precise Resistors or less well matched trainsistor pairs.... Or its just getting rid of one QA check. You drew a correct picture of what is important to be considered in the design.
In this case I would assume it is the "insurance" for the company that even if some parts are not perfectly fine in the signal chain causing some DC it will not leave the CD Player.
Enthusiastic Hobbiests like us are doing more quality checks then the company it self.
Just simply by checking if the device still works after mods with internal crosschecks at logically given "Testpoints".
I'm not sure if any CDP with blocking Caps will have ever seen a Multimeter at the RCA's to check if there is DC. Mostlikely I have done it dozen of times which is exactly a dozen times more then at the factory.
Throwing in some DC blocking caps for a few cents is way cheaper then check every device if DC is in an acceptable range.
One thing to add. The mods are intended to push the device to its full potential without having the cost cutting measures of companies in mind. Which almost mostly means that the mods recommend the use of parts with very tight tolerances. Which does also reduce the potential amount of DC due to the reduction of imbalanced circuits.
But at the end I have to admit that throwing out parts just because some people claim it sounds better should not be done without the consiousness of knowing potential consequences. If this educated approach is ignored the caused failures lead to a steper learning curve
p.s.: my Amp has input caps so in the case that there is some DC after years of operation it will simply not pass.
Correct me if I'm wrong - but AFAIK those are intended to prevent having DC on the Audio Signal. If it would be passed through the whole chain the Coil of the Speaker might get burned (among other issues with the Amp itself). Does it increase the soundquality in combination with all the other tweaks? mostlikely yes.
First of all. Companies are lazy because sometimes adding one part can save money because you might be able to choose 10 cheaper parts at other locations instead e.g. less precise Resistors or less well matched trainsistor pairs.... Or its just getting rid of one QA check. You drew a correct picture of what is important to be considered in the design.
In this case I would assume it is the "insurance" for the company that even if some parts are not perfectly fine in the signal chain causing some DC it will not leave the CD Player.
Enthusiastic Hobbiests like us are doing more quality checks then the company it self.
Just simply by checking if the device still works after mods with internal crosschecks at logically given "Testpoints".
I'm not sure if any CDP with blocking Caps will have ever seen a Multimeter at the RCA's to check if there is DC. Mostlikely I have done it dozen of times which is exactly a dozen times more then at the factory.
Throwing in some DC blocking caps for a few cents is way cheaper then check every device if DC is in an acceptable range.
One thing to add. The mods are intended to push the device to its full potential without having the cost cutting measures of companies in mind. Which almost mostly means that the mods recommend the use of parts with very tight tolerances. Which does also reduce the potential amount of DC due to the reduction of imbalanced circuits.
But at the end I have to admit that throwing out parts just because some people claim it sounds better should not be done without the consiousness of knowing potential consequences. If this educated approach is ignored the caused failures lead to a steper learning curve
p.s.: my Amp has input caps so in the case that there is some DC after years of operation it will simply not pass.
Yes I did - years ago. The type of Transformer might not change the sound much but If you can find a custom made one with separated windings for the major circuits you will have a cleaner separation of the individual subcircuits. In my case it has improved the sound quite a lot. So it happend to be a toroid but the "effect" can be pinned down to the separation.
Last edited:
I have not used the Rayregulators but some compareabale LM317 based regulators. Cant recall all details but it had caps at the set pin to reduce noise, Impedance, PSRR.
Comparebale to the idea of the rayreg but without gyrator involved.
To be honest I got way better results with diy made LT3045 based regulators. Especialy at the DAC. Lowering the Noisefloor around the DAC is very beneficial for the sound. LT3042 and LT3045 are very nice devices for achieving that.
I have used the Reference design given in the Datasheet and the Application notes of the LT3045 (LT3042). LDOs can be quite tricky so reading the datasheet thoroughly is a must. Otherwise you might have some Oscillating and maybe overheating devices or they just do not reach the stellar performance. The LT30xx seem to be less tricky than the e.g. TI TPS7Axx ...at least for me.
AFAIK you can also find some prebuild ones on ebay where you have to adjust the voltage by simply soldering a bridge on a dedicated junction on the pcb. Some of them also resemble the recommended ref design including the pcb design style of the evaluation board..
Last edited:
concerning Scope - as a student I also started the hobby with a bad soldering iron and without scope....just my "golden ears". Not reliable I know. And from the scientific POV not even considerable.
Yes scope is the very best tool to have for this hobby along with a soldering iron. And a must have...
But as a broke student it took me quite a while to find a good old analog (correctly working!) scope below 100€. The beloved Tektronix 2225 is around 150-200€ (without probes) price has not changed much in 10years.
AFAIK in the US its cheaper but I think this is because the military had stockpiled them.
And after years those have flooded the market. And considering that they are almost indistructable. The same amount of devices are getting passed on from one generation to the next. So stable offer/demand equilibrium.
Doesnt count for the EU.
I was able to get the 2225 around 100€ with probes. Worth the investment and a real 1:100 bargain. The only alternative would have been an Philips Scope with 10 or 20Mhz Bandwith but with this you can't get the 16.933Mhz clocksignal rendered precisely enough to see if it is a square or sine. .
Yes scope is the very best tool to have for this hobby along with a soldering iron. And a must have...
But as a broke student it took me quite a while to find a good old analog (correctly working!) scope below 100€. The beloved Tektronix 2225 is around 150-200€ (without probes) price has not changed much in 10years.
AFAIK in the US its cheaper but I think this is because the military had stockpiled them.
And after years those have flooded the market. And considering that they are almost indistructable. The same amount of devices are getting passed on from one generation to the next. So stable offer/demand equilibrium.
Doesnt count for the EU.
I was able to get the 2225 around 100€ with probes. Worth the investment and a real 1:100 bargain. The only alternative would have been an Philips Scope with 10 or 20Mhz Bandwith but with this you can't get the 16.933Mhz clocksignal rendered precisely enough to see if it is a square or sine. .
If you look at the schematics of any integrated amplifier or preamp-poweramp combo then you will probably find at least half a dozen coupling capacitors in the signal path preventing DC reaching the speakers. I think that in worst case I counted more than dozen of these in one integrated amplifier.
And I do not remember a schematics without a capacitor in the power amplifier input.
So removing e.g. half of these (certainly not randomly
) can only be of benefit to the sound signal.Your "another source" is just too loud - volume pots are usually not too good (channel balance wise) in the 8...9 o'clock region.
You if you get desired sound level around 12-1 knob position then I would say that this is rather an advantage than a problem.
