Thanks Chris - yes I was initially reluctant to 'name names' - but I go by what is inside equipment, experience, and testing. There are compromises in all equipment, it is easy to improve equipment because designs are determined by profit margins, until the equipment ends up someplace of which that isn't a restriction, or less of one.
Obvious areas of upgrade in this amp are the ADC and DAR circuits, if I had more time (and if it was within the customer's budget) then I would be certainly designing daughterboards for both, which would by default upgrade the buffers feeding it that are the subject of this thread.
In the case of the DAR which isn't shown, the manufacturer has implemented a plug-in module, but has then subsequently regressed in the design for some reason, looking at the various circuit revisions. I reverse engineered some of the digital circuits because the module was different to the official schematics, have been comparing them and considering upgrade paths in that area, but we have had to draw a line now for the time being. It would make far more sense if the ADC was a 96kHz device, considering the rest of the unit is capable of that or higher.
Obvious areas of upgrade in this amp are the ADC and DAR circuits, if I had more time (and if it was within the customer's budget) then I would be certainly designing daughterboards for both, which would by default upgrade the buffers feeding it that are the subject of this thread.
In the case of the DAR which isn't shown, the manufacturer has implemented a plug-in module, but has then subsequently regressed in the design for some reason, looking at the various circuit revisions. I reverse engineered some of the digital circuits because the module was different to the official schematics, have been comparing them and considering upgrade paths in that area, but we have had to draw a line now for the time being. It would make far more sense if the ADC was a 96kHz device, considering the rest of the unit is capable of that or higher.
Hi Jim,
We think along the same lines. Scary!
There are several other industry professionals who are members here. You may even know some of them. I think you would add a lot for the membership here if you pop by more often. Additionally, there are service people learning their craft who could use another source of information on how the business should run, ethics and all that. Safety is another area of knowledge that simply isn't taught any longer. Most of the training you and I received is simply non-existent these days. A few comments from you might help in that area where needed.
As there are working designers who are members, you might pick up some additional training as well. There is nothing wrong with coming straight out with questions in your own thread. I'm sure we can give back over the long term.
Take care and thanks for posting.
-Chris
We think along the same lines. Scary!
There are several other industry professionals who are members here. You may even know some of them. I think you would add a lot for the membership here if you pop by more often. Additionally, there are service people learning their craft who could use another source of information on how the business should run, ethics and all that. Safety is another area of knowledge that simply isn't taught any longer. Most of the training you and I received is simply non-existent these days. A few comments from you might help in that area where needed.
As there are working designers who are members, you might pick up some additional training as well. There is nothing wrong with coming straight out with questions in your own thread. I'm sure we can give back over the long term.
Take care and thanks for posting.
-Chris
First of all, we have no evidence, as lease this thread of discussion has not produced any of it, indicating that either these caps are making troubles or they are the bottle neck of the performance by design in that piece of equipment. Although it is believed, we do not know replacing the caps will improve performance in what respect and by how much, let alone the possibility of making damage to performance by replacing the caps, until a set of measurement before and after can be made, which is unlikely. It is quite likely replacing the caps makes no practical appreciable difference in performance of the equipment. In such circumstances I would do the mod in a least damaging risking way, that is, enhancing instead of replacing.
Secondly, the e-cap and the SMD cap help each other when doubled-up. The e-cap helps lower the signal voltage at low frequencies making the SMD caps problem components no more had they ever been, the SMD cap helps with lower esr and lower inductance at high frequencies.
Thirdly, The SMD cap, with one of its ends soldered down intact, will also provide mechanical support to the other end being soldered - without it one would be soldering a leaded part onto a floating SMD pad that is easily damaged as the temperature is well above glass transition temperature of the resin the PCB is made of, and a SMD pad does no have a plated hole to help hold it in place.
By doubling up the caps the PCB pads experience at least one less heat cycle than would the case replacing them.
You are assuming I am not measuring the equipment. My original post shows one capacitor removed from the left channel, and the other kept in place on the right channel.
The reason for this was to firstly measure the capacitor's standalone performance out of circuit, and secondly afterwards compare measurements of many different types of capacitors in place of the original that has been removed.
Keeping the original reference in circuit on one channel makes testing quicker by simultaneous analysis of the two channels / stereo signal path, before reaching a conclusion about which type of capacitor would eventually be used on both channels during the upgrade.
Of course... as mentioned, this is a very small part of a complex repair job most of which is now complete, and it was only really out of curiosity and surprise that I originally posted.
However comparisons of various substitutes will be done. I never assume any circuit design cannot be improved and always look at options while repairing equipment, taking into consideration the equipment as a whole and where the bottlenecks are and the viability and cost effectiveness of any modifications that seem within reason, or research any that are requested by my clients.
However comparisons of various substitutes will be done. I never assume any circuit design cannot be improved and always look at options while repairing equipment, taking into consideration the equipment as a whole and where the bottlenecks are and the viability and cost effectiveness of any modifications that seem within reason, or research any that are requested by my clients.
Hi nattawa,
Jim is operating in an industry standard and responsible manner. I guess you're not in the industry and really are not aware of how things are normally done. However, you needn't worry about cowboy tactics as it was clear early on that Jim isn't one. Another name you might hear to describe unprofessional conduct with a soldering iron is "plumber". Both very derogatory terms when used to describe the qualities of a technician.
Jim mentioned that he is using equipment made by Audio Precision. Certainly equipment viewed as an industry standard, something I can't afford in addition to the equipment I already use. It's not as sensitive as Jim's, but far beyond what a typical technician uses, having industry accepted performance being HP / Agilent / Keysight. He also had explained that he did make measurements before, during and after each segment completed of the job. His (and my) experience has shown that high-K ceramic capacitors are not suitable choices for high performance equipment. Stated earlier.
Your assumptions were unfounded, as was the assumption that either Jim or I had suggested we would use a leaded component for the replacement. I intended the use of SMD type capacitors, the film types (which I normally stock).
Stacking SMD components is not supported by the industry or component manufacturers. Neither is using a surface mount part as an anchor for a leaded component. Before raising an objection, maybe read the posts over again to make sure you are on the right track - or ask. We're friendly and would answer your questions.
-Chris
Jim is operating in an industry standard and responsible manner. I guess you're not in the industry and really are not aware of how things are normally done. However, you needn't worry about cowboy tactics as it was clear early on that Jim isn't one. Another name you might hear to describe unprofessional conduct with a soldering iron is "plumber". Both very derogatory terms when used to describe the qualities of a technician.
Jim mentioned that he is using equipment made by Audio Precision. Certainly equipment viewed as an industry standard, something I can't afford in addition to the equipment I already use. It's not as sensitive as Jim's, but far beyond what a typical technician uses, having industry accepted performance being HP / Agilent / Keysight. He also had explained that he did make measurements before, during and after each segment completed of the job. His (and my) experience has shown that high-K ceramic capacitors are not suitable choices for high performance equipment. Stated earlier.
Your assumptions were unfounded, as was the assumption that either Jim or I had suggested we would use a leaded component for the replacement. I intended the use of SMD type capacitors, the film types (which I normally stock).
Stacking SMD components is not supported by the industry or component manufacturers. Neither is using a surface mount part as an anchor for a leaded component. Before raising an objection, maybe read the posts over again to make sure you are on the right track - or ask. We're friendly and would answer your questions.
-Chris
Hi Andrew,
I bought sm film capacitors in 805 and 1206. I heavily prefer the 1206 size for capacitors. But you are correct where 1uF and higher values are concerned.
-Chris
I bought sm film capacitors in 805 and 1206. I heavily prefer the 1206 size for capacitors. But you are correct where 1uF and higher values are concerned.
-Chris
Hi Guys,
I'll update this thread once the repair is complete, and post some test results for those who are interested. Several substitutes are being considered and components have been ordered, but I am currently working on other jobs in the meantime and will update with the final outcome of this section of the repair soon.
Thanks to all, Jim 🙂
I'll update this thread once the repair is complete, and post some test results for those who are interested. Several substitutes are being considered and components have been ordered, but I am currently working on other jobs in the meantime and will update with the final outcome of this section of the repair soon.
Thanks to all, Jim 🙂
Using a leaded component was not an assumption. When a film type SMD cap of a suitable capacitance and voltage rating that will fit into a 0805 footprint is nowhere to be found, a leaded component most probably is the only choice for the mod in question. You seemed to have never realized it until posts #28 and #29.
No one was talking about stacking SMD components either. Did it ever cross your mind that I could've suggested replace with a film type SMD cap of the same ratings that suits the available space and pad pattern?
Doubling up a SMD cap with a leaded cap in this particular case not only has its merit, but also seemingly is the only shot. When it is the only shot, then we do it not only because we can, but also because we should. Ask you wife if you don't believe me. 😀
Don't worry about the support of your industry. History can be made.
Hi nattawa,
FYI the two substitutes that have been order for testing are
Panasonic ECPU1C105MA5
Cornell Dubilier FCA1210C105M-G2
They are of package size 1210 of which there is physical clearance by original design for on the PCB. Their electrical specifications indicate they will significantly outperform the existing device, because they are designed for coupling applications whereby the original high-k ceramics are not. Maximum ratings such as voltage have obviously been also checked to be safely within the range that they will be operating.
I'll post my own test results too out of interest when the repair is being finalised.
Cheers, Jim
FYI the two substitutes that have been order for testing are
Panasonic ECPU1C105MA5
Cornell Dubilier FCA1210C105M-G2
They are of package size 1210 of which there is physical clearance by original design for on the PCB. Their electrical specifications indicate they will significantly outperform the existing device, because they are designed for coupling applications whereby the original high-k ceramics are not. Maximum ratings such as voltage have obviously been also checked to be safely within the range that they will be operating.
I'll post my own test results too out of interest when the repair is being finalised.
Cheers, Jim
Thanks for the update, Jim. It appears they over-sized the footprint enough in PCB design to accommodate a 1210. I think the soldering pads are undersized by quite a bit for 1210 parts and you wont be getting the best fillet at the components ends. But it's acceptable for a mod work as long as you can fit them in and have no problems getting the heat to the pads. I'll be interested in your test results, many will too I believe, as they may tell how far one can go with miss-using a part for what performance expectation.
While the original cap seems to be the 'wrong color' to be a C0G/NP0, I have to add that the modern C0G/NP0 caps using nickel electrodes (the 'base metal electrode' caps) are made with Calcium Zirconate, which is a completely non-polar dielectric with no piezoelectric effect. The consensus here seems to be that these are not good for audio, but honestly, they are an improvement compared to film caps with possibly faulty end terminations, which can and do cause measurable distortion. Murata, AVX and Kemet make these modern C0G caps, and the nice part is that they are cheaper than the older precious metal electrode C0G/NP0, which actually used a different and slightly worse dielectric. I'm using some Kemet 1812 sized 0.1µF caps because they're thinner than other 1206 caps, but there are a number of 0.1µF and a few 0.22*µF C0G ceramics, and you can always stack them for larger values. Not sure what value you need, but unlike any other SMD film cap, their dielectric is far more linear and there will be no possible microphonic effects. Remember, you can't get the 'nice' polymers in SMD because the reflow temperature is too high, so you're stuck with PPS, acrylic, PEN, and mylar, all of which are not as nice as Calcium Zirconate.
- Status
- Not open for further replies.