The larger surface mounted caps that I have seen swapped and changed to Oscon on the 4258 D>A Dac. Both larger caps near the DACs are filter to ref Grd. Super low ESR desired here ?
The 100uf is now an OSCon SEPc
The 100uf is now an OSCon SEPc
Interesting the Formulae for ESR bears no relationship to say the stated ESR for a Panasonic FM cap. I get the ESR is quoted at 120 Hz and the ESR at orders of magnitude higher but I still cant get this to get close ?
https://forum.digikey.com/t/calculating-capacitor-esr-from-tan/2633
https://forum.digikey.com/t/calculating-capacitor-esr-from-tan/2633
Loss tangent can be due to several mechanisms, not just ESR - for instance the dielectric losses can show up in the tangent, but doesn't limit the max current like ESR does.
I am still learning and fond this helpful and interesting. https://www.analog.com/media/en/training-seminars/tutorials/MT-101.pdf
Not given up yet and are using a home built spread sheet to calculate ESR, also wondered if anyone has used these and might be worth looking at ?
https://uk.farnell.com/peak-electro...dible-alert-capacitor/dp/2917878?st=esr meter
https://uk.farnell.com/peak-electro...dible-alert-capacitor/dp/2917878?st=esr meter
For those interested the Silmics are very good for ESR, a 470uf 35v Silmic II compared to a 470uf 35v Panasonic. Shows the Silmic is .03 Ohms, the Panasonic .04 Ohms Okay they are physically large but they measure better than a Panasonic FR.....
Might be useful for some a number if 470uf 35v caps, Original Panasonic fitted in 9000 is the worst for ESR and near bottom for spec
I chose the Silmic for how it sounded in comparison to others, interesting or coincidental it measured the best for ESR
I chose the Silmic for how it sounded in comparison to others, interesting or coincidental it measured the best for ESR
If you want low ESR, 40mΩ is easily beaten by ordinary size, industrial parts. being 10mΩ better than another, at some arbitrary and unknown frequency, with a 2 probe measurement (as opposed to the more suitable bridge/4 wire measurement) at the very end of the meters useful range, without taking inductance and other board parasitics into account, isnt all that useful on its own IMO. Its great you are using objective tools, but all you are doing is using them to justify a dubious choice IMO. Add in the long extra lead wire vs something that fits and see how much of your 10mΩ is left.
Also, just an example, the plain vanilla old Sanyo Oscon SEPC in 100uf/16V that you mention replacing above, has an ESR of 10mΩ, with pages and pages of similar size and cost capacitors around this number.
Also, just an example, the plain vanilla old Sanyo Oscon SEPC in 100uf/16V that you mention replacing above, has an ESR of 10mΩ, with pages and pages of similar size and cost capacitors around this number.
Last edited:
InspectorGadget I know you are keen to dismiss those with less electrical knowledge, I have no problem with humility. I have used SepC's in combination of lower value caps and listened as well as measuring. I have swapped changed and learnt, very happy with the results so far. I have and are still considering using a Merdidian HD621 to allow me to run LPCM into 4 stereo DACs. I had the opportunity to listen to the internal multichannel DAC after my novice work so far in comparison to a DacMagic and a R2R NOS Metrum Musette Dac using SPDF Digital out from my BR player. It was an experiment to see how much of an upgrade 4 external DACs would offer. All those who heard and listened including those that didn't know what was playing preferred the novice modded DAC. I am not saying that the mods have moved the DAC into the extraordinary but I am saying I have made improvements with limited knowledge and experimentation. That's what I am here for, I thought that was the point of DIY Audio ?
I'm not dismissing you, i'm dismissing Silmic II in this job in general, but especially if using 'low esr' to justify it, when the caps that were in there to begin with are several times better in that metric. Not really worth mentioning Silmic's 0.03Ω as a plus, when the cap its replacing is 0.01Ω. That isnt an improvement (not that low ESR is even required/desired here).
The rest of your post i'm afraid is not convincing. I'm genuinely glad you are happy, but your 'improvements' are anecdotal at best. I'm guessing the fact you hadnt mentioned the SEPC ESR being 10mΩ, means you hadnt done due diligence when replacing a part. You mention the ESR of other cheap caps from other brands being worse, but not the part you are attempting to match/improve on? There are 266 types of 100uf/12-50V capacitors less than 40mΩ just in the polymer/al-polymer hybrid section at digikey. The Nichicon FPCAP al/pol 100uf/16V is $1.36AUD in singles and is 7mΩ
The rest of your post i'm afraid is not convincing. I'm genuinely glad you are happy, but your 'improvements' are anecdotal at best. I'm guessing the fact you hadnt mentioned the SEPC ESR being 10mΩ, means you hadnt done due diligence when replacing a part. You mention the ESR of other cheap caps from other brands being worse, but not the part you are attempting to match/improve on? There are 266 types of 100uf/12-50V capacitors less than 40mΩ just in the polymer/al-polymer hybrid section at digikey. The Nichicon FPCAP al/pol 100uf/16V is $1.36AUD in singles and is 7mΩ
Last edited:
I did actually research and looked through Farnell and lots of posts and threads, I used information on sites like Authentic Cinema, conversations with a few people to understand in most locations what was important. I spent several hours reading the Panasonic brochures for FR,FM and the old FC an well as the OSCONS that are now Panasonic and no longer Sanyo branded. I looked at Kemet, Nichicon and others as well. I even built a calculator in excel to add two caps with known ESR, the calculations from tan dont seem to align if ESR isn't known. I did search and found one site where a few Silmics were measured as well. The reason I tried Oscons was based on several sites where caps had been experimented with, like Lampizator, I have also heard his production DACs which are excellent (not in my system though)
In the end I tried and listened and experimented. In a number of locations I now have Silmic and Oscons and in some just Oscons, also Panasonic FMs, most Nichicons have ended up being replaced interestingly. I have made mistakes and revered changes on several occasions.
In the end I tried and listened and experimented. In a number of locations I now have Silmic and Oscons and in some just Oscons, also Panasonic FMs, most Nichicons have ended up being replaced interestingly. I have made mistakes and revered changes on several occasions.
Last edited:
Hard to say, the parts are not those i'm familiar with, but going by the apparent vintage of the schematic, I would guess a large number of caps need replacing and will improve performance; if you really wish to conserve this unit.
What is it?
What is it?
SMG sounds like United Chemi-Con or Nippon Chemi-con (not the same thing as Nichicon) 85C 2000h rated for general purpose radial cap. Often there are date / lot codes on the caps as well to help determine the age, if the unit was purchase second-hand, possibly re-capped already, etc..
Sony MDS-JB930. Shouldn't there be low-impedance capacitances in the digital power supply of the DAC?
Not necessarily, no. Depending how good the layout is, you might want some damping, in the form of lossy caps.