View previous topic :: View next topic |
Author |
Message |
audiolabtower VIP Contributor 500+
Joined: 06 Jan 2009 Posts: 686
|
Posted: Tue Aug 28, 2012 2:33 pm Post subject: |
|
|
It looks about right apart from C5 which on later 105s should be 80 uF non polarised at least 50V working.
The series bass capacitor is made up to 480 uF on early 105 2s and I believe 450 uF on late 105 2s so I'm not sure if your 900uF made up of two 450s is absolutely correct. Maybe Guru can clarify? |
|
Back to top |
|
|
Dev Junior Contributor 20+
Joined: 17 Aug 2012 Posts: 24 Location: India
|
Posted: Fri Nov 16, 2012 7:30 pm Post subject: |
|
|
Quote: | speakerguru wrote: | Most multimeters will not measure capacitor loss factor. You need to get an LCR meter which has this feature. |
|
Hello , I have been away for long hence could not complete the cap measure. Finally, the results are : (All are blue ALCAP,+,-,10% printed )
original value measured value
450 MFD 479 MFD
450 MFD 476 MFD
450 MFD 468 MFD
450 MFD 481 MFD
30 MFD 30.3 MFD
30 MFD 30.17 MFD
10 MFD 10.6 MFD
10 MFD 10.5 MFD
3.3 MFD 3.61 MFD
3.3 MFD 3.40 MFD
3.3 MFD 3.38 MFD
3.3 MFD 4.07 MFD
All of them are on higher values, (there's a serial number on all of them 93-xxxx. Is this the year of manufacture?)
Now, the big question, shall I change these caps? If so, which ones to go for?
I am willing to spend $ 500 If needed. All the gurus' up there, please guide.
Dev |
|
Back to top |
|
|
speakerguru Über Contributor 1000+
Joined: 18 Nov 2005 Posts: 1192 Location: Green Hut, Tovil
|
Posted: Sat Nov 17, 2012 9:43 am Post subject: |
|
|
Where both sides are within 1% of each other and within +/-10% of their nominal values, I would leave them alone. Where they fall outside these limits, without the original factory build charts, you will have to decide whether to replace with nominal or some other values. In any case, I would recommend that maintaining good pair matching should be your primary aim. |
|
Back to top |
|
|
Dev Junior Contributor 20+
Joined: 17 Aug 2012 Posts: 24 Location: India
|
Posted: Sun Nov 18, 2012 7:03 pm Post subject: |
|
|
Quote: | speakerguru wrote: | Where both sides are within 1% of each other and within +/-10% of their nominal values, I would leave them alone. Where they fall outside these limits, without the original factory build charts, you will have to decide whether to replace with nominal or some other values. In any case, I would recommend that maintaining good pair matching should be your primary aim. |
|
Pardon my lack of knowledge but I have read of KEF using caps with 1%+or- tolerance value. But these caps have +,-,10% as printed on them ! They are also high in value, almost 10%.
Shall I carry on with them only & add the missing C5 or change all to polypropylene or other type which are better with sonic reproduction? |
|
Back to top |
|
|
speakerguru Über Contributor 1000+
Joined: 18 Nov 2005 Posts: 1192 Location: Green Hut, Tovil
|
Posted: Sun Nov 18, 2012 11:24 pm Post subject: |
|
|
KEF bought +/-10 or sometimes even 20% capacitors. They were all measured in the factory and sorted into 1% bands. For Reference series all the inductors were wound +/-1% to complement the stock and spread of capacitors. If the capacitors were high then the inductors would be wound low, etc. This together with drive unit matching, would keep all the critical crossover and eq frequencies within 1% and the frequency response within tight limits.
I have no idea how capacitors (and drive units) drift after all this time. That's why you'll have to make your own decision re absolute values. But you won't go far wrong if you major on keeping the pair matching of crossover components close.
Don't change reversible electrolytics caps to polypropylene, because the loss factor difference will result in audible frequency response changes. You would need to add series resistance if you do and you want to maintain the original frequency response.
Have you checked loss/dissipation factor? Electrolytic capacitors can dry out and become high resistance...... |
|
Back to top |
|
|
clubsport911 Senior Contributor 100+
Joined: 26 Aug 2012 Posts: 165 Location: Cheltenham, UK
|
Posted: Thu Nov 22, 2012 9:13 pm Post subject: |
|
|
Ref adding polypropylene caps etc... Is there a formula for the series reisistor you need to add? There are as many theories as there are Internet forums. I feel the one closest to my mind is 0.1ohm / 5 uf |
|
Back to top |
|
|
audiolabtower VIP Contributor 500+
Joined: 06 Jan 2009 Posts: 686
|
Posted: Thu Nov 29, 2012 7:34 pm Post subject: |
|
|
You can work it out from R=d/2pifC where d is around 0.04 for small caps and around 0.08 for large ones. Of course it is frequency dependant so I calculated around the crossover frequencies. For a single 3.3uF at 2.5kHz it was around 0.77 ohm, but at 5 kHz for example it would be around 0.4 ohm. I found a high power 1 ohm at the output of T52 and B110 around right, but it may depend what your room and position is doing to the bass response. Whether right or wrong I did not fancy putting series resistance to the B300. |
|
Back to top |
|
|
|