Do not forgett to solder fast. Caps are fragile and too long soldering tasks move the capacitance of the lythics...if precision matters. Cause you don t know at 10 to 12% with lytics precision if you don t measure.
Tip: Jantzen cross caps and SCR smart cap serie are good and precise...and cheap...
Tip: Jantzen cross caps and SCR smart cap serie are good and precise...and cheap...
Update
I tried other capacitors instead of yellow Ceraphine.
1) Elna RBD Bi-polar Electrolytic Capacitor 100v
This is the best capacitor that I tried by now. It gave best sound, very neutral, everything sounds smooth, no frequencies are too forward and I have kind of good sound picture in my head when listening to speakers with these capacitors inside.
It is closest brother to ELNA Ceraphine that was in place in original design.
But sound is not that wide open and clear as with two guys that I will describe below.
2) Jantzen Audio Premium ELKO Bi-polar – Electrolytic capacitors 70V 5% tolerance
Seller mistakenly sent to me wrong capacitors instead of this but I had other (almost analog) capacitors:
22uF 70V Mundorf ECap AC PLAIN electrolytic capacitor
Installed them and got pretty bad sound. Bass became more clear, but it does not go low at all. Mids are too forward and speaker's sound feels like curved. Highs are good but voices sound not natural at all.
I guess I got this result because of too low ESR value of new capacitor. Ceraphine have not neglectable ESR and looks like it played huge role in original design.
So if I want to try to improve the sound (at some level) I need to buy 10W resistors with several values (like maybe 0.5 Ohm to 1.5Ohm) and try to listen to sound with one resistor in series with new capacitor. This is going to be an attempt to simulate higher ESR value. But I am not sure it will work out good enough, as ESR of lytics is changing with frequencies. Will be glad to get some stories of success of this approach from your experience.
3) Jantzen Audio Cross Cap Bi-polar MKP 400V
This capacitor caused same sort of issues as Mundorf Electrolytic above.
Solution? Same as with Mundorf Electrolytic above.
I tend to keeping ELNA RDB in place (will order new pair, as current pair survived two soldering sessions and can be damaged).
I tried other capacitors instead of yellow Ceraphine.
1) Elna RBD Bi-polar Electrolytic Capacitor 100v
This is the best capacitor that I tried by now. It gave best sound, very neutral, everything sounds smooth, no frequencies are too forward and I have kind of good sound picture in my head when listening to speakers with these capacitors inside.
It is closest brother to ELNA Ceraphine that was in place in original design.
But sound is not that wide open and clear as with two guys that I will describe below.
2) Jantzen Audio Premium ELKO Bi-polar – Electrolytic capacitors 70V 5% tolerance
Seller mistakenly sent to me wrong capacitors instead of this but I had other (almost analog) capacitors:
22uF 70V Mundorf ECap AC PLAIN electrolytic capacitor
Installed them and got pretty bad sound. Bass became more clear, but it does not go low at all. Mids are too forward and speaker's sound feels like curved. Highs are good but voices sound not natural at all.
I guess I got this result because of too low ESR value of new capacitor. Ceraphine have not neglectable ESR and looks like it played huge role in original design.
So if I want to try to improve the sound (at some level) I need to buy 10W resistors with several values (like maybe 0.5 Ohm to 1.5Ohm) and try to listen to sound with one resistor in series with new capacitor. This is going to be an attempt to simulate higher ESR value. But I am not sure it will work out good enough, as ESR of lytics is changing with frequencies. Will be glad to get some stories of success of this approach from your experience.
3) Jantzen Audio Cross Cap Bi-polar MKP 400V
This capacitor caused same sort of issues as Mundorf Electrolytic above.
Solution? Same as with Mundorf Electrolytic above.
I tend to keeping ELNA RDB in place (will order new pair, as current pair survived two soldering sessions and can be damaged).
Nope, I did not check this.
My 20$ tester was dead on arrival ) Will need to order another one.
Not sure if its correct gear to use here as it will give ESR only at some 1 frequency, when we need to get kind of graph ESR vs frequency (ESR in range of frequencies).
But I know that both 22uF 70V Mundorf ECap AC PLAIN electrolytic capacitor and Jantzen Audio Cross Cap Bi-polar MKP 400V have really low ESR. Relatively to ELNA RDB and Jansen Audio (M.D.L.) which both give good speaker sound with my setup.
You can calculate ESR value directly from Loss angle [tan a] that is provided by manufacturer. So using this formula for ELNA RDB I got 5.3 Ohm at 120 hz.
ESR=tan/(2* π*f*C)
0.09 / (2 * 3.14 * 120hz * 22 * 10^-6F)= 5.3 ohm
But Loss angle [tan a] for Mundorf ECap AC PLAIN is given for 1khz, so no point in comparison.
As for Jantzen Audio Cross Cap Bi-polar MKP 400V loss factor is also given at 1kz, but I don't understand the numbers there:
https://hfc-fs.s3-eu-west-1.amazonaws.com/s3fs-public/cross-cap.pdf
Any way they are REALLY low in comparison to any electrolytic capacitor.
If we look on original Ceraphine capacitor
http://www.dalojan.nl/tube/elna cerafine datasheet.pdf
then we see that it has loss angle of 0.12 (vs 0.09 of Elna RDB). This gives us ESR or ~7 ohms at 120hz.
So I guess that this was best match for my speakers.
But there is no way of getting such new capacitor today (not produced any more?).
Maybe putting some additional resistor of 1 ohm in series with Elna RDB can bring some additional improvement!
Don't know.
My 20$ tester was dead on arrival ) Will need to order another one.
Not sure if its correct gear to use here as it will give ESR only at some 1 frequency, when we need to get kind of graph ESR vs frequency (ESR in range of frequencies).
But I know that both 22uF 70V Mundorf ECap AC PLAIN electrolytic capacitor and Jantzen Audio Cross Cap Bi-polar MKP 400V have really low ESR. Relatively to ELNA RDB and Jansen Audio (M.D.L.) which both give good speaker sound with my setup.
You can calculate ESR value directly from Loss angle [tan a] that is provided by manufacturer. So using this formula for ELNA RDB I got 5.3 Ohm at 120 hz.
ESR=tan/(2* π*f*C)
0.09 / (2 * 3.14 * 120hz * 22 * 10^-6F)= 5.3 ohm
But Loss angle [tan a] for Mundorf ECap AC PLAIN is given for 1khz, so no point in comparison.
As for Jantzen Audio Cross Cap Bi-polar MKP 400V loss factor is also given at 1kz, but I don't understand the numbers there:
https://hfc-fs.s3-eu-west-1.amazonaws.com/s3fs-public/cross-cap.pdf
Any way they are REALLY low in comparison to any electrolytic capacitor.
If we look on original Ceraphine capacitor
http://www.dalojan.nl/tube/elna cerafine datasheet.pdf
then we see that it has loss angle of 0.12 (vs 0.09 of Elna RDB). This gives us ESR or ~7 ohms at 120hz.
So I guess that this was best match for my speakers.
But there is no way of getting such new capacitor today (not produced any more?).
Maybe putting some additional resistor of 1 ohm in series with Elna RDB can bring some additional improvement!
Don't know.
Last edited:
I bought capacitors from some popular site for hifi, so if to trust them then they do matching of values good. They should not sell branded stuff that is out of specs too.
About original capacitor.
Guy from Set Crossover said that it measured ESR of 155 ohm on 1khz. Lets suppose this is true.
So, we take 22uF 70V Mundorf ECap AC PLAIN electrolytic capacitor and look at the spec:
https://cdn.shopify.com/s/files/1/0809/2387/files/Ecap_plain.pdf?9887667108540766049
We see there that the tan loss is 0.025 at 1khz.
Then its ESR will be :
ESR=tan/(2* pi*f*C)
0.025 / (2 * 3.14 * 1000hz * 22 * 10^-6F)= 0.18 ohm
This is not 0.0003 ohm, but very low too. And it is zero in comparison to Ceraphine's 155ohm.
Even if the guy was mistaken and Ceraphine's ESR lets say 1.5 ohm, then still it is 10 times higher then ESR of Mundorf ECap AC PLAIN.
About original capacitor.
Guy from Set Crossover said that it measured ESR of 155 ohm on 1khz. Lets suppose this is true.
So, we take 22uF 70V Mundorf ECap AC PLAIN electrolytic capacitor and look at the spec:
https://cdn.shopify.com/s/files/1/0809/2387/files/Ecap_plain.pdf?9887667108540766049
We see there that the tan loss is 0.025 at 1khz.
Then its ESR will be :
ESR=tan/(2* pi*f*C)
0.025 / (2 * 3.14 * 1000hz * 22 * 10^-6F)= 0.18 ohm
This is not 0.0003 ohm, but very low too. And it is zero in comparison to Ceraphine's 155ohm.
Even if the guy was mistaken and Ceraphine's ESR lets say 1.5 ohm, then still it is 10 times higher then ESR of Mundorf ECap AC PLAIN.
Last edited:
I guess you would mean to say 're-branded'
I don't understand much in capacitors market
Yes the Elna have 2 to 3 x times more ESTR than plain Lytics caps from Jantzen or Mundorf... (0.5 to 1 circa if I remember the datasheet of the Elna bipolar)
But the particular sound due to the model (assemly, materials, lytics formula, etc), notice that this ESR as we said before in the thread is certainly wanted in the crossover and match what they sim & measured... So nothing bad to swap with something close as you did and btw you like how it sounds ! 
... race over !
But the particular sound due to the model (assemly, materials, lytics formula, etc), notice that this ESR as we said before in the thread is certainly wanted in the crossover and match what they sim & measured... So nothing bad to swap with something close as you did and btw you like how it sounds
! ... race over !
I am with ELNA RDB 100V 22uf bipolar right now. Sound seems OK to me.
Its interesting how cheap China stuff (replicas) can sound.
I tried to buy some cheap stuff, but could not find anything available.
My next try will be Nichicon DB (if they are available at market):
http://products.nichicon.co.jp/en/pdf/XJA043/e-db_gb.pdf
They are only 50V and their ripple currents are several times higher then ELNA RDB or CERAFINE (740 vs 120 mAms).
The Nichicon GB series seems like less suitable - too low loss and too high ripple current.
Its interesting how cheap China stuff (replicas) can sound.
I tried to buy some cheap stuff, but could not find anything available.
My next try will be Nichicon DB (if they are available at market):
http://products.nichicon.co.jp/en/pdf/XJA043/e-db_gb.pdf
They are only 50V and their ripple currents are several times higher then ELNA RDB or CERAFINE (740 vs 120 mAms).
The Nichicon GB series seems like less suitable - too low loss and too high ripple current.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.