I just finished my BIIIse with IVY III output stage. I am only using the single ended output for now as my balanced amp in no where near being finished.
After about 8 hours of listening I feel that the resolution and level of detail is excellent perhaps even exceptional compared to anything I've built or modded to date.
However, the DAC seems bright and missing some low end. I tried it with my Dynahi 1.0 and my BH Crack. The crack tames it somewhat but it's still bright. Headphones used were HD650 and Headphile Darth Beyers V2 (Bass heavy). I intend to give it at least 100 hours of burn in but I suspect it will only mellow slightly. Basically I want to see what's involved in warming it up a bit and to see what others have done.
My first area to look at would be the opamp doing the BAL/SE conversion. Who has rolled the opamp? What did you use or settle on, and did you achieve better results without sacrificing detail or resolution. The second area would be to change cap values, and if so which ones? Also if I switched the resistors to an "audiophile" brand is 1% ok?
Page 4 of the manual shows the ones that would or could be changed, but I'd need direction on which ones to change and what values to use.
After about 8 hours of listening I feel that the resolution and level of detail is excellent perhaps even exceptional compared to anything I've built or modded to date.
However, the DAC seems bright and missing some low end. I tried it with my Dynahi 1.0 and my BH Crack. The crack tames it somewhat but it's still bright. Headphones used were HD650 and Headphile Darth Beyers V2 (Bass heavy). I intend to give it at least 100 hours of burn in but I suspect it will only mellow slightly. Basically I want to see what's involved in warming it up a bit and to see what others have done.
My first area to look at would be the opamp doing the BAL/SE conversion. Who has rolled the opamp? What did you use or settle on, and did you achieve better results without sacrificing detail or resolution. The second area would be to change cap values, and if so which ones? Also if I switched the resistors to an "audiophile" brand is 1% ok?
Page 4 of the manual shows the ones that would or could be changed, but I'd need direction on which ones to change and what values to use.
Last edited:
I don't think modding the IVY will change the balance you are seeing. I have used the IVY and the Legato, I think that the Ivy had a tighter bass than the Legato. Switching power supplies on the Legato to the latest Placids made the Legato bass faster and more dynamic, which I thought bought it closer to the Ivy.
What power supply are you using for the Ivy?
What power supply are you using for the Ivy?
Lots...
of options here... First of all, 1% is not good enough for the IVY-III resistors. You want to match all resistor pairs to at least .1%.
The output filtering in the stock IVY-III is already pretty strong, so I would not change that. But you could try different passive parts. Things like Takman Carbon Film resistors are a little warmer than most metal film types, but you will need to purchase a bunch of 1% resistors to get enough to match to .1%
For the filter circuit caps in the IVY-III you might try the Amtrans (Parts Connexion has them) capacitors in place of the stock Wimas as well.
Also, if you are using TPA Shunt regs to power it, try upping the CCS current. TPA mentions shunting 60 mA or so, try shunting 100 mA and see if you hear a difference (watch the heat when upping the CCS current).
Also, please describe the whole chain: what is the input to the B-IIIse? Are you running the fast or slow roll off FIR filter?
of options here... First of all, 1% is not good enough for the IVY-III resistors. You want to match all resistor pairs to at least .1%.
The output filtering in the stock IVY-III is already pretty strong, so I would not change that. But you could try different passive parts. Things like Takman Carbon Film resistors are a little warmer than most metal film types, but you will need to purchase a bunch of 1% resistors to get enough to match to .1%
For the filter circuit caps in the IVY-III you might try the Amtrans (Parts Connexion has them) capacitors in place of the stock Wimas as well.
Also, if you are using TPA Shunt regs to power it, try upping the CCS current. TPA mentions shunting 60 mA or so, try shunting 100 mA and see if you hear a difference (watch the heat when upping the CCS current).
Also, please describe the whole chain: what is the input to the B-IIIse? Are you running the fast or slow roll off FIR filter?
Prudent advice- the CCS does make a difference... Wima caps suck too btw, as do most metalized caps. They always have, even when there were fewer options.
Not metalized...
The filter caps provided in the TPA kits are Wima FKP2, these are actually film/foil polyproplene capacitors and not metalized types. But I do feel there are better sounding options which will fit, like Amtrans... REL RTE polystyrene film/foil can be made to fit as well, but the Amtrans are a tad "warmer", perhaps.
Try the slow roll off filter as well...
The filter caps provided in the TPA kits are Wima FKP2, these are actually film/foil polyproplene capacitors and not metalized types. But I do feel there are better sounding options which will fit, like Amtrans... REL RTE polystyrene film/foil can be made to fit as well, but the Amtrans are a tad "warmer", perhaps.
Try the slow roll off filter as well...
Source is a Music Hall CDP. Amplifiers are a single ended Dynahi and a Bottlehead Crack. Headphones are HD650, Headphile Darth Beyer V2, Joe Grado SR200
I found this post from barrows on another forum
PRP resistors matched to 0.1% for feedback and filters
Takman Carbon Film resistors matched to ~0.1% for output resistors
Vishay Roderstein .1uF film caps for the supply decoupling to the opamps (better than Wima)
Amtrans 8.2 nF film/foil polypro caps for the analog filters (vs the stock 15nF Wimas)
My question is with the amtrans caps. What effect does reducing the value to 8.2 nf have?
PRP resistors matched to 0.1% for feedback and filters
Takman Carbon Film resistors matched to ~0.1% for output resistors
Vishay Roderstein .1uF film caps for the supply decoupling to the opamps (better than Wima)
Amtrans 8.2 nF film/foil polypro caps for the analog filters (vs the stock 15nF Wimas)
My question is with the amtrans caps. What effect does reducing the value to 8.2 nf have?
This is pin compatible on a brown dog adapter. Has anyone tried this?
Also it looks like I would have to parallel two caps for C1 - C4 to get 0.015 uf
Is there any drwaback to doing this other than having to make it fit?
I have done some experiments with ivy
For me the sound was to sterille ,cold.
Some resistor changes that you suggest are good. I think i used Wishay.
Panasonic FM or FR are to cold-sounded for audio filtering. I use Nichicon KZ because midrange is more powerfull and energic.
I reduced filtercaps on 4,7n Styroflex capacitors
Lm4562 changed first for lme49720 metallcan... better but still some coldness in music to hear.
Best option was lme49990 with more than 100pf (100-200pf styroflex or silvermica) on feedback ( without cap lme49990 oscilated but lme49720 metalcan didn*t) and about 5-10mA AClass biasing ( think i used BF245Bor C selected for right biassing) . That is up to you what bias you will choose.
Thin was silver-gold Mundorf.
http://www.audio-kontakt.com/forum/...start=45&sid=1e3d2e72cc23186c4c391bd2d38e5e4a
This was middle version.
For me the sound was to sterille ,cold.
Some resistor changes that you suggest are good. I think i used Wishay.
Panasonic FM or FR are to cold-sounded for audio filtering. I use Nichicon KZ because midrange is more powerfull and energic.
I reduced filtercaps on 4,7n Styroflex capacitors
Lm4562 changed first for lme49720 metallcan... better but still some coldness in music to hear.
Best option was lme49990 with more than 100pf (100-200pf styroflex or silvermica) on feedback ( without cap lme49990 oscilated but lme49720 metalcan didn*t) and about 5-10mA AClass biasing ( think i used BF245Bor C selected for right biassing) . That is up to you what bias you will choose.
Thin was silver-gold Mundorf.
http://www.audio-kontakt.com/forum/...start=45&sid=1e3d2e72cc23186c4c391bd2d38e5e4a
This was middle version.
Lme49860 is +-22V lm4562 for me similar sound.
Lme49720 is better... 2X49990 on adapter is even better , low noise, more organic, highs are brillant, fine ...bas has more weight than lm4562 and lme49720.
opa627 is to soft for my taste in this configuration. Mycrodinamic and details are not so good as lme49720 or lme49990. Opa627 and lme49720 both need no compesation cap, but lme49990 does. Maybe all others have tried it without cap... then the IC oscilate and sound is wooly, shy, like a curtain over the speakers.
Lme49720 is better... 2X49990 on adapter is even better , low noise, more organic, highs are brillant, fine ...bas has more weight than lm4562 and lme49720.
opa627 is to soft for my taste in this configuration. Mycrodinamic and details are not so good as lme49720 or lme49990. Opa627 and lme49720 both need no compesation cap, but lme49990 does. Maybe all others have tried it without cap... then the IC oscilate and sound is wooly, shy, like a curtain over the speakers.
Changing C1-C4 lowers/highes high cutoff frecuency 8n2 from 55khz to 100khz. So the phase margin is not so deep in 20khz range
(Sample)RC Low-pass Filter Design Tool - Result -
Feedback cap i meant across resistor R22 and R24 (2,2kohm...) you add 100-220pf styroflex or silver mica cap when you use lme49990.
If you don*t have scope you better use lme49720 metallcan !!! not plastic case. Nichicon KZ or Silmic 2 -100uf capacitors plus 1uf mkp and 100nf mkp or pps on power supply decoupling.
I also changed opamp in AVCC regulator do some more filtering and give better decoupling on both other regulators..... clock regulator had huuge oscilations...measure and you willl see it...that oscillation even goes in both other regulators. By me supply for +5V was Salas shunt.
(Sample)RC Low-pass Filter Design Tool - Result -
Feedback cap i meant across resistor R22 and R24 (2,2kohm...) you add 100-220pf styroflex or silver mica cap when you use lme49990.
If you don*t have scope you better use lme49720 metallcan !!! not plastic case. Nichicon KZ or Silmic 2 -100uf capacitors plus 1uf mkp and 100nf mkp or pps on power supply decoupling.
I also changed opamp in AVCC regulator do some more filtering and give better decoupling on both other regulators..... clock regulator had huuge oscilations...measure and you willl see it...that oscillation even goes in both other regulators. By me supply for +5V was Salas shunt.
I see what you're talking about. Someone went down to 8.2nf and you went down to 4.7nf I'm just not sure if I should keep the stock value or lower it. I really want it to sound not as cold or forward on the high end. So I think I would want to lower the value of C1 - C4 correct?
I should switch to a pair of LME49990 so I will add the feedback cap across R22 and R24
I don't have a scope so I'll go with LME49990 and Silmic II caps
I'll revisit this a little later on
If you have current versions of Trident they are absolutely stable. Older versions can benefit from a comp cap on the error amplifier. AVCC is very stable - newer versions have improved error amplifiers. 🙂
As for "Warmer" sound, you could experiment with raising the output impedance. And possibly increasing the filter capacitors - this will shift the high end phase a bit and tend to accentuate the mid band.
The latest iteration of the Placid BP is a must have for either the IVY or the Legato. This last iteration has the benefit of exceptional user feedback which has influenced its final form. It has taken the supply to a new level in terms of low noise and low output impedance. The difference it makes in an output stage can be stunning.
As for "Warmer" sound, you could experiment with raising the output impedance. And possibly increasing the filter capacitors - this will shift the high end phase a bit and tend to accentuate the mid band.
The latest iteration of the Placid BP is a must have for either the IVY or the Legato. This last iteration has the benefit of exceptional user feedback which has influenced its final form. It has taken the supply to a new level in terms of low noise and low output impedance. The difference it makes in an output stage can be stunning.
Does your strong Placid BP recommendation also apply to Buff II with current gen Tridents + Legato v2? I'm currently using the LCBPS and I've been thinking about a PS change.
Also, is the original 30VA 15+15V trafo adequate for the above setup if I change to the Placid BP? I notice your current offering is 50VA.
BK
Also, is the original 30VA 15+15V trafo adequate for the above setup if I change to the Placid BP? I notice your current offering is 50VA.
BK