A wire is also the quickest way to fry your outputs - when you connect to an amplifier, "modded" in the same way by another DIY, that has some 50-100mV offset.
The difference between dual caps (nice ones) and wire is insignifiant in audio range (even up to 100kHz). Really, did you test to hear the difference? I did and was impossible to discern when it was the caps and when the "short" wire across them.
The difference between dual caps (nice ones) and wire is insignifiant in audio range (even up to 100kHz). Really, did you test to hear the difference? I did and was impossible to discern when it was the caps and when the "short" wire across them.
Last edited:
I've done the same testing, repeatedly over the last 30 YEARS, and I clearly hear the difference EVERY time. I've eliminated dc blocking caps on HUNDREDS of players, dacs, preamps, tape decks, power amps, etc., and have NEVER had anybody "fry" anything as a result. If you didn't hear any difference, the rest of the system was not adequately good.
Well, this is very interesting and informative post, thank you.
However, looking on the schematics, and especially in the PCB traces, both R707/708 are unusually far away from the I/V IC pins; does it mean that designers count on trace capacitance and because of that did not put Cfb?
Just asking, I have no idea
Well, this is very interesting and informative post, thank you.
However, looking on the schematics, and especially in the PCB traces, both R707/708 are unusually far away from the I/V IC pins; does it mean that designers count on trace capacitance and because of that did not put Cfb?
Just asking, I have no idea
I don't think anyone ever considered trace capacitance.... you'll notice parallel combination of C702 and C704 that negate any possibility that someone took trace capacitance in to consideration...you'll also notice the virtual ground point is at R704 and R708 junction - right were the PCM 63P BPO signal is tapped from -> which is correct & presents excellent opportunity for current IC I mentioned before (AD812).
What R709 needs is 47pF – 100pF capacitor in parallel, or even better - the combination of 100 ohm resistor and 47pF capacitor.... this is with VFB IC’s. Placing the CRO on it would help get the capacitor value just right, but 47pF is safe bet if one does not want to use CRO…
I know the whole stage would sound much better with AD812, if the guidance I posted earlier is implemented correctly – there is same amount of work needed for either VFB / CFB IC anyway, so why not try sthe uperior CFB IC like AD812….
Boky
At the moment I don't have AD812 (but they are on their way
), but I have lot od other Analog devices and BB chips.I've cut the trace between pin 6/7 and put 1k SMD resistor - I don't know if it is pure psychological efect, but even the VFB chips sounded much better than before (I've tried LM6172, OPA2604 - the deafult in this player - AD825, 826, 827)... for now, I left the old op275, which seems to be most balanced.
I did not tried Cf still (I had none on hands in the moment), but I will
Do not worry about Cf if you are going to try the current feedback IC - the AD812....
The 1k resistor is very important because it will prevent the CFB IC from oscillations and self-destruction due to large heat dissipation
Let me know if you need any further help…… and if you liked the AD812...
Good luck,
Boky
The 1k resistor is very important because it will prevent the CFB IC from oscillations and self-destruction due to large heat dissipation
Let me know if you need any further help…… and if you liked the AD812...
Good luck,
Boky
I did this - skipped the FT-3 though because of space considerations and bypassed with Ero 1837 0.1uF instead. The sound is improved some except that it's now quite bass-heavy, whereas before it was rather bass-shy. The effect is most noticeable on HDCD discs, where it at times sounds quite unnatural, especially in piano recordings. I'm rather theoretically deficient in electronic matters, but I couldn't help but notice that there are two resistors following these caps before the output - 100 ohm to the output and 560 ohm to ground, like a fixed-position volume control. Should either or both of these be changed in value as well to avoid this drastic change in frequency response? I seem to remember in my dim dark past that a cap followed by a resistor constitutes a filter. Have I upset part of the design by reducing the value of the output cap 100x?
PS AD812 is next!
Last edited:
I looked up the formula on Wikipedia (yup, that's about my speed at the moment) and calculated the half-power point at around 16 Hz for the original arrangement (100uF, 100 ohm). That would correspond pretty well with what I heard before the change (assuming the response would roll off smoothly down to this point), which was OK but a little thin compared to my reference player. That would make the current half-power point 1600 Hz in the present arrangement (1uF, 100 ohm). My question is if it's still a high pass filter, why is the bass so present? Does the filter response curve have a minimum and then rise again below that point?
I suppose I could restore the original curve by increasing the value of the resistor 100x, but that would make for high output impedance. Are these resistors really necessary? I suppose they're there for a reason. Would I run an unreasonable risk if I just jumpered the 100 ohm resistor (and maybe removed the 560 ohm to ground as well)?
Last edited:
Success!!
Several sites suggest replacing the op-amps in RCD-971 with AD812, so I thought I'd try it. I connected pins 6&7 with a 1K SMD resistor and cut off pin 6, because I didn't want to go to the trouble of cutting a trace. I took out the OPA2604's and installed sockets (I want to try other opamps later) and popped in the AD812s. Result - lots of noise (hiss) and a weak signal. Very discouraging, but I was determined to fix this myself (the DIY spirit, you know). I had bypassed the output resistors with jumpers in an effort to tame the bass (it helped a little), so thinking that was the problem I removed the jumpers. No sound and the opamps got too hot to touch! Yikes! So I thought - I'd heard of opamps oscillating, but I don't have a scope to check it out - so I'll just try to fix it and if it works, all is good. Tangent's site was a mine of information on "cranky opamps", and the first I tried (0.1uF ceramic caps from power in to ground) worked! This is probably small potatoes to most of you, but I was very happy to be able to fix this, and I thank Tangent for maintaining his site. Now for some burn in. Pulling the output filter caps helped the overall response, so the bass doesn't seem quite as far out of whack as it was now that there's more treble to balance it, and the AD812's seemed to clear up the midrange some. This is fun, when it works!
Several sites suggest replacing the op-amps in RCD-971 with AD812, so I thought I'd try it. I connected pins 6&7 with a 1K SMD resistor and cut off pin 6, because I didn't want to go to the trouble of cutting a trace. I took out the OPA2604's and installed sockets (I want to try other opamps later) and popped in the AD812s. Result - lots of noise (hiss) and a weak signal. Very discouraging, but I was determined to fix this myself (the DIY spirit, you know). I had bypassed the output resistors with jumpers in an effort to tame the bass (it helped a little), so thinking that was the problem I removed the jumpers. No sound and the opamps got too hot to touch!
Yikes! So I thought - I'd heard of opamps oscillating, but I don't have a scope to check it out - so I'll just try to fix it and if it works, all is good. Tangent's site was a mine of information on "cranky opamps", and the first I tried (0.1uF ceramic caps from power in to ground) worked! This is probably small potatoes to most of you, but I was very happy to be able to fix this, and I thank Tangent for maintaining his site. Now for some burn in. Pulling the output filter caps helped the overall response, so the bass doesn't seem quite as far out of whack as it was now that there's more treble to balance it, and the AD812's seemed to clear up the midrange some. This is fun, when it works!
The AD812 is a current feedback device rather than voltage feedback of "conventional" opamps. That's a fundamental difference in the way it should be used.
I haven't looked at the Rotel stages but I doubt the AD812 is a suitable device. If you can hear "changes" in the tonal response when experimenting with it then I think it will be oscillating and worse,
http://www.diyaudio.com/forums/anal...u-have-checked-see-its-stable-havent-you.html
I haven't looked at the Rotel stages but I doubt the AD812 is a suitable device. If you can hear "changes" in the tonal response when experimenting with it then I think it will be oscillating and worse,
http://www.diyaudio.com/forums/anal...u-have-checked-see-its-stable-havent-you.html
Yeah, I noticed the current feedback thing on the data sheet, but since others (Octave Electronics, for example) had got good results with this opamp I thought I'd try it. I assume the 1k resistor is to compensate for the difference in feedback, but I don't know much about these things as you can probably tell. On paper the AD812 may not look like an appropriate device, but if it works and sounds really good, who cares? At least it's not getting hot now. I'm also going to try LM6172 and LM4562, again on the strength of the positive experience of others, which has to be my guide until such time as I learn a lot more about electronics. I'm not going to pop opamps in at random! But thanks for your feedback, and I'll take a good look at your link.
Didn't experience this until I hooked it up to my new DCB1 with 25K PEC pots - wow, the bass simply disappeared! It sounds pretty normal into any other preamp, from a Lightspeed to a 6922 tube buffer to a Lite Audio A-15 or even a standard B1. The DCB1 sounds OK with other players - a little lean maybe, but pretty good otherwise.
AD812 works perfect. I had a problem with first chips I purchased ( SMD on adapter), meaning they were non working...and these days found DIP ones for a fair price
It really transforms this player in a very good way, but the most prominent is the bass area - controlled, fast and a bit dry, resembling very much to Krell sound (i'm using KAV400XI). I was never satisfied with stock opamps (2604), but nor with the other upgrades -there was always somewhat prominent mid-bass, unreal IMO. Not anymore:-D.
Mids are fair and detailed, and highs very clear, but not clean...but that's only after several hours working, rpobably the chip needs some burn in.
Another big improvement is incredible headroom; mostly of tried opamps are on listening impression are in some way "saturated", and thus flat on dynamic jumps. But this chip changed everything, really.
Boky, aprreciate for advice, this really works
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.