| filholder |
Hi,
Ok i built a set of monobloacks based on Brian GT's Lm3886 boards. They have been playing through a NAD C350 for a while and have sounded absoltutely great. However i thought that using the pre-amp section of the NAD, good kit though it is, must be in someway muddying the sound. I thought that a passive pre-amp was in order but since the hifi is also used for movies etc. decided to a remote controlled passive pre-amp based on the Lite Audio MV-01 boards. These are very reasonably priced and fitted my needs.
I finally finished this project over the weekend. The passive pre is wire entirely with home grown audio solid silver teflon insulated wire. Basiclly tried to make all wiring as short as possible. As my main source I am using a heavily modded Arcam Alpha 5+.
Anyway to cut a long story short the sound now sound less dynamic, the bass wollier, less slam, less impact. I am very dissapointed. Now since this is passive i have go any capacitor to bed in so thi sis it.
What should i do. From readin here i guess the issue is the impedence. The preamp was built into a tiny case and i havent got any space to put in an active stage. Can i buid a seperate buffer stage and what would people recommend.
I know there is the massive GC Preamp Suggestions thread but there dont appear the be any boards for sale.
Any advice greatly recieved.
Fil |
|
|
| analog_sa |
The passive resistive pre represents a classical dilemma. It offers crystal clear and undistorted sound but often lacks bass definition, power and guts. Even if impedance requirements are satisfied. Getting the passive as close as possible to the power amp (preferably in the same case) is often good but the sound may remain gutless all the same.
You need some form of buffering so the pot does not have to drive any length of cable directly. Many possible solutions but very hard to find one which sounds truly transparent. |
|
|
| carlosfm |
| quote: | Originally posted by analog_sa
You need some form of buffering so the pot does not have to drive any length of cable directly... |
...and the low input impedance of these bipolar-input (power) op-amps.;) |
|
|
| filholder |
Thanks guys,
So would a buffer in a seperate box but hardwired into the preamp be OK? Has anyone got a ciruit based on a high quality op amp, say Ad825 or the BB 627 which I could implement on say verio board?
I thought i way i was doing it was the same as the most people would seem to use just GC and the either a stepped attenuator or a pot. But after listening to it for so long thorugh the NAD C3350 pre it does sound very flat.
Are there no buffer kits?
Fil |
|
|
| jaudio |
| quote: | Originally posted by carlosfm
...and the low input impedance of these bipolar-input (power) op-amps.;) |
What is the input impedance of the lm3886? |
|
|
| carlosfm |
| quote: | Originally posted by jaudio
What is the input impedance of the lm3886? |
It's not specified in the National datasheets, but for bipolar-input op-amps it is usually around 300k and lower (instead of several Mohms of the fet-input op-amps).
But the problem is that to prevent dc-offset you have to use a low value input resistor resistor to ground.
Bipolar-input devices are very sensitive to this value, while the fets are very immune to this.
That's why a bipolar-input op-amp varies dc-offset according to the position of the volume pot.
I say like I'm repeating for years: the amp needs a pre.:D
Yesssss.
Dynamics, SLAM, soundstage, transparency.:cheerful:
:cloud9: |
|
|
| AndrewT |
Hi,
I stopped using passive pre about 20 years ago but recall being very pleased with the sound.
100k pot into 2m of 50pF/m cable rolled off the top end very slightly but bass seemed very natural.
Could your existing pre have a big bass hump that leaves your new sound seeming thin? |
|
|
| filholder |
Its not just bass though, the bass is looser but also it about the general presentation. It just lost a lot of impact all acroos the spectrum.
Even my girlfriend who basiclly think i am off my head with this audio stuff said it sounded boring and for once she is right.
Please guys what are my options. |
|
|
| carlosfm |
| quote: | Originally posted by filholder
So would a buffer in a seperate box but hardwired into the preamp be OK? |
Yes.
But ideally you should put it inside your passive pre, after the pot, with short wires.
| quote: | Originally posted by filholder
Has anyone got a ciruit based on a high quality op amp, say Ad825 or the BB 627 which I could implement on say verio board? |
Try this. |
|
|
| AndrewT |
Hi,
I like the ClassA switch.
Can you change the ClassA setting on the fly or do you need to power down? DC pulses etc?
A word of warning; the ClassA setting will drop out on excessive current. Take care with high capacitance/long cables and low impedance inputs. |
|
|
| filholder |
Thanks a million Carlos,
Very kind of you. Actually dont mean to take the P@ss but have you got an actual circuit board image of that design. I have recently aquired a old lazer printer from work and i might be a nice project to try and print my first PCB with.
If you havent fair enough i will have a go myself. Been a few years since GCSE electronics and i have a job read circuit diagrams these days but then maybe no pain no gain!!.
Thanks very much.
fil |
|
|
| nina |
| quote: | Originally posted by analog_sa
The passive resistive pre represents a classical dilemma. It offers crystal clear and undistorted sound but often lacks bass definition, power and guts. ...
..
You need some form of buffering .... |
Analog_sa,
I am facing the same thing as you mentioned via the passive amp. I am thinking of having a "simple" buffer opamp (unity gain, directly connect the output to i- without res) between my nOS DAC and the pot, then put a long length to the chip amp on seperate box.
As having four of LF356 on hand, will they good enough for that purpose??
:scratch: |
|
|
| carlosfm |
| quote: | Originally posted by AndrewT
I like the ClassA switch.
Can you change the ClassA setting on the fly or do you need to power down? DC pulses etc? |
No, this is not meant to be switched on-the-fly.
Actually it is meant to be a pair of jumpers on the board, so that you can use whatever op-amp you like, biased or not.
I like to count with everything, and this makes the board quite flexible, or should I say "universal".
| quote: | Originally posted by AndrewT
A word of warning; the ClassA setting will drop out on excessive current. Take care with high capacitance/long cables and low impedance inputs. |
No problems, even with 5m cables and 10k input impedance on the amp.
This is not the pre I use on my main system, but I have several boxes laying around here...:D |
|
|
| Nuuk |
| Filholder, you may care to read this . ;) |
|
|
| filholder |
Very interesting stuff. I fancy building a buffered based on opamps i think simply beacuse i have a few about and they seem easier to implement. Also i have read that the op-amps reject power supply noise better and since i will have to built that as well maybe it woldnt be an optimal design.
However Nukk's strip board design looks the easiest since he has given a picture which i can understand rather than a cicuit diagram which i dont. Also meant to sound good.
I am going to have to sleep on this.
Oh the other thing is what sort of size tranny would a buffer need? Tiny i guess 20-30VA?
fil |
|
|
| AndrewT |
Hi,
a stereo buffer could draw upto 100mA if driven really hard.
This still equates to only 3w6.
However these tiny transformers have enormous regulation some approaching 20%.
You must avoid excess voltage on the supply pins when drawing quiescent current.
Your 20VA sounds about right, but go for 18v + 18v and use good regulators to control the voltages back down to +-17v or 18v if your opamp can go this high. |
|
|
| Nuuk |
| quote: | | Oh the other thing is what sort of size tranny would a buffer need? Tiny i guess 20-30VA? |
Yes, that would be plenty! I use 80VA because I can then take off suppies for other items like my phono stage, sub crossovers etc.
The three transistor buffer is surely easier to build than an opamp circuit but as the power supply is 80-90% of the cost, you could build an opamp buffer and the discrete transistor version and compare them both! ;)
You can find my buffer/pre-amp PSU on DD as well. Several people here have built one and found it to be very good. :att'n: |
|
|
| filholder |
Nuuk,
Ok this sounds pretty good. I fancy building a decent phono stage at a later date since i am only using a NAD PP2 at the moment. If i build the PSU with a 80VA tranny to share with the phono stage how would i do it. Build your standard PSU in one box and then build seperate second stages of regulation in each of the served devices (i.e. the buffer and phono stage)?
Also if i build the transistor buffer i guess i would not need any input or output caps since all my sources and my GC power amps have then as well. I think might be the route i will go since i have some BC 547 felt over from repairing my Arcam Alpha CD player.
This does all seem very scarey, will be the first ciruit i have ever built myself and i feel like i know too little. A long way from buildin a chip amp kits which was very easy, only difficultty being the stupidly small chassis i keep building all my stuff in. I guess i will learn at some point.
Anyway thanks again
Fil |
|
|
| Nuuk |
| quote: | | Build your standard PSU in one box and then build seperate second stages of regulation in each of the served devices (i.e. the buffer and phono stage)? |
That's the idea! May be build the PSU in a box and have three (or more) outlet sockets so you can just plug in extra circuits later.
The PSU was one of the first things that I built so don't worry too much. Take your time and fully understand what goes where. Avoid any short circuits and you shouldn't smoke anything. Test each stage as you do it, ie output from transformer, output from rectifier, output from regulator stage 1 and 2 etc.
The transistor buffer must have the output cap - there is over 600mV going into it! Some people have found they need a cap on the input but try it without first. ;) |
|
|
| filholder |
Nuuk,
Are the 100nf caps mentioned without a farnell order code X2 or Y2 class caps? Thats all i can find. Sorry it this is a stupid question.
Phil |
|
|
| Nuuk |
| quote: | | Are the 100nf caps mentioned without a farnell order code X2 or Y2 class caps? Thats all i can find. Sorry it this is a stupid question. |
No not stupid! ;)
I use polypropylenes but the ones I use are no longer available. Farnell code 387-8480 will do fine! :att'n:
As a rule, you will only use Class X2 or Y2 with mains voltages. :att'n: |
|
|
| dsavitsk |
I can vouch that Nuuk's buffer is really really good.
Compared to my Melos pre (that was a stereophile recommended component) with super high end tubes in it (amperex 7308) there is more detail and dynamics with Nuuk's buffered passive. The Melos may be slightly more refined sounding, but that is likely due to the fact that I prefer tube sound. For SS, however, the buffer is great.
| quote: | Originally posted by Nuuk
The transistor buffer must have the output cap - there is over 600mV going into it! Some people have found they need a cap on the input but try it without first. ;) |
I needed to put both a 1uF cap and a 1K resistor in series after the pot and before the buffer (I did the cap first and it was not quite enough. adding the resistor fixed the problem, but I did not try it again with just the resistor which may work as well).
For my PS, I used a velleman (essentially a lm317) that has been snubberized and modded with larger voltage caps (63v) running at 30v. I then used a tle2426 to split the rails (2 actually, one for each side as I was not sure that one could provide enough current for both channels) and put 1 1000uF cap and a couple of assorted smaller film caps on each rail (so, that's 4x1000uF plus some other stuff) and this seems to work great. The transformer is about 30VA.
-d |
|
|
| postpunk |
I have just built Nuuks discrete buffer along with the discrete regulators, and I have to say that my gc just doesn't sound like my gc anymore.
I thought that perhaps this buffer-talk was a bit over exagerated, but that is apearantly not the case.
The drums hit harder, but with "air" around them, I can clearly hear hiss on some older recordings which I did not notice before. It's like having a new system. |
|
|
| Nuuk |
| quote: | | I needed to put both a 1uF cap and a 1K resistor in series after the pot and before the buffer (I did the cap first and it was not quite enough. adding the resistor fixed the problem, but I did not try it again with just the resistor which may work as well). |
I'll update the GCpreamp page with that in case anybody else needs it! ;) |
|
|
| Nuuk |
| quote: | | I have just built Nuuks discrete buffer............ |
Another happy customer! :) |
|
|
| dsavitsk |
| quote: | Originally posted by Nuuk
I'll update the GCpreamp page with that in case anybody else needs it! ;) |
This was with a 50K pot, if that is important.
-d |
|
|
| Nuuk |
| quote: | | This was with a 50K pot, if that is important. |
Yes, it seems to be pots higher than 10K that may cause a problem. Don't ask me why though - I just follow the instructions and pray for success! :D |
|
|
| carlosfm |
| quote: | Originally posted by postpunk
The drums hit harder, but with "air" around them, I can clearly hear hiss on some older recordings which I did not notice before. It's like having a new system. |
Yes, yes, yes! :cheerful:
Now listen to Ben Harper (which are mostly analog recordings) and delight yourself with the hiss.
If the hiss isn't there, very noticeable in the low passages, there's a blanket somewhere - or everywhere - in the system.:cool:
Oh glory, oh joy... can you now listen to your favourite music without tapping your foot?
Sing along, enjoy.;) |
|
|
| filholder |
Nuuk,
Your a star , if only i could get such quick responses out of Dell ever time one of our servers go down!!
The more people talk about Nuuk's buffer the more i want to do it, by the way is there a great deal of difference between the two types of secondary regulation. I guess the decreet method is more sophisticated however looks like it will take up a lot more space and the enclosure i have in mind is pretty small.
Also which buffer did people build out of the 2 described, just interested. I like the idea of the simple one. So few components seems to appeal after build the GC.
Cheers,
Fil |
|
|
| Nuuk |
| quote: | | by the way is there a great deal of difference between the two types of secondary regulation. I guess the decreet method is more sophisticated however looks like it will take up a lot more space and the enclosure i have in mind is pretty small. |
IMHO, the discrete regs are marginally better but it is very close. As regards space, there isn't a lot in it and if you look at the DD GC pre-amp
you can see it all fits in quite a small space!
| quote: | | Also which buffer did people build out of the 2 described, just interested. I like the idea of the simple one. So few components seems to appeal after build the GC. |
If you mean the three transistor (BJT) buffer and the Jfet, then I can't tell you how many of each have been built. I have built both and the BJT is certainly easier/cheaper to build, and more compact. Sound-wise, they are both excellent! :) |
|
|
| Upupa Epops |
| Nuuk, in every time, when I see your roller, I'm quite fascinated ;) . |
|
|
| Nuuk |
| quote: | | Nuuk, in every time, when I see your roller, I'm quite fascinated |
Ha, ha Upupa. In the UK, a roller is the slang for Rolls Royce but I don't own one of those!
These 'tubes' can be good hi-fi cases. If I had a lathe to turn the front and back panels, it would be a much quicker job! :smash: |
|
|
| AndrewT |
Hi Nuuk,
don't tell us all your secrets.
We might try using them to digitally control a six channel analogue attenuator behind all that woodwork.
Certainly different & wood suit my decor.
How about some thoughts on using a single log cut off centre and skewed to show the grain and no joins at the front or sides. Now that would be a secret (without a lathe). |
|
|
| Nuuk |
| quote: | | How about some thoughts on using a single log cut off centre and skewed to show the grain and no joins at the front or sides. Now that would be a secret (without a lathe). |
That's my next project Andrew - just as soon as I have fully trained my team of master-craftsmen woodworm! :D |
|
|
| Upupa Epops |
| To Nuuk : OK, not by roller, but by rolling pin I'm fascinated ;) . |
|
|
| Nuuk |
| quote: | | To Nuuk : OK, not by roller, but by rolling pin I'm fascinated |
You found me out! I was making pastry when I had the idea of the pre-amp case! ;)
How many people have pre-amps that are useful in the kitchen? :att'n: |
|
|
| Upupa Epops |
| Also another machine you can use in kitchen ;) , class A amp for example is ideal for preparing " ham and eggs " :D . |
|
|
| Nuuk |
| Or a Teres turntable for icing cakes! :clown: |
|
|
| filholder |
Hi Nuuk,
Just to let you know i have ordered all the bits for the psu and buffer. I will have a go at sticking all together hopefully this week, depending on Maplin and farnell delivery time. Can't wait.
Afraid there also might be a few more questions about assembley.
Anyway thanks a lot, can't wait to hear the buffer in action.
Fil |
|
|
| carlosfm |
| You will soon find out that it's very easy to make much better than the Nad C350 preamp seccion. |
|
|
| AndrewT |
Hi Nuuk,
when we refer to your buffer, are we talking about the last version (discrete) or the classA biased opamps? |
|
|
| filholder |
Hi Carlos
"You will soon find out that it's very easy to make much better than the Nad C350 preamp seccion."
That is very much what i am hoping!!!!
OK, look i know the NAD is no where near top dollar hifi, for the price i payed (less than half retail price) it was reasonable.
What is interesting to me is that no matter how lacking in sophistication the NAD pre section is, it is far better than a passive one. And there are loads of people on this forum using either just a pot or switched attenuator. Gain clones need an active pre that is for sure, with a passive the sound has so much less athority and gets muddled by compliacted music.
I only hope that Nuuk psu and buffer stage can take things a stage further.
fil |
|
|
| Nuuk |
| quote: | | Afraid there also might be a few more questions about assembley. |
No problems Fil, that's what Carlos is here for! :D Seriously, I am happy to answer any questions.
| quote: | | when we refer to your buffer, are we talking about the last version (discrete) or the classA biased opamps? |
Andrew, in the context of this thread we are talking about the discrete, three-transistor buffer.
| quote: | | I only hope that Nuuk psu and buffer stage can take things a stage further. |
Faith can move mountains (and seriously improve your listening experience). :att'n: |
|
|
| Shoog |
There are passive preamps and there are passive preamps.
I did a side by side comparison of a basic Alps 10K pot and another 50K pot. My friend and myself definately found the little Alps 10K to sound better than the 50k. Infact we agreed that it sounded better than the basic JTL ECC88 buffer, and a Bride of Zen preamp. The only thing that bettered it was a little 5687 valve preamp, which sounded simply great and full on.
Your passive preamp may be just a poor implementation. Quality of parts becomes very critical in such a simple setup.
The bipolar transistor buffer given on Nuuks site is a fine performer for it simplicity. Very good PSRR, and the nearest thing to a valve sound I have come across. Its tone can be tuned by altering the bottom transistors bias resistor.
Shoog |
|
|
| Nuuk |
| quote: | | The only thing that bettered it was a little 5687 valve preamp, which sounded simply great and full on. |
Was it anything like this one Shoog? I have yet to try mine with a GC but it is one job on the 'to-do' list! ;) |
|
|
| Shoog |
No its not the one I built, but basically its very similar.
Heres the one I built,
http://www.diyparadise.com/simplepreamp.html
Simpler and with valve rectification. I didn't use a choke because I ran out of space. Mine is crammed into a dead PC powersupply case (thats including the two transformers it took). I had to go DC with the heaters because the 5687 like to hum. I daren't even contemplate taking this one apart to do any more work (its a mess of wires in there), still sounds fantastic and cost me about 100.00 to build.
Shoog |
|
|
| Nuuk |
| quote: | | I had to go DC with the heaters because the 5687 like to hum. |
Yes, mine does too and a DC heater supply is also on the 'to-do' list! :smash: |
|
|
| carlosfm |
| quote: | Originally posted by filholder
OK, look i know the NAD is no where near top dollar hifi, for the price i payed (less than half retail price) it was reasonable. |
I know the Nad C370 very well, the next model up in the range.
What limits it's performance the most is the pre.
But it's no exception, that's the problem of most integrated amps, independent of the price.;) |
|
|
| carlosfm |
| quote: | Originally posted by Shoog
My friend and myself definately found the little Alps 10K to sound better than the 50k. |
Naturally.
You should not use more than 10k for a passive pre.
The 10k pot could even be a cheap one, it could still sound better.
Sometimes it's not a case of using fancy components, it's just a case of picking the right values for the right job. |
|
|
| Shoog |
Some CD players will complain with a 10K load (though not many these days). The 50K pot was an afterthought on the Bride of Zen, it was simply taken off the pot as an extra output.
I stand by the statement that the quality of the components will have a bigger effect on the sound in a passive than in an active. I also agree that the most critical factor is the ohmic value of the pot.
Shoog |
|
|
| filholder |
Hi,
What would be a good value for the pot in my pre-amp once i have installed nuuks 3 transistor active stage?
fil |
|
|
| carlosfm |
| quote: | Originally posted by Shoog
Some CD players will complain with a 10K load (though not many these days). The 50K pot was an afterthought on the Bride of Zen, it was simply taken off the pot as an extra output.
I stand by the statement that the quality of the components will have a bigger effect on the sound in a passive than in an active. I also agree that the most critical factor is the ohmic value of the pot.
Shoog |
Open the CDP and change the output stage to low impedance.
That alone makes a huge improvement.;) |
|
|
| carlosfm |
| quote: | Originally posted by filholder
What would be a good value for the pot in my pre-amp once i have installed nuuks 3 transistor active stage?
fil |
10k, or 20k max.
The ideal is to have the pot on the active stage.
No cables.
Place the active stage and the input selector (if any) near the inputs for very short cables. |
|
|
| filholder |
Thanks carlos,
Anyone know of anywhere in the u.k. to get a cheap Alps motor pots. Looks like i need a new one.
fil |
|
|
| Nuuk |
| quote: | | What would be a good value for the pot in my pre-amp once i have installed nuuks 3 transistor active stage? |
I use a 10K (stepped attenuator) with mine! ;) |
|
|
| filholder |
Sorry i just wanted to ask if there is anyway of reducing the resistance of my current pot. Well i was guessing there is by putting a resistor beween the input and output, would this help?
fil |
|
|
| filholder |
My problem is that my whole setup is used for TV and films etc. so i need it to be remote controlled really. Well with my missus involved it is not even a option really no remote no go. So i bought the Lite Audio remote board since it was very cheap used good realys to control the input and had a Alps pot on. All for about Ģ30 delivered. Now seems the pot might be a big problem, it is 100k and in the passive pre i have got i havent got room for anything else.
Is there now way to lower the value? Alps dont do the same one in the right value. I can get a 10k linear but i imagine that if i did the law faking resistor on that it would make the value even lower.
Another mess i have got myself into.
Fil |
|
|
| Nuuk |
| quote: | | Sorry i just wanted to ask if there is anyway of reducing the resistance of my current pot. Well i was guessing there is by putting a resistor beween the input and output, would this help? |
I'm really not sure about that one Fil. Perhaps one of the 'experts' can answer. ;) |
|
|
| pinkmouse |
| That would work, but you might be better off by adding another unity gain buffer stage before the pot. |
|
|
| filholder |
From a simple calculation i guess i would need something like a 10-15k resistor to lower the value to somewhere in the region of 10-13k. Does that seem right?
Fil |
|
|
| filholder |
Hi,
yeah i saw them on hifi collective, bit pricey and the 'blue beauty' one are 70 mm long. Too long for my case.
Is sticking a 20k resisor in series with the pot a really bad idea. That should give me 16.7k. The onlt thing is my pot is log and i gues this will act like a law-faking one so i might get a weird curve on the volume control.
Might try it anyway. Hardly expensive and might be worth a go.
Fil |
|
|
| rabbitz |
Have a look at this page on the Leach amp and down near the bottom he shows a method of to make a lower resistance pot by connecting a resistor between input and wiper terminal. I've never tried it but someone here might be able to tell us how successful it is.
http://users.ece.gatech.edu/~mleach/lowtim/part2.html |
|
|
| filholder |
Thanks a lot,
Pretty much what i had planned to but maybe i have the wrong values.
I cant see it can do much harm (fatal last words). Just remove them if if doesnt help.
I was going to use 20k so that 20k*100k / 20k+100k equals 16.67k which sounds ideal.
Best bite the bullet as see.
Also all the bits for my 'Nuuk' active stage and psu arrived today.
Good god those 4.7uf poly caps are huge!!!
How the hell am i going to fit some of them in my cd player?
Anyway christams came early.
fil |
|
|
| Shoog |
I would sujest using the buffering both sides of the pot. This would make it a universal preamp suitable for all situations. The single power supply should cope with both easily, and the circuit is so simple and small that finding room for it shouldn't be a real issue.
First off i would see how it sounds with the pot you have and without modification - you might find everything hunky dory with the buffer added.
Shoog |
|
|
| dsavitsk |
| quote: | Originally posted by filholder
Good god those 4.7uf poly caps are huge!!! |
By the way, the quality of this part is important. It is a good place for a boutique type cap. |
|
|
| filholder |
I got some from Maplin that Nuuk had previously recomended for the input caps on my gain clone. They arent particularly cheap at Ģ4 a peice. Should do for a starting point anyway.
Fil |
|
|
| postpunk |
I'm still happy with the preamp! :)
I fitted a second discrete regulator so that the two buffers are fed separately, but did not notice any obvious differences between that and single supply.
So now I have an extra regulator. That's why I am happy to see Shoog's suggestion of putting extra buffers before the pot :)
What do you guys think, should there be a capacitor on the output of the buffer before the pot, too? Will the DC offsets add up and become too much (or whatever bad) otherwise? |
|
|
| filholder |
I thought that you normally put the buffer after the pot?
Am I wrong? |
|
|
| postpunk |
| Filholder: You are right! Normally you put it after the pot, but I want to act on Shoog's suggestion of putting a buffer in front of the pot in addition to the one which is already positioned after the pot :) Didnīt want to confuse anyone, sorry :) |
|
|
| filholder |
Sorry i should have got that, you are talking about a twin buffer situation. All makes sense now.
Thanks,
Fil |
|
|
| postpunk |
| Filholder: My limited knowledge in the English language makes it tough for me to explain what I mean in a consice way. Thanks for clarifying my post, "twin buffer" thats an excellent way of putting it :) |
|
|
| filholder |
No problems at all.
Sorry i am very new to all this so can i clarify something.
I have nearly finished the PSU and am starting to think about the actual buffer circuit.
On Nuuks diagram for the buffer circuit you have a +15 and a -15v and a ground. The ground i presme is the 0v on the psu board. What i dont understand is the ground for the input and output. Are these also connected to the power ground or do just connect together as in only the input and output go to through the buffer and the grounds just connect together.
Sorry now my english is failing me as well as my experience at this type of thing.
Fil |
|
|
| postpunk |
| All grounds are to be wired together. The ground is always common. Sometimes ppl are talking about power ground and signal ground, but that is all just about which order to connect stuff to the common ground. The grounding order of this preamp does not seem to matter though. |
|
|
| filholder |
So the signal ground wires of both input and output would be connected to the 0v line on this diagram. Actually they would be connected after the second stage of regulation but you see what i mean.
Thanks for your help by the way, i find grounding the hardest part of all this stuff, seem to get confusing very quickly.
FIl |
|
|
| postpunk |
| Yes! That's correct ;) |
|
|
| filholder |
Thanks very much,
i think you just saved me a few problems!! |
|
|
| Shoog |
Good to seeyour making progress.
I personally am not certain you will get much of a benefit from the buffer before the POT, I think there is a potential for it to sound just as good without it. I would incorperate some simple method of bypassing it to see which you prefer. I for one would be interested in the result as I was contemplating this myself but never got around to it.
Shoog |
|
|
| filholder |
Hi Ahoog,
Do you mean bypassing the pot with the resistors to lower the resistance? I will defintely try this in the next couple of days. Been very busy at work so time on the project is limited at the moment. However Nuuks PSu was very easy to build, did most of it in a night. Just need to test out the output from the rectifier before i connect it to the cascade of caps i have already wired up.
Fil |
|
|
| Shoog |
No I mean bypass the first buffer before the pot. Adding in the buffer before the pot also adds in an extra cap in the signal path - you may just find that the improvement gained by having the first buffer is then waisted in going through the extra cap. Theres no easy way of knowing without actually trying it in both configeration, therfore I advocate a bypass for the first buffer so that you can make an empirical judgment for yourself.
Don't take anyones word for the benefits to be achieved from one implementation over another. Their experience and taste may be completely different from your own. I would value your own experience above all.
Shoog |
|
|
| dsavitsk |
A couple of notes on Nuuk's buffer:
1) I built mine with all Kiwame carbon resistors. I didn't think much of it at the time, but I recently built another with all Holco and PRP metal films. The metal film version sounds horrible. Unlistenably sterile. The Kiwame's give the circuit a warmth and presence that is missing with metal film resistors.
2) I added 0.22 uF Wimas at the + and - voltage input points and it added a good punch to the bass (fwiw, my PS is a welborne ps1 with the 0.1 film caps replaced by 4.7uF ones).
-d |
|
|
| Nuuk |
| I have noted similar changes to the sound when using carbon film resistors with Gainclones! I am just finishing another GC variation at present then I will turn my attention back to the buffer and try those variations myself. ;) |
|
|
| filholder |
Hi Nuuk,
Can i just ask you on your PSU design which out of UAR and LAR are the N and P resistors.
Sorry i trouble you just couldnt work it out last night, but then i had had a drink or two so that isnt saying too much.
Thanks,
Fil |
|
|
| Upupa Epops |
| " Warmth of carbon resistors, sterility of metal ones " - all is about distortion, guys :D and at any music you can't find it prescribed :cool: . |
|
|
| Nuuk |
| quote: | | Can i just ask you on your PSU design which out of UAR and LAR are the N and P resistors. |
N is the UAR and P is the LAR. My apologies for the confusion and I have now updated the information on DD to make it clearer. ;) |
|
|
| postpunk |
dsavitsk; I just tried, like you described, small caps on the+/- V inputs. I used 1uf low impedance electrolytics. The sound got less hard on my ears :) Thanks for the idea!
Upupa; Do you think that the buffer is a good design? I asked a Swedish engineer and he thought that it at least SHOULD have low distortion. Thanks for you input, it is much valued. |
|
|
| filholder |
Cheers Nuuk,
Thanks for that.
P.S.
I like the new colour sceme at DD much better now, the old one was a bit vibrant to look at for extended periods!!!
Thanks,
Fil |
|
|
| Nuuk |
| quote: | | 2) I added 0.22 uF Wimas at the + and - voltage input points and it added a good punch to the bass (fwiw, my PS is a welborne ps1 with the 0.1 film caps replaced by 4.7uF ones). |
We'd better make it clear - are these in series or to ground? :att'n: |
|
|
| Nuuk |
| quote: | | I like the new colour sceme at DD much better now, the old one was a bit vibrant to look at for extended periods!!! |
I always aim to please Fil. And of course I have to cater for those of you who like a drink or two while viewing the site. :D |
|
|
| postpunk |
| Nuuk; mine, at least, are connected to ground. |
|
|
| Upupa Epops |
| To postpunk : Good designed buffer with ultralow distortion is good thing. If have high input and low output impedance, you can find many applications, where you can use it. :cool: |
|
|
| postpunk |
| Upupa; Yeah, but how do you like the design of this particular buffer with the discrete regulators? :) |
|
|
| postpunk |
Adding small value capacitors after the discrete regulators.. why does that make the amp sound audible different? Can that be, from a technical point of view?
Perhaps I have made a mistake somewhere?
I mean, if hearable differences arise from such a small(?) change, there must be distortion involved?
This distortion issue worries me. If I had an oscilloscope, would it be easy to measure if there are big differences after adding caps and such? I can't say that I dare to trust my ears. |
|
|
| filholder |
Hi Nuuk,
Having nearly finished the main board of your PSU design i am about to start building the second stages of regulation. I am going for the simpler repeat design but will probablyt go with twin second stages ie one for each channel. You talk about adding some capacitance on this second psu but dont really say where. Would it be before or after the regulators or both. And should these be bypassed as well?
So for the constant questions, i really am very new to all this.
Thanks very much for all yours help,
Phil |
|
|
| postpunk |
I am not Nuuk, but the caps should be added before the second regulators. I think that Nuuk (?) would have bypassed them, that seems to be common practice.
Edit: OTH he does not write on the page about bypassing them. It is probably unnecessary to do that before a regulator. |
|
|
| filholder |
Hi postpunk,
Thanks, what sort of values should i use? I have got lots of duorexii 100uf and some 1000uf as well would these be any good. Also what value of bypass 1.0uf then 0.1uf sound any good. And like you said all this is before the regulator and nothing after is that right?
fil |
|
|
| postpunk |
Of your available values I think that you should use 1000uf before the regulators, and use the 1uf and 0.1uf after the regulators, like in the first stage.
Sorry I was not clear, donīt bypass the electrolytics before the regulator, after the regulator you put the filmcaps 1uf, 0.1uf. Like in the first stage. |
|
|
| filholder |
Ok so no bypass caps.
Fair enough.
edit.
Ok that sounds fine. Thanks for the help.
Cheers,
Fil |
|
|
| Nuuk |
Sorry for the delay in replying - I have been wall-papering and completing my discrete regulated IGC! (Don't tell Carlos :eek: )
10uF is plenty to put on the inputs of the second regulator stage(s). That is 10uF from rail to ground and no need to bypass them IMHO. ;) |
|
|
|
