| BobK |
I have recently completed the attached prototype of what I guess can be called a 3way monoblock Gainclone. It is a chassis that has a 3way active crossover with 3 of Brian's Gainclone boards.
The attached picture is before I completed the final grounding scheme. I am using this prototype with a open baffle speaker design I have been working on for some time. This design has a load impedence of less than 4 ohms in both the woofer and midrange passband. I find that the GC's run out of power on demanding rock concert DVD's but otherwise sound fantastic. I know that the GC is not the best choice on driving loads below 4 ohms but the sound of this configuration is so encouraging I would like to explore all posibilities.
I am in the process of installing a higher power Torroid but am curious if anyone has had any experience with using larger capacitors between the rectifier board assembly and Brian's gainclone boards?
I'm aware of the minimal design approach of the gainclone but am also aware of the fact that under dynamic demands the 1 to 1.5Kuf capacitors on these boards will definitely not be able to deliver the instantaneous requirements of some program material/speaker efficiency demands.
Are these requirements best met by the rectifier board assembely and its associated interconnect wiring and all of it's reactive impedence considerations, or could this dynamic requirement be met better by a capacitor bank between the rectifier assembly and the GC board?
BobK |
|
|
| carlmart |
| quote: | Originally posted by BobK
I'm aware of the minimal design approach of the gainclone but am also aware of the fact that under dynamic demands the 1 to 1.5Kuf capacitors on these boards will definitely not be able to deliver the instantaneous requirements of some program material/speaker efficiency demands.
|
That should be easily solved if you using a three-way active approach. Simply add more external capacitance on the module going to the bass unit. Up to 10000uF should be finely handled by the diodes and would provide a good punch control.
A 200 to 300VA toroid should be enough for dynamics and should not let the units run out of power.
Carlos |
|
|
| Nuuk |
I understand that the 'best' way to add capacitance for a GC is to use a regulated supply. I am currently building one with 10,000 uF caps and LM338 regulators.
You can get more information on Pedja's web site . ;) |
|
|
| BobK |
I have 250VA transformers on order and was also considering 10kuf's on the bass amplifier. This would allow the reserve energy for the low end demands but keep the mid and highend performance everyone likes about the GC.
Pedja's postings about regulated supplies are interesting, I might give one a try. In the past I have not been impressed with the sound of amplifiers with regulated supplies but they have been evaluated running full range. |
|
|
| tbla |
| quote: | | In the past I have not been impressed with the sound of amplifiers with regulated supplies |
may i ask - what amps did you listen to....? |
|
|
| BobK |
| It's been a number of years ago that I listened to a regulated supply amp, and am embarrased to say I don't recall the commercial amplifier we auditioned. One unit was a heavily modified Dynaco 400 than one of my friends built. The unit tested very well but sonicly was somewhat disappointing. |
|
|
| carlosfm |
| quote: | Originally posted by Nuuk
I understand that the 'best' way to add capacitance for a GC is to use a regulated supply. I am currently building one with 10,000 uF caps and LM338 regulators.
|
I made a regulated PSU for my sub (two paralleled OPA549s) with LM338 and the result is fantastic.;)
Yes, with 10,000uf before the regulators.
A few days ago I was given a nice box.
I'm also scavenging an old amp, that has a (apparently) good trafo, but voltage is +/- 31v.
So, by coincidence, I'm just starting a regulated PSU with LM338 for another GC... NOW!:D |
|
|
| BobK |
| Carlos, good to hear you've had good success with your regulated supplies. Again, the experience I've had is dated, and was always evaluated full range. I could see advantages in a sub-amp application. In my design the three way unit I posted will be used with a sub, so the woofer amplifier will be used from 80 to 300Hz. |
|
|
| carlmart |
| quote: | Originally posted by Nuuk
I understand that the 'best' way to add capacitance for a GC is to use a regulated supply. I am currently building one with 10,000 uF caps and LM338 regulators.
|
That's not necessarily the best way, and I would probably not think of a regulator as a way to add capacitance. Perhaps a capacitance multiplier might be that.
The only thing in common between capacitors and regulators is that both lower the impedance, but the former does so in just one area.
A regulator should have other more important chores, like keeping a constant voltage no matter the voltage/current demands, lower and flatten the output impedance of the supply, lowering noise, etc.
The problem is that to do that they imprint their mark on the sound, because most use feedback to do so. It maybe a matter of taste also, so it's a debatable question whether a regulated amp sounds best than a properly designed one.
By properly designed I mean the transformer should be large enough and the capacitance attend the demands of the sound. On the tests I have done it became a taste option, if you liked good and controlled bass, whether to use up to 1500uF caps for better mid and highs, or more than that for better lower frequencies. In my opinion you should power the speakers separately.
There's an important tip though: a regulator should have better specs than the amp it's powering. Like higher bandwidth. In fact a regulator IS an amplifier.
Pedja's regulator, using an LM338, should be a very good place to start from.
Carlos |
|
|
| Peter Daniel |
I would say the best way to add capacitance is to use batteries.
It seems like the batteries form a sort of buffer and even when charging circuit is connected the sound is not that bad. It is much worse when only charger is powering the amp (without batteries attached);) |
|
|
| carlosfm |
| Like an alternator in a car doesn't make the sound much worse.:D |
|
|
| carlosfm |
| quote: | Originally posted by carlosfm
I'm just starting a regulated PSU with LM338 for another GC... NOW!:D |
This afternoon I :smash: :smash: :smash: and it's done.:cool:
It took a little longer, I had to open the trafo, it was a CT.:bawling:
Regs with trimmer pots, adjusted for EXACTLY +/- 31.0 volts.:cool:
PSU is done, my modules with LM3886 are done for months, now to finish the box, some days.
Let's see if this thing is even better than my GC power amp, on my main system.
Later I'll report if it's :up: or :down: .
:D |
|
|
| Pjotr |
| quote: | Originally posted by carlmart
............
There's an important tip though: a regulator should have better specs than the amp it's powering. Like higher bandwidth. In fact a regulator IS an amplifier. |
Hi,
Indeed a regulator IS an amp. But there are some major differences over an audio amp. A regulator is usually loaded with a big capacitor. This will give it low impedance at higher frequencies relaxing bandwidth requirements, but anyway GBP can be made easily 100MHz (without the loading cap) when build with discrete transistors. Second, a linear regulator is always running in class-A.
What is also important is its transient behaviour. A regulator feeding an audio amp should have no overshoot and ringing artefacts. Most IC regulators do have that however.
IMHO such a regulator should have constant output impedance from DC to the end of the band the connected amplifier is used, to be “invisible”. At low frequencies the regulation action of the regulator takes care of that and at higher frequencies the loading capacitor will do. Building a good regulator that have no imprint on the audio is not an easy task.
On the other hand the small smoothing cap in the original Gaincard PSU is part of its flavour :D
Cheers ;) |
|
|
| miguel2 |
| quote: | | considering 10kuf's on the bass amplifier |
Are you using different diode bridges for the sub amp with the same tx? Using only 1 tx a second bridge must be used to have not 10000u in the GCs.
Miguel |
|
|
| BobK |
| Miguel, I had hoped to use only one transformer and one bridge for each amplifier channel. Each channel would have three GC's (woofer, midrange, tweeter) |
|
|
| carlosfm |
It's all mounted and tested, without the box yet.
The whole thing has been :hot:-in since last night.
It's a bunch of aligators.:D
Gonna test this tonight on my main system to see what it's worth.
Stay tuned.;)
I've got a feeling...:) |
|
|
| carlosfm |
Summarizing, the only difference from this GC (let's call it RGC - regulated, with LM338s) and my power-amp on my main system (let's call it URGC) is really the PSU.
Both are BIGCs, on my PCBs, with OPA627 as buffer.
The amp modules are exactly the same: same caps, same regs for the 627s, same everything.
Both use MUR860 as rectifiers, on my small PCBs.
My RGC has 4,700uf caps on the rectifier board, and another 2,200 uf on the regulator board.
The URGC just has 1,000uf.
Results?
The RGC has very goof bass, as I expected from the results i've got on my sub.
The rest... ****!:bawling:
That midband magic disappeared, that treble got muted, that impressive transient speed of my URGC is gone, detail is gone, harmonics disappeared. :eek: :xeye:
This seams like a normal amp.:bawling: :bawling: :whazzat:
Hi-Fi, not high-end.:att'n:
I don't believe it's the **** test wires and connectors, because they are the same in every amp I make.:dodgy:
I'm gonna keep this thing :hot:-in some days on my bench, and then test again.
I'd like to hear other people's oppinions.
Nuuk, have you made the PSU?:bawling: |
|
|
| Nuuk |
Carlos, I was advised to remove the 100 uF's from the chip pins and use 100 uF instead.
Also you need to have the regs as close as possible to the chips and those 100 uFs act as the output caps for the LM338's.
All details from the totally brilliant and hard-working Pedja who also finds time to study philosophy as well! :D How does he do it? :xeye: |
|
|
| carlosfm |
| quote: | Originally posted by Nuuk
Carlos, I was advised to remove the 100 uF's from the chip pins and use 100 uF instead.
|
You mean replace the 1000uf for 100uf?
That makes sense to me.
I'll try that, I have this as a bunch of wires, no problem.
Maby I'll remove the 100uf I have on the output of the LM338s
:D and (I'm reluctant to leave them with nothing:D ) put 0.1uf caps.
Yes, maby an RGC needs to be quite different.:cool: |
|
|
| Nuuk |
If you remove the 100's from the output of the regulator circuit but have the regs as close as possible to the chip pins, then they are still effectively on the the ouput of the regulators!
| quote: | | You mean replace the 1000uf for 100uf? |
Exactly Carlos - your English continues to improve wonderfully! :D |
|
|
| tbla |
dear carlos.....:)
put the regs as close as possible to the powerchips and use a 470uF at the regs output and no film - put the film on the regs input instead, bypass the adjust resistor with 100 uF to ground. 10000 uF before the reg will be nice too.....:) |
|
|
| Pedja |
Hi Carlos,
When I tried the LM338 for the first time the set-up was:
- two 4700uF caps per rail;
- LM338s physically mounted to the second of these caps;
- LM338s adj pins were bypassed by WIMA MKS 1uF caps;
- other than local caps mounted to the LM3875s, no additional caps after the regulators were used;
- regulators were relatively close to the power IC’s;
- for the start I have used usual 1000uF at the LM3875’s pins.
I liked it instantaneously, without any :hot:-in. I can not remember if :hot:-in brought any improvement or not, so if it id, it was not huge.
Usage of lower value, higher quality (BG) local LM3875 caps was quite meaningful.
Nick, that part with philosophy actually happened years ago. ;)
Pedja |
|
|
| carlosfm |
Guys, let me tell you, this is special.:)
It's getting late, I can't turn up the volume as I wish to, but what I hear now is that magic returned.:nod:
More auditions tomorrow when I arrive from work.
Here's what I did:
- removed the 100uf cap at the reg's output and put small 47nf polyesters.
- Removed the 1000uf caps at the chip's PCBs and put 100uf.
S' beautiful.:bawling:
I'm gonna let it :hot:-in 'till tomorrow.
Doesn't matter if it does a difference or not, at least the new electrolythics need it.
I'm now :drink: -in' a pure malt to Nuuk and Pedja.:grouphug: :D
Gotta go listen to Keith Jarrett on vinyl.:up: :cool:
BTW Pedja, I have the Adj. resistor bypassed with 47uf caps.:devilr:
And protection diodes. |
|
|
| carlmart |
| quote: | Originally posted by carlosfm
BTW Pedja, I have the Adj. resistor bypassed with 47uf caps.:devilr:
And protection diodes. |
The norm for these regulators is probably the same as for LM3X7s: small cap at the output (up to 150uF at most) and big cap on adj. resistor. Protection diodes are a must if bypassing with electrolytics.
Also small caps close to gainclone, I'd say 47uF Pana FC. Rowland's seems to use SMD 'lytic, which value I can't say, but it's small. Perhaps 4.7uF or so.
Carlos (the other) |
|
|
| BobK |
Carlos, Very interesting results. Since I have limited time right now to prototype, I have to live vicariously thru every one else's success. I am now lifting a fine higher octane adult malt from Scotland on your progress.
Not that it matters, but what type of speakers are you doing your evaluation on?
BobK |
|
|
| carlosfm |
| quote: | Originally posted by BobK
Carlos, Very interesting results. Since I have limited time right now to prototype, I have to live vicariously thru every one else's success. I am now lifting a fine higher octane adult malt from Scotland on your progress.
Not that it matters, but what type of speakers are you doing your evaluation on?
BobK |
Cheers.:D
My main-system speakers, and the ones I always trust for a final test, are Epos 11. |
|
|
| carlosfm |
| quote: | Originally posted by carlmart
Also small caps close to gainclone, I'd say 47uF Pana FC. Rowland's seems to use SMD 'lytic, which value I can't say, but it's small. Perhaps 4.7uF or so.
Carlos (the other) |
Carlos, I had some 47uf caps looking at me, but I decided to take on Nuuk's suggestion for 100uf first.
Anyway, this report is not final, I'm gonna listen to this thing, fine-tune if I must and listen again.
As I always do.
Let's see the final result.
I expect to finish the whole thing (with box) this week, I hope soon to have some pics to show.:up:
Forgot to say: both amps are with LM3886 chips.:cool: |
|
|
| Nuuk |
| quote: | | Protection diodes are a must if bypassing with electrolytics. |
You beat me to it Carlos(M)! Carlos (in Lisbon) if your hangover is lifted, please take this advice ;) |
|
|
| carlosfm |
Nuuk, as I said, and as I always do with the LMs, I used protection diodes.;)
When I arrive home today I'll first listen.
Then I'll change the 100uf caps for some good Philips 33uf/40v and listen again.
Let's see...;) |
|
|
| mAJORD |
| quote: | Originally posted by Nuuk
You beat me to it Carlos(M)! Carlos (in Lisbon) if your hangover is lifted, please take this advice ;) |
why is that? :( |
|
|
| peranders |
| quote: | Originally posted by carlmart
The norm for these regulators is probably the same as for LM3X7s: small cap at the output (up to 150uF at most) and big cap on adj. resistor. Protection diodes are a must if bypassing with electrolytics. |
If you read the datasheet and the application notes carefully, these diodes are necessary if the incoming voltage decreases fast when the power is disconnected. Normally you don't need these, unless see quote below.
| quote: | when either the input or output is shorted. Internal to the
LM117 is a 50Ω resistor which limits the peak discharge
current. No protection is needed for output voltages of 25V
or less and 10µF capacitance. Figure 3 shows an LM117
with protection diodes included for use with outputs greater
than 25V and high values of output capacitance. |
|
|
|
| Nuuk |
Thanks P-A, that's what I was going to point to.
Carlos - glad to hear that you are practising safe regulation! ;) |
|
|
| KT |
Carlos,
Great to see you got the regulator to work. Please let us know what improvements and changes you hear over the next few days.
Best,
KT |
|
|
| carlosfm |
The regulator always worked.:clown:
I will report my tests, stay tuned.;) |
|
|
| carlosfm |
Left this thing playin' since last night.
Arrived home, put "Focal/JMLab - Le grand spectacle du son", turned the colume up up up and...
:bigeyes: :bigeyes: :bigeyes: :bawling:
INCREDIBLE!!!! :bawling:
Bass is amazing!:bawling:
I'm surrounded by detail!:bawling:
It's a festival.:D
Now excuse me, gotta go change those cheap 100uf caps I put last night for something better.:D
Wait... I have to gain some courage to power this thing off...:bawling:
:D |
|
|
| Nuuk |
| I'm convinced - where do I get a 'colume' for my GC? :tilt: |
|
|
| carlosfm |
...Changed the cheap :down: 100uf caps for Philips 33uf/40v.:up:
Changed the 100uf before the OPA627 regulators, on the GC board, with 10uf (they have resistors before them).
Bypassed the 2,200uf caps before the LM338s with 470nf polys.
Connected the whole bunch of aligators.
Power on...
:eek:
Does it get any better than this?:bawling:
Guys, I think I'll stop here, or else I'll put a 1pf cap and that's it.:devily:
No, seriously, this thing IS special.;)
Nuuk, are you there?
Try it.:cool:
Just a conclusion:
- Very small cap after the LM338 (I used 47nf poly), or no cap at all, if you use them very near the chips.
- From 4,700uf to 10,000uf total capacitance before the regs (I use 4,700uf on the MUR860's board and 2,200uf on the regulators' board).
- Small cap on the chip (I'd say between 22uf:eek: and 100uf).
Gotta go listen, something made "click" and I can't stop listening!:bawling: |
|
|
| carlosfm |
| quote: | Originally posted by Nuuk
I'm convinced - where do I get a 'colume' for my GC? :tilt: |
:D
I can't edit the post now...
It's 'volume', I hope my neighbours are still alive.:clown: |
|
|
| Pedja |
Well, was interesting. I have not even tried to use any caps after 338’s other than 3875’s local caps. Unfortunately, I don’t have these regs working at the moment so can not check it.
Does it get better than that? Heh, yes I’d say. ;) While I can concur with that the simpler non-f/b supply from my page showing regulated GC supplies sounds rather different than better than LM338 (though I am honestly leaning toward it), that second non-f/b supply is better.
Pedja |
|
|
| tbla |
| i guess its time for peter daniel to try the regulated supply.......;) ;) ;) ......:) |
|
|
| carlosfm |
Listening to good music 'till 2 AM...:)
A good test for transient speed, harmonics, treble detail, dynamics is two vinyl LPs I have from Keith Jarrett (they are called "Volume 1" and "Volume 2").
Very good ECM pressings, as always.
This amp is impressive to say the least.
Even with a much weaker trafo (compared to the big toroid I have on the URGC) salvaged from an old amp.
I can take the conclusion that with these chips and an unregulated PSU you always have a compromise: between 1,000uf and 1,500uf capacitance is the ideal on a sonic point of view (I made tests), but of course it's always a compromize.
With a regulated PSU you can have much more capacitance BEFORE the regs, and it pays off.;)
If only I could send you all the sound of this thing...:bawling:
:D |
|
|
| Nuuk |
Carlos, I have the Keith Jarret LP's "Colume 1" and Colume 2" :D
Now enough joking, I am going to my 'workshop' to do more work on my regulated GC! :smash: |
|
|
| carlosfm |
:cool:
Then listen to track 2 of side one of "Volume 2".;) |
|
|
| Peter Daniel |
| quote: | Originally posted by tbla
i guess its time for peter daniel to try the regulated supply.......;) ;) ;) ......:) |
How would you compare reg PS to batteries? |
|
|
| carlosfm |
Hi Peter.:wave2:
No batteries to compare, so...:dunno:
But believe me, this is special.
Try it.
Because it seams that I'm alone with Pedja on this.:bawling:
For now. ;)
Nuuk?:Popworm: |
|
|
| Nuuk |
| quote: | | Because it seams that I'm alone with Pedja on this. |
Not quite Carlos, several people to my knowledge have tried regulated supplies! ;)
I'm still working on my case. :smash: |
|
|
| carlosfm |
| quote: | Originally posted by Nuuk
Not quite Carlos, several people to my knowledge have tried regulated supplies! ;)
|
And they don't say anything?:confused:
Or it may be a case of not doing it the best way, as this thing is picky with capacitance after the regs.;)
| quote: | Originally posted by Nuuk
I'm still working on my case. :smash:
|
Much listening to do before that.
The case is the last thing I do.
I try the amps without the case, with test wires, connectors, heatsinks.
It's all prepared for R&D.:D
I can dismantle the whole thing in seconds, change a component and test again.
Forget the case for now or you may find out too late that you made some unnecessary holes:bawling:. |
|
|
| tbla |
i have never tried batteries......but i'm shure that they have some good sides as well as some less good sides.
you should try the lt1083 regs.....:) |
|
|
| tbla |
| quote: | | And they don't say anything? |
i did mention this along time ago......:)
as a side note - a few weeks ago i bought a krell fpb300, this amp also have a regulated supply for the output powersection, its a rather big and powerfull amp.
i haven't listened to it yet because its a 115 volt/60hz and the voltage conversion takes a bit longer time than i expected, 8 - 10 days and i hope its ready........almost can't wait to be blown away....;) |
|
|
| carlosfm |
Some (all?) Mark Levinson amps use regulated PSUs too.
We're achieving/surpassing this level.:yikes: :D |
|
|
| Nuuk |
Carlos, I am currently listening to the OPA627 buffered IGC that I built some time last year and it is still on a plank of wood!
The new amp is partly the result of my R&D with that GC and now I want to get it in a case rather than having another plank of wood in my rack for the next nine months?
I am using the modular design that I used for the VBIGC and (hopefully) any mods will be easy to do by removing the appropriate module.
I decided sometime ago to make this a pleasurable hobby again rather than be a slave to time schedules. This is easier to do of course when you already have a good system to listen to although that is not to say that I am not keen to hear the regulated PSU and new buffers!
'The ox is slow but the earth is patient' |
|
|
| carlosfm |
Take your time, Nuuk, I was joking.;)
I'm now gonna make the crime to power this thing off and mount it all in the box.:) |
|
|
| tbla |
| quote: | | Some (all?) Mark Levinson amps use regulated PSUs too. |
only the two top of the line models (33, 33h) have regulated powersupply for the output stages.......:att'n:
| quote: | | We're achieving/surpassing this level. |
we'll see (hear)...;)
but one area will not be surpassed - loudness, my speakers need much more power than delivered by a gainclone......:) |
|
|
| carlosfm |
| quote: | Originally posted by tbla
only the two top of the line models (33, 33h) have regulated powersupply for the output stages.......:att'n: |
So we're getting to/surpassing THAT level.:yikes: :yikes: :yikes: :D
| quote: | Originally posted by tbla
but one area will not be surpassed - loudness, my speakers need much more power than delivered by a gainclone......:) |
Regulated bridge/parallel LM3886?:hot: :redhot: :up: |
|
|
| tbla |
| quote: | | Regulated bridge/parallel LM3886? |
not a chance.........thekrellsmuchlouder.....:D |
|
|
| carlosfm |
tbla, what speakers do you have?:eek:
I've been:smash: :smash: :smash: the box.
Now it's late, let's see if tomorrow I can finish this.:xfingers:
Can't wait to :hphones: again. |
|
|
| tbla |
| i have a pair of ATC scm100sl - big 3-way with a 12", a 3" dome and 1" dome, its not difficult to drive but it needs many watts to really come alive.....and it can take it...;) |
|
|
| carlosfm |
:eek:
Yes, those speakers need a krell-type amp.:)
Two regulated bridge-parallel LM3886 amps, in bi-amp config.;)
Very good speakers you have there.:up: :nod: |
|
|
| carlosfm |
Yesterday :smash: :smash: :smash: the box.
Almost done, must finish it today.
I miss the sound of this thing.:bawling:
BTW I'm not receiving mails from this forum, are you all?:confused: |
|
|
| jarkaa |
Hi,
I´ve been listenig to gc with regulated supply for three weeks now and I am amazed of it´s performance. I have 7000uF before regulators and 100,1uF on rails and it is excellent. Even though that the gc is based on opa549.
The soundstage is very clear and especially the stereo imagining is excellent.
I am yearing to get a tube buffer in front of the amp but I am not capable of designing that myself.
The amp is NIGC
One more thigng...
I found that 2.2uF input cap for each block gave the best results for opa549.
Still waitin for a good schematic for tube buffer...
Jarno |
|
|
| Variac |
This isn't close?
He says that the conversion to non inverted is probably not big deal.
http://www.diyaudio.com/forums/show...3693#post403693
I like the tube buffer/ regulated power supply combo.
More is better- in this case it just might be!!
I'ts possible that with a buffer, the inverted sounds better than the non inverted- one more experiment:rolleyes: |
|
|
| carlosfm |
Oh yes, it's done.:)
Photos in a few days, I've just finished a film on my camera.:devily: :p
It's singin' again, and very well thank you.:up:
Photos and a new thread in a few days, be patient.
I think I'll name the thread... "High-end GC for my pre".:D |
|
|
| Dr.H |
Carlos,
Can you confirm your PS set-up? As far as I can see you have
MUR860 diodes
Some smoothing caps
LM3xx regulator for + and -
No cap after regulator
A small cap on the LM3875
Please state your cap sizes. I tried the regulated PS cometime ago, but the sound was tizzy and overwhelmed by bass. Maybe it was because I was using 1000uF on + and - legs of the chip at the same time as using the regulator.
Does the use of the regulator require less capacitance on the LM3875?
Thanks
Ryan |
|
|
| carlosfm |
Ryan, have you read all of this thread?
It's everything here.
Look at my first impressions, it was ****!
I have a small 47nf cap on the output of the regulators, and changed the 1000uf caps on the LM3886 PCBs for 33uf.:eek: :D
High-end.:cool: |
|
|
| Nuuk |
| quote: | | I have a small 47nf cap on the output of the regulators |
Carlos, do you mean on the output as in between the output pin of the reg and the cap on the input of the 3875?
Or do you mean bypassing the adjust pin?
And what is the length of the wire between the output of the 338 and the 3875?
(Yes, mine is still progressing slowly) ;) (Like a good port) |
|
|
| moving_electron |
| quote: | Originally posted by Nuuk
Carlos, do you mean on the output as in between the output pin of the reg and the cap on the input of the 3875?
Or do you mean bypassing the adjust pin?
And what is the length of the wire between the output of the 338 and the 3875?
(Yes, mine is still progressing slowly) ;) (Like a good port) |
Since the info is throughout the thread I also feel compelled to make sure I really am interpreting Carlos' implementation correctly. I am comparing the comments to the regulated power supply design on Predja's site mentioned in one of the early responses:
http://users.verat.net/~pedjarogic/...ne/supplies.htm
I interpret Carlos' comments to be that he obtained his best results with 47nF for C3 and C4 and changed the 1000uF caps on the gainclone board to 33Uf.
But I am not sure if that is correct or if he still uses the 100uF-470uF caps in Predja's design. But it sounds like he does not use the 100uf-470uF on the power supply board since there are 33uF caps on the gainclone board.
And I am pretty sure he likes the sound! :) |
|
|
| carlosfm |
| quote: | Originally posted by Nuuk
Carlos, do you mean on the output as in between the output pin of the reg and the cap on the input of the 3875?
|
:confused:
No Nuuk, I mean on the regulator, I used a 47nf cap.
Between the output pin and ground.
Check the datasheet, page 6.
I use 47nf for C1.:eek:
They usually recommend 100uf cap here, but that sounds bad to me.
Look at the photos of my amp on my thread, the amp is in a small case (22cm wide), the regs are not far from the LM3886s, let's say some 10~15cm of wire. |
|
|
| moving_electron |
| quote: | Originally posted by carlosfm
:confused:
No Nuuk, I mean on the regulator, I used a 47nf cap.
Between the output pin and ground.
Check the datasheet, page 6.
I use 47nf for C1.:eek:
They usually recommend 100uf cap here, but that sounds bad to me.
|
OK so looking at Predja's schematic a summary of changes that describes your implementation is:
- MUR860 diodes
- C1 and C2 = 47nF
- No caps for C3 and C4 although there are 33uF caps instead of the 1000uF caps at the amp chip power ins
- Variable resistor to be able to precisely set voltage
- Resistor changes to provide different voltage target than Predja's |
|
|
| carlosfm |
Completely wrong!:bawling: :bawling:
People have difficulty in reading the threads, I've explained the PSU in detail.:bawling:
Looking at Pedjas's schematic, I used 47nf for C3 and C4 , 47uf for C1 and C2 and a diode from Adj to out.
33uf on the LM3886s instead of 1000uf.
The amp's schematic is here:
http://www.diyaudio.com/forums/show...82&pagenumber=3 |
|
|
| moving_electron |
Thanks that for clarifying and summarizing.
I was not successful integrating the information interweaved across 70 posts. |
|
|
| netgeek |
According to National the CMRR is so good that regulated supplies may not be needed - which would make sense for most of their target market (in order to reduce costs). Does it really make that much difference - in the opinion of people who have tried both ways? And has anyone actually measured the effect (positive or otherwise) on the output and performance?
Regards,
Bill |
|
|
| tbla |
| quote: | | Does it really make that much difference |
YYYYYYYYEEEEEEEEESSSSSSSS it does.........and the best of the chip regulators are the LT1083 (also the most expensive)........................:) |
|
|
| Nuuk |
| quote: | | No Nuuk, I mean on the regulator, I used a 47nf cap. |
Sorry Carlos, I meant on the output between the regulator and the chip but connected to ground!
You hadn't mentioned the cap on the adjust pin. Pedja had suggested removing the cap on the output of the reg, and just using a 100 uF on the pins of the amp chip, hence I was wondering if the 47 uF talked of was on the output of the regulator or on the adjust pin. It seems that you have both!
It can get very confusing talking about circuits, especially for those that haven't built one.
I'm sure that Pedja won't mind if we put his circuit here to show exactly what we are talking about (even if some of us have used slightly different values).
BTW, do you use the second protection diode as suggested in the application notes for the LM338? I have included it on mine. ;) |
|
|
| carlosfm |
Nuuk,
Looking at Pedjas' schematic, I used 47nf polyester for C3 and C4.
I don't like to leave a regulator without a cap on the output, but that's me, and this is a very small one.
For C1/C2 I used 47uf, and a diode from Adv pin to Output pin.
I didn't use D1.
On the amp's PCBs I used 33uf instead of the 1000uf I had.
The result is to :bawling: for.
:D |
|
|
| carlosfm |
| quote: | Originally posted by tbla
YYYYYYYYEEEEEEEEESSSSSSSS it does......... |
:yes: :yes: :yes:
:up: |
|
|
| Nuuk |
Thanks Carlos, I see what you mean about keeping a cap right on the output pin and using the 47 nF to do that job!
Be careful with all that crying, you may accidently cause a short circuit and then you will have nothing to do accept to go to the football! ;) |
|
|
| janneman |
I looked at the schematic above, is there a reason why the transformer is not used in CT mode, thus saving 4 rectiifer diodes?
Just curious.
Jan Didden |
|
|
| carlosfm |
Hi Jan,
The reason is simple, we're using two positive voltage regs (LM338) for both rails.
As you can see, we join the G and + of the two regs at the output to make the ground (the other extremes are + and -).
I can't see other way to do this other than with independent secondaries and independent bridges.
The trafo I used for this amp was scavenged from an old amp and was a CT.
I opened it and separated the wires, transformed it for independent secondaries.
It's always a thing that people can do easily. |
|
|
| janneman |
Ahh yes, I see. I thought there would be a complement to the 338, apparently not.
BTW looking to the data sheet, it seems that the 338 is better than the 1086 at higher frequencies (ripple rejection, output impedance). I didn't know the 338, seems like a good choice here.
Jan Didden |
|
|
| NealG |
| quote: | | Does it really make that much difference - in the opinion of people who have tried both ways? And has anyone actually measured the effect (positive or otherwise) on the output and performance? |
Yes sure does!
Going from memory with the non-regulated supply the + and - supply rails sag when the amp is under load. At about 1W o/p you can observe the supply rails dropping on a'scope and also supply rail modulation, can't recal the exact figures and I 'aint going to pull it apart to check but it's very noticable.
With the regulated supply there is no disernable sag even at full o/p under sine wave test and at 1W o/p there is hardly any modulation of the rails at all. At bass frequencies this supply can o/p full power for quite some time, I had to limit measurment due to the poor heatsinks I have on the 338's, when un-regulated the lines sag a huge amount I *think* I measured a drop from 26v nominal to 18v but don't hold me to that, I did not take notes. |
|
|
| carlmart |
| quote: | Originally posted by NealG
Going from memory with the non-regulated supply the + and - supply rails sag when the amp is under load. At about 1W o/p you can observe the supply rails dropping on a'scope and also supply rail modulation, can't recal the exact figures and I 'aint going to pull it apart to check but it's very noticable. |
Sorry to disagree on this point. If your AC is stable and connections are all fine, the main reason why supply rails will sag under load is because the transformer is too small. Going to a higher VA transformer will solve the problem or quite diminish it.
Supply modulation or ripple is directly related to capacitance, as well as capacitors type. If grounding is properly done and wired, higher capacitance will improve ripple.
Supply droping at 1W is certainly an indication of a small transformer. A transformer should only sag at the amp's highest output under heavier loads.
| quote: | Originally posted by NealG
With the regulated supply there is no disernable sag even at full o/p under sine wave test and at 1W o/p there is hardly any modulation of the rails at all. |
At a price, of course. You will be always wasting quite a few volts to regulate from input to output. You will also have to compute differences your AC line might have at some time in the day or week and add them to those input volts.
| quote: | Originally posted by NealG
At bass frequencies this supply can o/p full power for quite some time, I had to limit measurment due to the poor heatsinks I have on the 338's, when un-regulated the lines sag a huge amount I *think* I measured a drop from 26v nominal to 18v but don't hold me to that, I did not take notes. |
That's quite a huge drop! My advice would be to change your transformer, even if using a regulator.
Regulating or not regulating is a question of budget and taste. A properly designed unregulated supply should show little difference with a regulated one. If the difference is huge then there will be something wrong in the unregulated one. Of course that "little difference" between properly designed supplies might be huge to some, which I consider more a matter of taste.
It's a question of taste because some people do not like the sound of regulators, particularly those using feedback to regulate.
But I do agree that regulators are a shortcut to some problems, like lowering impedance and improving ripple.
Carlos |
|
|
| carlosfm |
Carlos, the problem here is that these chips have the best sound quality with 1000 to 1500uf capacitance, with an unregulated PSU.
I made extensive tests with capacitance.
This is a big compromise.
I didn't make measurements but I can tell you that my unregulated GC has a much bigger transformer (2x24v, 384 VA) compared to the normal trafo I used on my regulated GC, much smaller (scavenged from an old amp).
My speakers (Epos 11) are difficult to drive properly, they need power, so I have good conditions to compare.
If you heard the difference, even on a blind test, you would say when switching for the regulated GC:
Oh yes, this is the Krell.:D
Bass is much better, tight like there's no tomorrow, and that brings improvements in midband, treble, soundstage, everything, as always.
It's like putting a good clock on a "jittery" CDP.:clown: |
|
|
| tbla |
| quote: | | Oh yes, this is the Krell. |
i'll tell you in a week.........;) |
|
|
| carlosfm |
tbla, you mean you haven't made it yet?
I thought you had a regulated PSU/GC with LT1083.
Not yet? |
|
|
| NealG |
Hi Carlos, don't be sorry to disagree!
Curiosity go the better of me and I have just gone back to un regulated to check. I have a 300VA mains TX, I was way out on the voltage sag, it's no where as bad as I remembered, I have just measured a 8% approx. drop at full o/p before clip at 1KHZ and approx. 12% drop at 100HZ, this is with just 2000uF per rail setup favoured by some.
At 100HZ I seen to have LF instability and will have to look into that one later! At 1KHZ I see the o/p wave form on the supply rail at approx 250mV and if I increase capacitance to 10,000uF / rail it drops to approx 50mV at full o/p. Time to go back to regulated and try again! |
|
|
| tbla |
| quote: | | tbla, you mean you haven't made it yet? |
i was referring to my krell fbp300......its not playing yet, it will be converted to 230V/50hz next friday:cool: :cool: :cool: i hope......;)
and then i will compare it to the triamped 12 x lt1083 regulated 6 x 300VA 12 x 10000Uf/63V cerafine bla bla bla supersystem;) ;) ;) ;) .......... |
|
|
| carlosfm |
| :bigeyes: :bigeyes: :eek: :yikes: |
|
|
| chris ma |
| quote: | Originally posted by tbla
i was referring to my krell fbp300......its not playing yet, it will be converted to 230V/50hz next friday:cool: :cool: :cool: i hope......;)
and then i will compare it to the triamped 12 x lt1083 regulated 6 x 300VA 12 x 10000Uf/63V cerafine bla bla bla supersystem;) ;) ;) ;) .......... |
I think it is not easy to compare Tri-amp system with another system. That will not be a amp to amp compare. I had similar situation once but bi-amp and the closest I've done was to replace a set of the amps that drive the mid-high and compare, but ideally I wish I have the same number of amps each side to really compare them within a particular system. So how are you planning on with this compare? I'll be interested to see what other options are there?
Regards,
Chris |
|
|
| carlosfm |
Chris, tbla is going to compare a brutal amp with another one, DIY, with chip amps and much lower power (around 50 watts RMS).
Triamp, in fact, but interesting, no?:cool: |
|
|
| carlmart |
| quote: | Originally posted by NealG
Curiosity go the better of me and I have just gone back to un regulated to check. I have a 300VA mains TX, I was way out on the voltage sag, it's no where as bad as I remembered, I have just measured a 8% approx. drop at full o/p before clip at 1KHZ and approx. 12% drop at 100HZ, this is with just 2000uF per rail setup favoured by some. |
Now that seems about right. A larger transformer may probably sag even less, which seems to be the advantage of using them. My guess is maybe yours are not toroidal, as my experience with sagging transformers were usually EI types, smaller ones.
The lack of control some have told about at low frequencies, which I could myself listen on the tests I did, seem to be related to smaller capacitors. That's why I thought this arrangement using 1000uF capacitors was better for active bi-amping, powering mid/tweeter. Using larger caps on woofer side should provide the best of both worlds.
| quote: | Originally posted by NealG
At 100HZ I seen to have LF instability and will have to look into that one later! At 1KHZ I see the o/p wave form on the supply rail at approx 250mV and if I increase capacitance to 10,000uF / rail it drops to approx 50mV at full o/p. Time to go back to regulated and try again! |
Regulators act like isolating filters, which in your case the bigger cap seems to be doing. Larger capacitors store the charge better, allowing smaller transformers work less.
But once I read that the best way to deal with Class B amp's supplies was to deal with them as if they were class A. Big transformers, big capacitors. I think that's about right.
Pity there's no way to modify the idle current setting in the 3875/3886, as that might bring further improvements.
Please pay attention that I never said that an unregulated amp should sound better or worst than a regulated one, or viceversa. They are just different.
Carlos |
|
|
| carlosfm |
| quote: | Originally posted by carlmart
Pity there's no way to modify the idle current setting in the 3875/3886, as that might bring further improvements.
Carlos |
You can treat them as op-amps, wich is what they are.
A resistor from V- to output.:D
But some people have tested that and it seams that in this case there's no advantage. |
|
|
| carlmart |
| quote: | Originally posted by carlosfm
You can treat them as op-amps, wich is what they are.
A resistor from V- to output.:D
But some people have tested that and it seams that in this case there's no advantage. |
Then it's not the same. Using the "class A trick" on chips never sounded worst, as some reported here for the GC.
Carlos |
|
|
| NealG |
Hi Carlos,
The transformer is a toroidal one!!
I am now back with regulated supply and large heatsinks on the 338's. :D First thing I noticed was large oscillations on both power rails with the negitive rail voltage dropping from 25V to 22V at full o/p 1KHZ sine wave. :bawling:
I had not measured under load before so did not notice this, it works out that the .1uF caps I used on the o/p of the 338's where causing it, removing them the oscillations went away and increasing to .47uF they also went away so there must be some interation causing a resonant circuit somewhere, I am now at .47uF on the 338 o/p and all is well. :angel:
This regulated supply can supply full power at 50HZ with no supply rail drop, for me with 25v rails it's about 35W :cool: and sounding great! |
|
|
| carlosfm |
| quote: | Originally posted by NealG
This regulated supply can supply full power at 50HZ with no supply rail drop, for me with 25v rails it's about 35W :cool: and sounding great! |
:up: :drink: :yes: |
|
|
| carlosfm |
| quote: | Originally posted by carlmart
Then it's not the same. Using the "class A trick" on chips never sounded worst, as some reported here for the GC.
Carlos |
It may be me, but I have doubts of the advantage of puting an unregulated supply with lots of ripple on the output of the chip.
:rolleyes: |
|
|
| carlmart |
| quote: | Originally posted by carlosfm
It may be me, but I have doubts of the advantage of puting an unregulated supply with lots of ripple on the output of the chip.
:rolleyes: |
Who says an unregulated supply has lots of ripple? It certainly will not if capacitance is enough. The minimum capacitance can be computed to have minimum ripple according to the current you will be demanding. There's a formula to get that number.
Carlos |
|
|
|
