I also got a tiny ripple on the rising edge of another very different amplifier, just like your 20kHz sq is showing.
I could not hear the effect of that tiny ripple, so I left it alone and continue to listen to it.
I could not hear the effect of that tiny ripple, so I left it alone and continue to listen to it.
You have just confirmed my findings about R14 ( in my case it took me a little while to figure it out...) . You should probably at least put a small value (2.7 ohms) like I did to have this ground lifted from power gnd...
I believe that an input cap has been reported more "harmful" to sound than a properly chosen NFB blocking cap...
Fab
I suppose I could try a smaller value resistor in R14 but with the 10R in there that oscillation is instant once both channels are connected to the source. Paulo's fix is to send an extra wire directly to the star so there is no lifting there.
The schematic that Paulo posted says to use a link for C1 if the source has no DC and to use 1-5uf if it is present. No where does it suggest bypassing C9. I also see that he uses R14 to lift the ground. The extra wire to star must have been a fix he came up with later. There is no mention of it on the schematic. I'm just going to do away with R14 and call it a day. I don't really like extra wires running all over the case.
Blessings, Terry
The schematic that Paulo posted says to use a link for C1 if the source has no DC and to use 1-5uf if it is present. No where does it suggest bypassing C9. I also see that he uses R14 to lift the ground. The extra wire to star must have been a fix he came up with later. There is no mention of it on the schematic. I'm just going to do away with R14 and call it a day. I don't really like extra wires running all over the case.
Blessings, Terry
The extra wire is standard by Greg Ball's instructions.
In a stereo build (not monoblock or dual mono) Greg also advises the use of a 100nF polyester capacitor across R14.
If you want to keep C9 and minimize it's influence in the sound you should also use C1 and fully AC couple the amp. Choose a value that gives a time constant sqrt(2) times greater for the NFB capacitor, so C9 sees almost no AC signal across it. Of course, with a C1 fitted now you have to worry about it's influence in the sound. A good quality polypropylen should not be a problem (Wima MKP10 for example).
In a stereo build (not monoblock or dual mono) Greg also advises the use of a 100nF polyester capacitor across R14.
If you want to keep C9 and minimize it's influence in the sound you should also use C1 and fully AC couple the amp. Choose a value that gives a time constant sqrt(2) times greater for the NFB capacitor, so C9 sees almost no AC signal across it. Of course, with a C1 fitted now you have to worry about it's influence in the sound. A good quality polypropylen should not be a problem (Wima MKP10 for example).
Last edited:
Nice of him to leave out that little tid-bit when he released it to public domain. No mention of that on the schematic and don't even show up asking questions on his website if you didn't buy from him. I replaced a lot of parts back when I built my first one because of that ghost oscillation. On the schematic it shows only R14 going to PSG. The signal ground is shown tying to NFB. I will have to look back through the thread but I don't remember any of the guys who had built amps from Greg mentioning that ground wire back then.
I select a much lower corner frequency with the NFB cap ( increase value) then with a decent low DC offset source when listening to real music ( with no subsonic content) I can leave out the input cap then prevents it's effect on the sound...
Fab
That gives a mixed AC & DC coupling.
Output offset often varies with volume setting.
And the DC coupling of the input and output still puts the speaker at risk.
However, one does not need to DC block the input signal twice.
If the Source has an output DC block, then the amplifier does not need to duplicate the DC block @ it's input.
Member
Joined 2009
Paid Member
I think he regrets releasing the design. I posted picies of my nice build into his forum and when I had trouble with it he was helpful. When I posted piccies of the working units looking very sexy - he deleted my posts 🙄
As I read it he only released the original circuit.
Some took that to mean they could divulge all the newer tweaks that had not been released, but were reserved for real customers.
Some took that to mean they could divulge all the newer tweaks that had not been released, but were reserved for real customers.
There are no new tweaks. The mods were old obsolete tweaks and thats why they don't show in the released schematics. I urge people to use the schematic as Greg reseased it. There are NO new improvements kept in secret.
So...
It depends on the circuit...and even so the level is minimal and well below 100mv which is accepted DC at power amp output. In my case always below 5 milivolts.
I have a DC protect circuit at output of power amp. The reliability of preamp or sources is better than power amp anyway and still we normally do not use power amp output cap ( when not mandatory)
Fab
When I already finished SKA, It's time to try next one...Today I checked both channels, and It´s working ! 🙂 Great work Jason... now I have to build proper PSU...
edit: sorry guys..I am in wrong thread 🙂
edit: sorry guys..I am in wrong thread 🙂
Attachments
Last edited:
Hi Guys, can anyone clarify something for me please.
I understand that no electroincs sounds perfect and anyone can criticise anything if they want to based on their own previous experience or expectations or whatever.
I read someone saying about the GB150 that "If only it had an ounce more detail it would be heaven"
The flip side is that the vast majority of people rave about this amp.
These days my preference for sound leans towards clean and clear. I've been enjoying my JLH 1969 with SanKen output transistors. To me this sounds very clean, and the depth I can hear into the recording is great. Of course this is not a perfect amp, but having heard what this simple JLH can do, it would be difficult to go to an amp with less detail. However I would apreciate more power so the dual rail JLH amp or JLH for ESL is on the cards for construction.
The GB150 has now caught my eye and seems like it also has a reasonably short signal path, so I am on the fence. Is anyone able to elaborate on the sound of the GB150 in comparison to the JLH (any version) in regards to fine detail? I have personally observed that lower power amps tend to sound nicer at lower volume than high power amps playing at lower volume. Does the GB150 need to be playing loud to sound best? how does it sound at low volume?
I have only heard one amp using Mosfet outputs before, and I found it to sound "smooth" or "relaxed" for want of better words. Is this a characteristic of Mosfet output amps? What do people think about Mosfet outputs vs BJT outputs? Can generalisations be made about the sound? Is Mosfet sound perhaps "smoother" than a BJT?
I understand that no electroincs sounds perfect and anyone can criticise anything if they want to based on their own previous experience or expectations or whatever.
I read someone saying about the GB150 that "If only it had an ounce more detail it would be heaven"
The flip side is that the vast majority of people rave about this amp.
These days my preference for sound leans towards clean and clear. I've been enjoying my JLH 1969 with SanKen output transistors. To me this sounds very clean, and the depth I can hear into the recording is great. Of course this is not a perfect amp, but having heard what this simple JLH can do, it would be difficult to go to an amp with less detail. However I would apreciate more power so the dual rail JLH amp or JLH for ESL is on the cards for construction.
The GB150 has now caught my eye and seems like it also has a reasonably short signal path, so I am on the fence. Is anyone able to elaborate on the sound of the GB150 in comparison to the JLH (any version) in regards to fine detail? I have personally observed that lower power amps tend to sound nicer at lower volume than high power amps playing at lower volume. Does the GB150 need to be playing loud to sound best? how does it sound at low volume?
I have only heard one amp using Mosfet outputs before, and I found it to sound "smooth" or "relaxed" for want of better words. Is this a characteristic of Mosfet output amps? What do people think about Mosfet outputs vs BJT outputs? Can generalisations be made about the sound? Is Mosfet sound perhaps "smoother" than a BJT?
Last edited:
I have a dual mono DC coupled GB 150 builg from Gregs boards, and also several other amps regarded as very detailed, Pass F5, and Aleph J. I find the SKA to be plenty detailed, but I also have the bias turned up a bit to 1.2v. I found that the SQ seemed to improve when I increased the bias. I highly recommend it, sounds great, lots of grunt.
PJN
PJN