RJM Audio Sapphire Desktop Headphone Amplifier

i'm confused with your wiring to the volume pot and phono sockets. it looks to me like you have run the GND to the input (-) bypassing the volume pot?

the GND of the volume pot is (-) not GND.

volume pot:
the phono socket (+) goes to the Pot IN.
the phono socket (-) goes to the Pot GND.
the (+) OUT from the Pot goes to the (+) IN on the Sapphire.
the GND from the Pot goes to the (-) IN on the Sapphire.

i had better results substituting the ''Module'' OA_GND for the (-) IN to the Sapphire.

no connection to GND on the board from the volume pot or case.

Richard has had best results using the Headphone socket GND to case as the board GND.
 
I think it's confusing because I have wires coming out of the volume pot. So Pot GND is actually a wire, not a point on the pot.

I've connected together Pot GND, RCA GND, and (-) IN of the board. They are soldered and I've put some Heat shrink tube.

Here a closer look, you can see three black wires connected together :
An externally hosted image should be here but it was not working when we last tested it.


I didn't understand the module GND part sorry.
 
i see what you have done now, thats fine.

there is a box mark on the silkscreen around the opamp, in the box there are large holes (6 of) for plugging in a discrete opamp. one of the large holes is marked OA_GND. i connected the IN(-) to that instead. 99% of people wont have to do what i done but it helped my ground 'hum' problem as i'm plugging a Raspberry Pi straight into the IN's of the Sapphire, so no volume pot.
 
Ok got it.
Well I've pluuged it to my dac and I have no Hum, while th gound is not connected to chassis, so I guess I'll leave it at that.

Now I have another problem, but a mecanical one for change : I can't fix the volume knob to the pot, the little screw might be damaged...

Apart from that everything is working fine, I'll let in burn in some more before serious listening.
 
I've redone it, is it okay now ?

Yes, much better! A significant improvement over before, good job. You hope bad things don't happen, but it is nice to know that if they do, you will remain safe.

The safety standards I mentioned earlier cover both Class I and Class II equipment. Class 1 equipment typically use 2-conductor power cords with no earth, and is designed so that the user cannot, even by accident, come into contact with AC. Numerous special precautions are required. This is actually very common in the market. My Onkyo system receiver (amplifier) has a 2-conductor cord,, therefore the amplifier ground cannot be connected to earth ground, but yet it works just fine. If you check around your house you might bu surprised how many Class I devices you find. My opinion (worth exactly what you paid for it) is that the safest approach with a metal chassis is the Class II approach, where the chassis acts like a safety shield, ready to provide a safe path to ground if something goes wrong.
 

rjm

Member
Joined 2004
Paid Member
I've redone it, is it okay now ?
I've disconnected ground from earth.

An externally hosted image should be here but it was not working when we last tested it.

I think there will be less noise if you connect the GND (black wires) which are now disconnected back to the chassis (the big lug where you now have the earth and transformer screens connected).

By the way, those are Plitron transformers with the screens?

Do you have a part number on those?
 
I've been listening to AD8610's on Brown Dog adapters in Saphire 3 since late July and I quite like them, but I've been distracted by other activities and have not done any more listening comparisons, that has to wait for dark winter nights.
just got round to putting these on brown dogs, blooody hell they are 'fast' and 'open'. who took the cotton wool out my ears.. lol. if you like a laid back sound these are not for you. if however you want to hear every single little sound and the room size they are for you just be preprepared to sacrifice the bottom end. very very good at taming sibilance. they are very very open sounding but also very fast so bass stops and starts so quick there is no overhang, no tubby bass here. highs are superb as are vocals. only gripe is the lack of overhang on the bass so they sound a bit light weight and can sound a bit snappy.

they are the complete opposite to the OPA134's

i rather like them :)
 
Re. Trying AD8610's

just got round to putting these on brown dogs, blooody hell they are 'fast' and 'open'. who took the cotton wool out my ears.. lol. if you like a laid back sound these are not for you. if however you want to hear every single little sound and the room size they are for you just be preprepared to sacrifice the bottom end. very very good at taming sibilance. they are very very open sounding but also very fast so bass stops and starts so quick there is no overhang, no tubby bass here. highs are superb as are vocals. only gripe is the lack of overhang on the bass so they sound a bit light weight and can sound a bit snappy.

they are the complete opposite to the OPA134's

i rather like them :)

I still have the AD8610's in my Sapphire and continue to enjoy them. Supposedly the 8610's, with their lower voltage requirement, are well-suited to battery-powered amps. The comment about bottom end is interesting, I haven't noticed anything I'd consider a deficiency, but then I remind myself that I'm also listening to the output cap's rolloff. I'm using .68uF Clarity Caps and even though I purchased other cap brands and values to try, I like these so much I've left them in. My headphones are 250 Ohm DT880 Pros, which I think are quite neutral across the audible frequency spectrum. It all seems to work extremely well.
 
My headphones are 250 Ohm DT880 Pros, which I think are quite neutral across the audible frequency spectrum. It all seems to work extremely well.
i use the DT880 600ohm with the DT770 ear pads and to me its a superb combination with thunderous bass so to say the bass was a bit light is me saying it makes the 8610's sound like the 880's have the standard pads on.
 
I believe noise spikes down at -100 dB or even lower, even while not perceptible directly as noise, do influence the sound quality as a loss of spacial detail or clarity at the lowest levels of resolution.

And I believe you are correct.

I've just taken time to wrap the toroid with a mu-metal band. I also changed the stupid-expensive Obbligato caps for a pair of Vishay MKPs, and I can't believe the difference these 2 changes made. The best way to describe it is the amp "sparkles" now. The SQ was very good without the mods, but it really shines now. There's a presence, a detail, a sparkle in the high end that wasn't there before, and it's very, very nice to hear.

Since both the Obbligato and the Vishay parts are polypro, I have to assume that changing caps had a lesser influence, leaving the mu-metal band as the more significant change.

I'm going to add mu-metal bands as part of the formula for future builds


PS - I also owe the group an apology. I had the Class I / Class II stuff backwards. Class I equipment uses an earthed metal housing to protect against shock, while Class II relies on double insulation for protection:

https://en.wikipedia.org/wiki/Appliance_classes

My bad for working from memory.





.
 

rjm

Member
Joined 2004
Paid Member
The Obbligato's are very dark, even dull sounding.

Also I've noticed that most of the transformer noise seems to couple into the circuit through the film capacitor. So even something as simple as flipping the lead connection (so the outer foil wrap is at higher or lower impedance) can have an impact.

But yes, I felt the mu metal brought a huge step up in clarity and resolution.
 
Obbligato are caps with warmer smoother sound. This is reason for Your changes. And reason, why I not bought Obbligato for Sapphire. I prefered too more detailed sound. My favorite is Mundorf Supreme Silver/Oil, but is expensive. Good reference about detailed sound is too for Russian K73 PETP caps, these are cheap.
I have in my Sapphire switch, that can bypass input capacitor. So if I have my amp permanently connected to my DAC, I have bypassed input capacitor. This is cheapest way for best result. But You must be sure, that DAC have not any DC voltage on output.

For transformer noise, best way is transformers in separate chassis. This I use in my Sapphire.

Edit: RJM was faster :D
 
Last edited: