Hello everybody ,
I' ve just finished my own Gainclone , which is visible here :
http://idefixes.phpnet.org/sections.php?op=viewarticle&artid=36
I' m sorry , the article is written in french , but the photos will show you how it is built .
It relly sounds well !
@ + Philippe
I' ve just finished my own Gainclone , which is visible here :
http://idefixes.phpnet.org/sections.php?op=viewarticle&artid=36
I' m sorry , the article is written in french , but the photos will show you how it is built .
It relly sounds well !
@ + Philippe
Translation
Sorry, my french is a little rusty (haven't used it much in over 20 years), but this should be a passable idiomatic translation:
-----------------------------------------------------------------------------
Gainclone with a switching power supply
Production of an LM3875-based GainClone
Why a GainClone? Why not?
I had some old switching power supplies from PCs on the shelves in my basement, which delivered 36VDC at several amps. It was an interesting starting point!
What amp can run on this voltage? I quickly thought of the famous Gainclone based on the LM3875, which is making a fury on many forums and sites.
The basis of my project was set.
Chronology
I first had to modify these fabulous power supplies to fit inside a chassis; 4 power supplies, two for +36VDC and two for -36VDC, in a double-mono configuration. Here is one of the power supplies before modification, seen from the component side.
And here it is from the solder side:
Here is a photo of the box which will accomodate this production. It's made up of one from an old PC, to which I have added heat sinks, flanges[????] for the connecting blocks, and a standard 19" rackmount face plate. A ????? of protection helps to avoid avoid accidental radiation [presumably EMF].
Details of the back section with the IEC receptable and two generous heat sinks.
Machine-work in progress
First implementation of the power supplies, side by side, lined up like in a parade:
A close-up view allows us to better visualize the connectors that I put in place to obtain the +/- 36VDC with two identical power supplies. The order of the cabling is as follows: A1+, A1-, A2+, A2-, which gives the +36, Ground, -36, connected together.
Verification made with a multimeter, we move onto the next part.
Installation of the first LM3875 module. Note the small number of components, I decided from the beginning to use point-to-point wiring with very short wires and the pots very close to the back, to avoid sending the signals to the front face plate.
Connections between the CINCHes [connecting blocks???] and the pots were made by twisting two wires together, and attaching these to the chassis mounts. The + and - of the HP [hot and neutral of high-potential AC?] uses the same path to avoid at all costs the ???? of ground.
The base plate area with a ferrite, varistors, and a shielded cable (which I happened to have handy). The earth ground is very strongly recommended for switching power supplies, it is relayed to all four power supplies and the metal chassis. The connection between the electrical grounds between the electrolytics are made via 150 ohm resistors on the bottom of the case.
At this stage, the work is not polished, we need to see if the idea holds water.
The measurements were correct, and a total absence of ripple encouraged me to finish the project. We took everything apart again, sanded, painted (with spray cans), and ????? [either something re-mounted itself, or the guy got high. Not sure which.
]
Here are the 1500uF electrolytics in parallel with 1uF capacitors. We really see the 100 ohm resistance.
The wires are fastened against one another. An elongator allows control of the pot from the front face plate.
My standards for for the symmetrical power supplies: + in red, - in blue, and ground in black.
No useless lengths of wire.
A little special-purpose sheet to improve the thermal contact between the LM and the heat sink [looks like Sil-Pad].
90 degree cabling; no loose/flying wires!
The protection grills for the swiching power supplies are put back in place. The 220VAC is never far away. We notice also the elongator on the axis of the Alps potentiometer.
Supplying power to the LED up front.
A little trip back to the rear to see the evolution as compared to photo number 1.
A view of the CINCHs and the high-pot connector; the back will stay aluminium-coloured.
An aerial view of the completed amp.
The front face with a nice, massive knob, a dual-pole switch and a power LED.
It's time to try some music now.
We are always happy when it works right on the first try.
A slightly enlarged photo to finish the article.
Now I still need to evaluate this amplifier more objectively. I will very soon build an electronic switch with some relays to make some tests against my other amps. I will draw inspiration from the schematic provided at the end of article about the 6L6 Idefixes[???] tube amp.
For now I am positively surprised that I have no noise: "My GC is not humming..."
Some long nights of comparative listening are planned for next fall.
-----------------------------------------------------------------------------
Oh, anything is square brackets is a comment of mine, ??? means "unsure of this word".
Cheers,
Wes
Sorry, my french is a little rusty (haven't used it much in over 20 years), but this should be a passable idiomatic translation:
-----------------------------------------------------------------------------
Gainclone with a switching power supply
Production of an LM3875-based GainClone
Why a GainClone? Why not?
I had some old switching power supplies from PCs on the shelves in my basement, which delivered 36VDC at several amps. It was an interesting starting point!
What amp can run on this voltage? I quickly thought of the famous Gainclone based on the LM3875, which is making a fury on many forums and sites.
The basis of my project was set.
Chronology
I first had to modify these fabulous power supplies to fit inside a chassis; 4 power supplies, two for +36VDC and two for -36VDC, in a double-mono configuration. Here is one of the power supplies before modification, seen from the component side.
And here it is from the solder side:
Here is a photo of the box which will accomodate this production. It's made up of one from an old PC, to which I have added heat sinks, flanges[????] for the connecting blocks, and a standard 19" rackmount face plate. A ????? of protection helps to avoid avoid accidental radiation [presumably EMF].
Details of the back section with the IEC receptable and two generous heat sinks.
Machine-work in progress
First implementation of the power supplies, side by side, lined up like in a parade:
A close-up view allows us to better visualize the connectors that I put in place to obtain the +/- 36VDC with two identical power supplies. The order of the cabling is as follows: A1+, A1-, A2+, A2-, which gives the +36, Ground, -36, connected together.
Verification made with a multimeter, we move onto the next part.
Installation of the first LM3875 module. Note the small number of components, I decided from the beginning to use point-to-point wiring with very short wires and the pots very close to the back, to avoid sending the signals to the front face plate.
Connections between the CINCHes [connecting blocks???] and the pots were made by twisting two wires together, and attaching these to the chassis mounts. The + and - of the HP [hot and neutral of high-potential AC?] uses the same path to avoid at all costs the ???? of ground.
The base plate area with a ferrite, varistors, and a shielded cable (which I happened to have handy). The earth ground is very strongly recommended for switching power supplies, it is relayed to all four power supplies and the metal chassis. The connection between the electrical grounds between the electrolytics are made via 150 ohm resistors on the bottom of the case.
At this stage, the work is not polished, we need to see if the idea holds water.
The measurements were correct, and a total absence of ripple encouraged me to finish the project. We took everything apart again, sanded, painted (with spray cans), and ????? [either something re-mounted itself, or the guy got high. Not sure which.
]Here are the 1500uF electrolytics in parallel with 1uF capacitors. We really see the 100 ohm resistance.
The wires are fastened against one another. An elongator allows control of the pot from the front face plate.
My standards for for the symmetrical power supplies: + in red, - in blue, and ground in black.
No useless lengths of wire.
A little special-purpose sheet to improve the thermal contact between the LM and the heat sink [looks like Sil-Pad].
90 degree cabling; no loose/flying wires!
The protection grills for the swiching power supplies are put back in place. The 220VAC is never far away. We notice also the elongator on the axis of the Alps potentiometer.
Supplying power to the LED up front.
A little trip back to the rear to see the evolution as compared to photo number 1.
A view of the CINCHs and the high-pot connector; the back will stay aluminium-coloured.
An aerial view of the completed amp.
The front face with a nice, massive knob, a dual-pole switch and a power LED.
It's time to try some music now.
We are always happy when it works right on the first try.
A slightly enlarged photo to finish the article.
Now I still need to evaluate this amplifier more objectively. I will very soon build an electronic switch with some relays to make some tests against my other amps. I will draw inspiration from the schematic provided at the end of article about the 6L6 Idefixes[???] tube amp.
For now I am positively surprised that I have no noise: "My GC is not humming..."
Some long nights of comparative listening are planned for next fall.
-----------------------------------------------------------------------------
Oh, anything is square brackets is a comment of mine, ??? means "unsure of this word".
Cheers,
Wes
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.