|
|
I have just finished ripping apart the preamp and installing a pair of new circuitboards that add the Pass Labs "X" differential circuitry, as well as adding a constant current source. As you can see from the original, space is limited, so I had to create an external power supply.
This variation on the BOSOZ is called the "X-CCS-BOSOZ."
I felt bad disassembling all my hand wiring, removing all that perfboard that took two years to put together. But, I used most of the original parts, resistors, MOSFETs, and capacitors in the rebuild. Performance is improved and it looks a lot better, although it takes up 2 boxes, rather than one. It also runs cooler than the original.
Click to Open In New Window |
|
| A high resolution picture of the power supply with the lights dimmed. |
|
This is the external power supply, top view. There is so much capacitance in the four power supplies (greater than 35,000 uF) that I added my Soft Start circuit to reduce turn-on power surge. The case is from Par-Metal. |
| Don't know why I put this in, but here is the underneath of the power supply. Feet are not yet installed. | |
| Preamplifier, Top View-- A lot of wires running around. I see a lot of very neat components made by other people but somehow the method of doing that eludes me. The inside is different, you see the red circuitboards and the added INA134 board for an additional pair of single ended outputs. The exterior of the preamp is the same (except for the fuses being replaced by WBT RCA jacks). Looks like I actually used the 9 pin dsub connector installed years ago. | |
|
|
Well I finally drilled ventilation holes in the preamplifier top cover. As you see, I don't have a special tool for this, its all done with a hand drill so it turns out quite poorly. I wanted to start using this right away so did not take the time to wait and have someone drill a nice grid into the cover. |
|
|
I've got a lot of switches on the front panel, they swith off all the LEDs, as shown. Note that the the 2 LEDs used for backlighting the gears cannot be turned off. |
|
|
Here is how I backlight the frosted glass of the microscope slide mounted behind the gears. Wire ties are fixed on the volume knobs with LEDs in them, using epoxy to fasten the LEDs. The LEDs have insulated leads. This arrangement allows the LEDs to be carefully aimed so that there are no hotspots on the frosted glass. |
|
|
Here is the rear view, note the pair of outputs where the fuses used to be. The Dsub-9 connectors are used to connect the power supply to the preamp. |
Since I was going with an external power supply, I had to add a bunch of features to the preamp, as I had room for more circuitboards in there.
This project was very difficult, took about 2 years and involved numerous design decisions and considerable expense. It was my first DIY project, yet I completed the GainClone and iPod DC Box during the pendency of this project. You see the numerous boards and components, all had to be carefully planned and everything barely fit into the case. I'm glad it worked.
If you don't know what a BOSOZ preamp is, click the link in the heading above or click hereAnyway, here are the many features of my design:
***New (5/05) *** Images will open in a new window
| Click on Graphic For Zoom | Description | |
|---|---|---|
|
|
Here are some exterior shots. At this point I had not drilled the vent holes in the cover yet. This needs vent holes as the MOSFETs get quite hot. Do you see the gears, yeah, there are a lot of close ups of those coming up. |
|
|
|
This is the "GainClone" type amplifier which was made during the middle of this project. Hopefully they match each other somewhat. Click here to see information on the amplifier. |
|
|
|
As this is a Nelson Pass preamp, its only fitting to use it with the Pass Aleph 3. Here you see the preamp with my Pass Aleph 3 (GainClone is disconnected). In the picture on the bottom, the switches on the right are in the "down" or "off" position. In this position they disconnect the 3rd transformer and all LEDs (save the 2 used for backlighting) and thus see the preamp operated in "quiet" mode; so no criticism that there are too many LEDs! |
|
|
|
A number of boards before being installed and some other items. | |
|
|
Here you see 4 boards installed: 1. The 5.7 volt power supply and LED relay board (near the xformers) 2. The Diode Matrix board (near the front panel) 3. The Input Relay Switching Board (near the back panel). 4. The DRV power supply board in front of the diode matrix board (almost touching the front panel). |
|
|
|
Close up of the 5.7 V board and the relay switching board.
the relay switching board grounds all unselected inputs to ground
via 100K resistors (for an easy load). |
|
|
|
Close up of the 5.7 V board (includes relay switching of illuminated current input) This board provides 5.7 V regulated DC output from any AC input, and also swithces any one of 4 pins to 5.7V to correspond to the selected input which causes the LED above the selected input to illuminate. |
|
|
|
Here are the the top and bottom of the L and R channel boards. You see the screws for on the underside of the boards for holding the heatsinks. Also, at this time I didn't know what a header was so I just inserted a few and connected wires to them. Many are upside down! The Solen output caps worked great, but were very large. These boards reflect new caps replacing the standard Panasonics. These board were made years before I finished the preamp. There is a lot go to go at this point. |
|
|
|
The L and R preamp channels are installed in the chassis. |
|
|
|
The L and R power supplies are installed in the chassis. Note that double the capacitance (4 x 1000 uF) is used for each PS. Almost all the boards are installed, one more which controls the voltage to many of the illuminated switches needs to be installed behind the two POTs, where the black circles are is where the feet attach. |
|
|
|
Well, the boards are in and now you see the wires connecting each board are installed. The power supply wires are shielded and many multi-conductor wires with multi-conductor pins and DIN connectors are shown. Many of these only carry current when a momentary pushbutton is depressed, so they don't require shielding and do not have the potential to affect the audio circuit. The wire going out the back is a power supply cable to power a DRV134 board to convert single ended input #3 to differential. This very tiny board has yet to be added. |
|
|
|
Another view from the rear angle. Note that the third transformer has been replaced; the prior one had heated up to 160 degrees F after about 45 minutes of use and then stopped working. That one was an unspecified transformer from a Kyocera CD player, it had the right secondary voltages but I had no idea of what its ratings might be; kind of hoped for the best. The new transformer is a 12VCT, 1A transformer, and I hope that I don't exceed its ratings. The wire sticking out the back where some RCA connectors should be is the power supply for the DRV board that mounts to the RCA connectors, this board is shown below. |
|
|
|
Attached to the faceplate you see that the frosted end of a "Fisher" microscope slide has been fixed via epoxy to the faceplate. This is for backlighting of the volume control. Yes, a package of slides was puchased on EBay for this reason alone. Also, switches and LEDs have been installed to the faceplate. |
|
|
|
The volume knobs are attached to shafts coupled to gears, which are coupled to a "balance" gear that can be pressed down to disconnect the L and R gears. This is the volume and balance control prior to the installation of switches and the frosted slide for the backlighting. Spacing of the gears was done mechanically after measuring gear diamater and pitch the triangles were set up. Note that the coupling 3rd gear allows for much error in the placement of the other two, although in my case it worked out ok either way. |
|
|
|
Although shown in the prior pictures, the "balance" decoupling structure is not clear. It is shown here. Its essentially a spring biased rotating shaft that is installed between and below the volume knobs, it is biased upwards by springs and a felt piece reduces friction on the shaft. It looks great, works but could be better. It took months and months to figure this out, I considered bearings and many types of shafts, all were tough to find. In the end I settled for this due to available materials, though I was seriously considering using a ball bearing instead. The right angle pieces were custom machined and the hinges were obtained at a hardware store. |
|
|
|
OK, these are cool, so here are a bunch of close ups of the gears and backlighting. As promised. |
|
|
|
Here is a zoom in of the front panel fully constructed with
gears and backlighting. |
|
|
|
Here is the front panel fully constructed and installed. You see that the selected input is indicated by a blue LED over the blue pushbutton switch. The top picture is pretty small but the other two are very nice if you click on them and display them full size! |
|
|
|
Nelson Pass says that the BOSOZ is a good single ended to differential converter, but I added another (very small) board using two DRV134s which do a single ended to differential conversion. This enables me to see which method works best for my setup. This tiny board attaches to the rear of input #3 RCA jacks. The rubber grommets fit over the end of the RCA connectors on input #3. Its very snug and the wire keeps it from coming off. The power supply at the front of the case powers this up via a shielded cable. You see one set of RCA connectors missing from some of the pictures above, and a unconnected cable. This board is inserted over the missing connectors and attaches to the cable. |
|
|
|
This shows the schematic and assembled pcb for the diode matrix board. This board does all the signal routing to cause the the depressed momentary pushbutton to "set" the appropriate relay and "reset" the other relays. Also shown are some test traces, as this was my first ever pcb layout, I was experimenting a little. Note the test footprint for the large relays that I was about to use in the relay switching board. I had to make my own libary for this part and tested the pattern on this board prior to making the board that would use the relays. |
|
|
|
Here is the relay board layout, blank pcb and assembled pcb, all in one big picture. All unselected inputs are grounded through 100K resistors to present an easy load to the source component. Note that I used four quad "latching" type relays with separate coils for "set" and "reset" which made the switching a lot easier to implement. I needed quad relays to switch the 4 balanced input signals for the L&R channels. |
|
|
|
This is the 5.7V regulated power supply layout and board. This board also includes dual latching relays (though I only use one set of inputs and outputs on each relay) for switching the 5.7 volts to the LED above the currently selected input (via appropriate resistors to reduce the voltage) so that the user knows which input is active. The relay actuation is also driven by the diode matrix board shown above. |