Aleph J illustrated build guide

I have assembled a few of their cases now. I learned not to tighten the bolts until I have fitted the pieces together. There is play here and there. In particular the rails on the top and bottom of the heat sinks can deform when tightened to the heat sinks. Leave just enough play under the screws until the front and back are on. Then try to square things up and then tighten everything. Even then, the top and bottom plates need finesse sometimes.

+1 Bone13

This is virtually the only way you can assemble the HiFi2000 enclosures. The hardware they package with the kits leaves a lot to be desired. I usually end up having to get additional hardware like longer M4 bolts, washers, lock washers and nuts. All that said, they do look nice completed.
 
Awajoy - you have the RCA ground attached directly to chassis... I.E., on the wrong side of the ground-lift thermistor. That needs to be on PSU ground.

But before changing that, try wiring the RCA directly to the amp PCB (bypassing XLR) and jumper -IN to GND at the amp board inputs.

Hi 6L6 and Zen Mod. Thanks for the suggestion on moving the ground from chassis to PSU ground, and connect RCA directly to the PCB. In an effort to locate the source of the turn on/off popping/cracking noise, I tested the PSU by itself without any of the amp PCB's connected. It was noted that as soon as the power is on, in a split second's time, the V+ registers 25.4 volts. My understanding is that with CL-60 thermistors, the voltage is to rise slowly until the thermistor is warmed up. Does the above indicate the CL-60's are not working properly and is also the cause of the turn-on pop?
 
The CL60 would limit the current only if the AMP PCB's are connected to the power supply PCB.

Without the current draw, i.e. the power supply PCB with no load connected, the CL60 would present a current limiting function only during a very short period of time -> only while the capacitor bank is fully discharged.

Once the capacitor bank is fully charged, AND still with no current draw (AMP PCB's still disconnected), the CL60 would act like a 10ohm resistor in series with the primary winding, with very small amount of current running through it. Hence, it will not heat up at all and its resistance will remain at 10 ohms.

The power supply voltage rails will sag from 24-25 V DC (plus and minus) to around 21-22V DC under the constant load caused by 4 MOSFET "series pairs" conducting approximately 1Amp of current per each pair (connected in series, between + and - voltage rails) due to Aleph J mode of operation (pure class A). Under these conditions, the CL60 will be exposed to constant current draw, that will heat it up and therefore its cold resistance (10ohm), will drop to around 1 ohm or less.
 
Last edited:
Lonepine510 - The front panel attaches to the brackets with the bigger holes in your photo. :)


@6L6-- thanks... but I don't see how the big holes fit. Of the smaller, tapped ones (indicated by the chopsticks in photo below) I can get both upper ones to fit, or lower ones, but not both.


IMG_4468.jpg
 
@6L6-- thanks... but I don't see how the big holes fit. Of the smaller, tapped ones (indicated by the chopsticks in photo below) I can get both upper ones to fit, or lower ones, but not both.


View attachment 895851

lonepine510,

The tapped holes you’ve indicated on the photo of the backside of the faceplate are the correct holes for mounting.

To align the slotted holes on the angle brackets for mounting the faceplate, I loosened the screws that mount the top and bottom angle brackets to the heat sinks and adjusted the top-bottom distance between the top and bottom angle brackets so that those tapped holes will line up properly.

First, you will need to take the baseplate off to make this adjustment. Believe it or not, this slight adjustment of the L-brackets will line up the slotted holes on the L-brackets with the holes on the face plate and rear plate. When you remount the baseplate, use M4 flat or oval head screws with a long enough length to attach the baseplate and heat sink subassemblies. I used a combination of M4 washers and M4 Nylon lock nuts. I didn’t tighten the M4 screws and Nylon lock nuts completely until I had the face plate and rear plate mounted properly. I also did this assembly on a flat, level surface to insure chassis assembly was properly squared up. Believe it or not, once you make these small adjustments, the chassis is solidly assembled.

Of course, this will require additional hardware that doesn’t come with the chassis kit, but it will greatly improve the chassis assembly process. In fact, a lot of the hardware that came with the chassis is sitting in a bag unused.

55E4BF59-AECB-445C-AA58-7A7A04840AF9.jpeg
 
Member
Joined 2006
Paid Member
Thanks @rhing and @Bone13- I loosened the L-brackets on the heat sinks, and that certainly worked. I'll try the baseplate idea too- I'm having trouble getting the bottom cover to align with the holes. Thanks!

I’m glad that worked! The fun part about the bottom plate is putting the screws in the bottom (all not quite lined up), is doing it through the feet, so you can’t see at all what’s happening.
 

6L6

Moderator
Joined 2010
Paid Member
The chassis are not immediately obvious in the way they go together because everything bolts into everything else and the tolerances are actually a little tighter than it seems, so sometimes they can be assembled a little akimbo.

Essentially the entire amplifier is attached to itself at the heatsink rails... But the rails themselves are not stiff enough until the heatsinks are attached... so the heatsinks need to be attached very squarely to the rails. BUT

There is actually no mechanism to make sure the rails are on perfectly. That said, trying to square them up by eye to the heatsinks is a good place to start.

SO... the best order of operation is -

Rails on the heatsinks
Heatsinks attached to front panel
Rear panel attached
Perforated baseplate
Bottom panel
Top panel

IF the top panels don’t seem to line up with the mounting holes the best thing is to loosen the rails on the heatsinks a little and let everything wiggle a little, and then you will be able to find the holes. Attach the bottom panel squarely and snug the heatsink screws. The top should be square as well at that point.
 
IF the top panels don’t seem to line up with the mounting holes the best thing is to loosen the rails on the heatsinks a little and let everything wiggle a little, and then you will be able to find the holes. Attach the bottom panel squarely and snug the heatsink screws. The top should be square as well at that point.
Also, the top and bottom panels seems to have a front and back. If holes don't fit, turn it 180deg.
 
Member
Joined 2019
Paid Member
...I'm always rising base plate from lower rails, using 10mm spacers and longer M4 bolts

I stole this idea along with flipping the inner base plate from ZM. It allows a nice bit of room to route my mains wiring etc. Also, I read somewhere that routing the wiring under the baseplate can potentially help with some types of noise. Since I have zero engineering-level understanding of that principle (something about a Faraday cage), I'll just pass it along as hearsay until I can find the original source in my notes.
 
I stole this idea along with flipping the inner base plate from ZM. It allows a nice bit of room to route my mains wiring etc. Also, I read somewhere that routing the wiring under the baseplate can potentially help with some types of noise. Since I have zero engineering-level understanding of that principle (something about a Faraday cage), I'll just pass it along as hearsay until I can find the original source in my notes.


Search for Faraday cage and kid and you'll get an idea :) What wiring should go under? Mains to transformer primary perhaps?



So- you invert the base plate, and elevate it from the lower L-bracket rail slightly? Why not keep it right side up (ie, bent edges facing down) for more clearance?
 
The CL60 would limit the current only if the AMP PCB's are connected to the power supply PCB.

Without the current draw, i.e. the power supply PCB with no load connected, the CL60 would present a current limiting function only during a very short period of time -> only while the capacitor bank is fully discharged.

Once the capacitor bank is fully charged, AND still with no current draw (AMP PCB's still disconnected), the CL60 would act like a 10ohm resistor in series with the primary winding, with very small amount of current running through it. Hence, it will not heat up at all and its resistance will remain at 10 ohms.

The power supply voltage rails will sag from 24-25 V DC (plus and minus) to around 21-22V DC under the constant load caused by 4 MOSFET "series pairs" conducting approximately 1Amp of current per each pair (connected in series, between + and - voltage rails) due to Aleph J mode of operation (pure class A). Under these conditions, the CL60 will be exposed to constant current draw, that will heat it up and therefore its cold resistance (10ohm), will drop to around 1 ohm or less.

Thanks for the explanation. That's really helpful.
 
Awajoy - you have the RCA ground attached directly to chassis... I.E., on the wrong side of the ground-lift thermistor. That needs to be on PSU ground.

But before changing that, try wiring the RCA directly to the amp PCB (bypassing XLR) and jumper -IN to GND at the amp board inputs.

6L6 and Zen Mod:
Made changes as suggested: twisted and connected the +IN and GND wires directly from amp PCB to RCA connectors, and jumpered -IN to GND at amp board. Disconnected the XLR connections.

Result: I think the change in grounding has reduced the turn-on/off pop on the both channels. On the right (problem) channel, the turnoff pop is still higher than the left. The cracking noise has reduced some in both.

With the RCA inputs connected to a source, there is now a slight hum with the left but still a very audible hum on the right channel, which also seems to carry an electronic noise component in additional to in the hum.

The offset immediately after power on for the right channel is as high as 2.4 Volts with a reading around 260mV on R19 and 300mV on R18. After it has been on for about 20 minutes, the reading would settle to around 12mV offset, and 349mV on R19 and 394mV on R18. Anything coming through the right channel is faint and distorted.

Does anything stand out to be the problem?
 
6L6 and Zen Mod:
Made changes as suggested: twisted and connected the +IN and GND wires directly from amp PCB to RCA connectors, and jumpered -IN to GND at amp board. Disconnected the XLR connections.

Result: I think the change in grounding has reduced the turn-on/off pop on the both channels. On the right (problem) channel, the turnoff pop is still higher than the left. The cracking noise has reduced some in both.

With the RCA inputs connected to a source, there is now a slight hum with the left but still a very audible hum on the right channel, which also seems to carry an electronic noise component in additional to in the hum.

The offset immediately after power on for the right channel is as high as 2.4 Volts with a reading around 260mV on R19 and 300mV on R18. After it has been on for about 20 minutes, the reading would settle to around 12mV offset, and 349mV on R19 and 394mV on R18. Anything coming through the right channel is faint and distorted.

Does anything stand out to be the problem?
 

Attachments

  • IMG_3718.jpg
    IMG_3718.jpg
    864.3 KB · Views: 260
  • IMG_3720.jpg
    IMG_3720.jpg
    755.7 KB · Views: 260
  • IMG_3716.jpg
    IMG_3716.jpg
    665.8 KB · Views: 254
  • IMG_3712.jpg
    IMG_3712.jpg
    817.4 KB · Views: 255