I am now very doubtful about my volume pot. Is there some way to check pot to see if it is defective or not?

[Edited by wnmnkh on 02-21-2009 at 06:49 AM.]

---

[Quoting wnmnkh]

I am now very doubtful about my volume pot. Is there some way to check pot to see if it is defective or not?

---

---

[Quoting amb]

---

[Quoting fc911c]

I would like to keep the B22 as is, will this be a problem? Are there any settings that need to be changed in this case?

---

No change is necessary (assuming your β22 is already wired with pre-out jacks). If your β22 is a 4-channel balanced configuration, the pre-out jacks should be 3-pin male XLRs, and don't forget to connect pin 1 to ground. If your β22 is a 3-channel active ground config, then the pre-out RCA jacks' ground should be connected to input ground, not the output of the ground amplifier.

---

Ok great, I have a ballanced B22 btw.

Thanks
Frank

I'm thinking of boxing up 3 B22 boards, the S22 board and leaving the transformer in its 'remote' box on a long cord.

that way there should be more control over placement of that darned thing, it shifts a lot of weight away from the main chassis and it lets me really tightly couple the PS with the 3 amp boards in a smaller chassis. I think this is the optimal config if you can afford to separate things like this, right?

so that means I'd have the 4 wires from the secondary going from the 'floor rat' box to the shelf installed B22 system box. any idea how long a cord I could get away with, and also what kind of wire you'd recommend? flexible is good since it helps the install. I'm thinking I can do 3' easily, but can I do 6'? should it even matter that much? my short distance should be from the spot AFTER the PSU regulation, not really caring much about the 'before', right?

any builds do it like this (pics?)

I know of a number of β22 built this way. An example is set300b's, found in the β22 website gallery.

Note that the option 2 diagram has the power switch on the transformer box. If this is inconvenient, then you could use a ε24 board in the transformer box, and put a momentary switch on the main amp box (plus two additional thin gauge wires to the umbilical) for remote operation.

I'm not looking forward to making the cable and drilling a huge round hole for that military plug/socket but I guess it has to be done

I want to have a safety so that when you plug jacks into the sockets, there won't be any live high-current wires and the relay will have to get an 'ok' from the amp before clicking on.

---

[Quoting linux-works]

I want to have a safety so that when you plug jacks into the sockets, there won't be any live high-current wires and the relay will have to get an 'ok' from the amp before clicking on.

---

my intention is to place JUST the transformer, a relay in series with all 4 wires, and fuse in the 'floor box' - and run the 4 secondary wires to the amp box. (plus some control wires but lets ignore those for now).

the power supply s22 board will be in with the amp boards and I don't plan on adding a power switch in THAT circuit; the s22 will have direct wiring to the 3 amp boards.

if I had the amp on and running for a while and I press the power button on my amp, I'm thinking that will break a continuity circuit (on the extra control wires in the umbilical cord) and the transformer box will unclick the 'big' relay in its box and the 4 30v power wires will be floating. so far, so good.

the issue you are describing is when you want to disconnect that umbilical and then re-connect it? you are thinking there could be a spark or jolt since the power supply is still storing charge in its caps (in the amp box) and connecting this to the tranny will cause a short?

but I'm thinking that if those 4 wires on the secondary are all relay controlled, and if you remove the cable, the relays click off creating a floating circuit. if you reconnect the cable AND keep the relays from auto-engaging (that might be the key part?) then you still have a safe and open/floating circuit.

next issue is: draining the local (amp box) PSU so that each startup is a 'cold start' so to speak. is it really worth that effort?

is there a trick I can do with a slow-start device (some kind of electrical series device) that will slowly ramp down its resistance? I've seen those before but not sure how they work or if they're the right thing to use here. is that an idea worth pursuing?

I do want to discuss this in more detail as I'd like to have all the issues resolved before drilling my first chassis hole

Instead of switching the transformer secondary with the relay, why not switch the primary? That way when the power is off, it really is off.

As for discharging the caps in the amp box, you don't need to do anything extra. The current draw from the amp boards will be high enough for this to happen pretty quickly.

---

[Quoting amb]

Never mind, it was late and I was thinking about a σ22 in a separate box scenario, which doesn't apply to you.

Instead of switching the transformer secondary with the relay, why not switch the primary? That way when the power is off, it really is off.

As for discharging the caps in the amp box, you don't need to do anything extra. The current draw from the amp boards will be high enough for this to happen pretty quickly.

---

so its all free, then?

maybe I was over-thinking things, as well. my goal was to ensure that if the power button was off (toggle button, but state=off) then no current would flow thru the wire and you are free to connect and disconnect worry-free until you finally have the wire connected AND you press the 'start' button.

would there be any benefit to moving the bridge over to the floor box along with the transformer and having rectified DC come across instead of pure AC? or even some local filtering, as well - just no regulation?

while we're drawing a line between what goes in the top box and what goes on the floor box (lol), I'm curious if there is reason to move other things as well.

this is 30v on 2 lines. putting that kind of voltage swing, in AC, near the signal wires is asking for noise pickup. but, if that is rectified and filtered (bridge+caps) then you are sending smooth DC along that distance and there is almost no chance of wire-based hum pickup.

Got a question regarding setting VR2.
With the cover open and fan blowing across the boards, I'm getting about 60mV across R34.
When i close the cover, and remove it 10 mins later to re-measure, I'm getting only 40mV across R34.Heatsink temperature at this point is about 58degC or 136degF
I understand this is due to the MOSFETs' negative temperature coefficient. Do you propose I reset VR2 so that R34 measures 55mV with cover closed or try to lower enclosure temperature

---

[Quoting amb]

ivant, the output quiescent current is up to you. There is no single correct setting, and is a compromise between how deep you want the output stage to be in class A, versus MOSFET device temperatures. As long as you don't exceed 65C or so at the heatsink surfaces, it should be safe to increase the bias a bit.

---

Thanks for the reply. I'll probably experiment abit on the temperature vs current setting to get an optimal value

I started my sigma and 3-channel beta builds but haven't decided how I'm going to do the enclosure thing yet. I'm leaning towards separate cases which leads to my question. Is there a good way of distributing the power from the connector to the three boards? I was just going to solder three wires to the incoming wire and shrink wrap them but that seems kinda inelegant. I saw someone in the gallery put together a little custom power distribution board but I was hoping for an off the shelf solution. Thanks!

But my amp brings another issue: the dc offset varies when connected to input. The right boards from BOTH sides, dc offset increases to ~20mV when connected to input.

Do you have any clue on this?

Edit : whoops I found out why

[Edited by wnmnkh on 03-01-2009 at 06:29 AM.]

Hey folks,

I have all my boards together and am about to get started with testing. But before I cut any wires, I wanted to settle once and for all on my enclosures. I'd like to put everything in one box but am a little worried about the transformer hum folks have discovered. As such, wanted to get some input as to the best way to mitigate this. My first thought was to fabricate a box similar to what others have done. Or maybe a box inside a box? Also, is thicker metal better or would more layers be better? Just looking for some guidance on how best to eliminate the noise.

I'm waiting on some parts to complete the s22 and I'll have some b22 boards sitting here doing nothing

I'm wondering, in the meantime, if I can get a clean 1amp or so per 24v rail and I simply connect one + to one - to form a middle ground, and I also manually trim the voltages so that both are exactly equal, can that get me by? or even suffice on its own as a mid-end solution?

how important is it that the + and - track each other? the dual TREAD approach is just pure lm317 linear. or I could even use the 317/337 combo for a more textbook +/- supply.

has anyone tried it? any reason why it wouldn't work? I'm assuming it will have worse noise specs than the s22 but for initial use (or budget use), any objection to it?