Salas hotrodded blue DCB1 build

Member
Joined 2006
Paid Member
I think the heatsinks are too small considering the resistor you want to use, and I would not use a clip, use a bolt and even a fender washer to hold the mosfet to a sink.
When hot rodded, floor mounting them is a start, and can glue heatsinks near the fets if needed as well.
I glued heatsinks to L-Channel and use that on both ends. I think my blog can show that.
 
Thinking about my hot rod operations on my wondrfully working DCB1:

Resistors: 4.5 Ohm 12 W Mills (I have two of these laying around)

At that current you are going to need a much more substantial hunk of metal. Horizontal mounting to the floor of a metal chassis has the added advantage of moving the radiated heat further away from other components, particularly the VRef LED strings.
 
What if I use one 10 Ohm 12W resistor? Would the heatsink be enough?

At that current you are going to need a much more substantial hunk of metal. Horizontal mounting to the floor of a metal chassis has the added advantage of moving the radiated heat further away from other components, particularly the VRef LED strings.

I think the heatsinks are too small considering the resistor you want to use, and I would not use a clip, use a bolt and even a fender washer to hold the mosfet to a sink.
When hot rodded, floor mounting them is a start, and can glue heatsinks near the fets if needed as well.
I glued heatsinks to L-Channel and use that on both ends. I think my blog can show that.
 
Thinking about my hot rod operations on my wondrfully working DCB1:

Resistors: 4.5 Ohm 12 W Mills (I have two of these laying around)

Heatsinks:

http://www.newark.com/jsp/search/productdetail.jsp?SKU=10WX1023
http://www.newark.com/jsp/search/productdetail.jsp?SKU=02M0324

I'll use silpads (not sure if I need Arctic Silver adhesive) to attach the clips to the heatsink. It will not be bolted to the chassis.

parodielin,

With those resistors you are going to pulling somewhere close to the magic 600mA and generating substantial heat. Small individual heatsinks won't cut it. Salas suggested something akin to a 30watt class AB amp. I am looking at the following found online, mostly eBay:

I am working on heat sinking for a stage 2, 600mA version.

  • Option 1) Use the case bottom as the heatsink, assuming a roughly 12x12x1/4" plate of aluminum with the MOSFETs bolted to them from the bottom of the PCB
  • Option 2) a single 140x190x32mm horizontal finned plate, mounted fins down in the case, with the PCB mounted on the back.
  • Option 3) A pair of 140x81x44mm vertical finned plates on each side of the PCB, MOSFETS mounted above board and bolted to the heatsink.
  • Option 4) As above, 250x100x55mm
Anybody successful at 600mA with any of these approaches?

Do any of you experienced DCB1 builders have some input for parodielin and I.
 
I would have been worried if you had said something like 5m to the actives about it may had needed higher value interfacing resistors than 220R not to oscillate on cabling capacitance, but 1.5m is very near to what all people use with DCB1 in home systems.

Both 1 & 2 on the graph ground pin 3 fully. So you had it grounded as your speaker's manual usually expects.

I am suspicious of possible bad joints, the computer's output as a source and what it may give out erratically through a DC preamp as the operating system does stuff, possible peculiarities in the unknown Dynaudio active plate amp, its preamp, filters, on board protection and limiters. As for what it may had happened to it and is lower but still has voice...don't know, from a tripped and stack limiter to burned stuff in a Class D power circuit. Check it with a balanced source with XLR output and XLR ready made cables first.
Alright, i got down to checking all the cable joints. They are all fine.

I changed the source to a CD player. When i tried it yesterday, there was some slight buzz/hiss but nothing major. Today, i shifted the speakers a little and when i turned on the left speaker, there was a huge, huge buzz. Right speaker is silent. I suspect it is because i shifted the left side closer to the TV.

But after some rewiring and shifting of cables, that was settled. Speakers are completely silent only when there is no source hooked up to the dcb1. Once i plug in the source, the noise comes back.

Is it really not possible to run it open case for now without getting *any* hum/buzz at all?

Is anything wrong with my internal wiring? And what should be done about it?
IMG_3006.jpg

IMG_3008.jpg

I know it's dangerous leaving the power wires exposed like that.. but it's temporary and i have no children/pets in the house so it'll do for now.

I'm thinking of trying out this alternative rca->xlr cabling arrangement. Any idea how well it'll work? According to the diagram, my cable is wired 'wrongly' although most literature elsewhere states otherwise.. so im not sure who to believe
wiring.jpg
 
I use uncased amplifiers regularly (same conditions, no kids, no wife, no pets).

While you are aware where the dangerous parts are, you are careful not to fall into them when you have been drinking (Oh I don't imbide either).

I find that the performance without a case is usually better than when inside a case particularly if the case is steel. I find steel cased two channel damned difficult to get completely quiet.

Generally I do not hear any interference coming into my gear while uncased and I rarely use any screened cable. Twisted pairs/triplets/quads are what I use.
 
Twisted pairs are missing.

Remember for every flow there must be a return route.
Electrical/electronic circuits are approximately what it says. The electricity flows around in a loop (circular route = circuit) all the way back to the start (the source of the emf).
Oh, okay. For the output, i understand the wiring. But i'm a little confused how to do a twisted pair at the vol pot <=> input section and the proper GND-ing scheme at the volume pot.

Because i need to connect the GND to two sides (RCA GND and DCB1 IN GND), how should it be done?

1) Each RCA IN's GND connects separately to individual (L and R) attenuator GND, then attenuator GND -> dcb1 GND?

2) Or attenuator GND does not -> DCB1 GND; instead, each RCA GND -> DCB1 GND (so there's two GND wires connected at the DCB1 GND)?

How should the twisted pair be implemented here?

Twisted pair for (RCA IN -> DCB1 IN) + (RCA GND -> attenuator GND)
OR
(RCA IN-> DCB1 IN) + (RCA GND -> DCB1 GND)?

Sorry for the novice questions. I read most of the grounding articles online but i still don't understand how to do it properly.

I tried these two methods and now i don't have the 60hz hum anymore when the source is plugged in without any volume level. But now there's some sort of funny high frequency noise and when i turn the volume up a little, the hum comes back. The higher the volume, the louder the noise
 
I’m building a balanced DCB1 hot rodded @ 600ma. I’m using a 111 x 222 10mm thick al anodized blocks to sink to the 3mm base plate 435mm x 353mm of the chassis.

This is my set-up internally:

imgPlace.com - Image Hosting - Free Image Hosting - Free Image Uploading - image 8314

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

John,

That seems a lot tidier than bulky heatsinks. I am surprised that a block of aluminum can do the job as well as a heatsink. Is your unit running yet or is this a build in progress? Once it is built, please let us know how well it cools. Thanks for sharing.
 
Michael,

By design fins are always a better option than a solid block for heat transfer. Fingers crossed my blocks work effectively, there’s certainly a large surface area for dissipation, I also plan to treat the surfaces between the base and al blocks with grease, avoiding any air pockets.

I’m still awaiting some parts, once up and running I will let you know the outcome :)
 
diyAudio Chief Moderator
Joined 2002
Paid Member
I tried these two methods and now i don't have the 60hz hum anymore when the source is plugged in without any volume level. But now there's some sort of funny high frequency noise and when i turn the volume up a little, the hum comes back. The higher the volume, the louder the noise

Is it like a whistle? Could be oscillations. Change your 220R series input/output resistors to 470R ones if the symptoms persist. Your now ones are they 220R for sure?
 
Is it like a whistle? Could be oscillations. Change your 220R series input/output resistors to 470R ones if the symptoms persist. Your now ones are they 220R for sure?
Sounds like a whistle. Yea they are 220R. Will order some 470R to try out and see if it's the problem.

When the source is not plugged into the input, the speakers are quite quiet - very very slight high frequency hash. The moment the source is plugged in, that whistle thing starts.

Can i say with certainty that it's not a ground loop problem anymore?

There's also this strange phenomenon: when the left speaker is ON and right speaker is OFF, unplugging the power cord from the right speaker causes a hum in the LEFT, even though the right side is OFF

And sometimes when both speakers are on, and i turn off one side, the other side will make some crackle/pop noise. I'm quite lost as to what to do now. Any help is really appreciated!
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Ground loops sound like hum and buzz. Zzzzz with some roar underneath. Whistle can be sign of oscillation between components. About your speakers power chord plugging/unplugging it certainly changes the mains ground route relationship to the common channels ground on the pre. Nothing is certain without expert data, but we try guess so you will not walk alone.
 
There's also this strange phenomenon: when the left speaker is ON and right speaker is OFF, unplugging the power cord from the right speaker causes a hum in the LEFT, even though the right side is OFF
!

This could be an earth issue, and as Salas pointed out, without expert data diagnoses is difficult at best. Do you have all of your gear running from one power socket via a power board? If not, maybe there is an earth issue with one of your power sockets and running everything from a single power outlet with a verifyable earth will at least eliminate that from the potential list of problems.

Good luck!
BK
 
Ah, i was looking around every part of the PCB to make sure i didn't miss out anything. Turns out one of the transformer wire wasn't secured properly to the terminal and it was loose.

The whistle/high frequency hash is gone - but the hum's back.

I've tried every conceivable way of wiring up the GND but i can't just get rid of the hum; the hum gets louder as i turn up the attenuator volume. This is even without any source plugged in.

Is it possible that i overheated any of the parts during soldering? Are the FETs particularly susceptible to overheating? But if that were the case, then wouldn't my measured values of the voltage, offset etc be weird/wrong?