Your values look reasonable to me. You might consider using a slightly different value for R27 (depending on the transistors used, see here). But the CCS current is not excessive, so I don't think that's the problem. Does the CCS heat up the same in both channels?
I have to say that my experience with the heat is limited to "open chassis" in a cold workshop, which certainly helps to keep the temperature low at the CCS heatsink. I don't know for sure what will happen when I box it up an move the amp to its new home; but I don't believe the CCS will heat up badly.
Thinking about this a bit more, and looking more carefully at your photo: did you use thermal paste to get proper heat transfer from the transistors to the metal plate? I used Kerafol insulators which provide good heat transfer without the paste, and the CCS temperature is stable within a few minutes (certainly much quicker than yours).
Neverwhere10,
Is that the new ClarityCap (purple) in the C1 position?
if it is, have you tried anything else in C1?
Ron
Is that the new ClarityCap (purple) in the C1 position?
if it is, have you tried anything else in C1?
Ron
That is the newer Clarity cap in C1. I do have a set of C1 from the BOM i can try if needed.
I did not use thermal paste on any transistors that i used the Keratherm insulators from the DIYA store on. The mica insulators i did not paste either.. should probably at least do those. OR i can cut some Keratherm and use those on the small transistors. What do you guys think? I'll have to order some paste as all i have around is computer thermal paste and i think that's conductive.
If they were not thermally bonded well enough to the heatsink wouldn't that mean the transistor would get really hot and the heatsink not? Or is the idea the heatsink wont saturate with heat fast enough and stabalize the transistor?
I used to have a heatsink half the size attached to my VAS section but i swapped it out since it was so hot when sitting in an enclosed case i figured it needed more. The smaller one is about the size of mbrennwa's. I'd be very curious to see how your heatsink does when in an enclosed case for an hour.
What is the purpose of this heatsink, is it just for cooling or is it to get the 3 transistors on the same thermal profile?
Read that post mbrennwa's mentioned, OS suggested getting my CCS current to 11ma for the MJE350/340's i have instead of 8ma for better performance of those trannys. Guess that means more heat!
I did not use thermal paste on any transistors that i used the Keratherm insulators from the DIYA store on. The mica insulators i did not paste either.. should probably at least do those. OR i can cut some Keratherm and use those on the small transistors. What do you guys think? I'll have to order some paste as all i have around is computer thermal paste and i think that's conductive.
If they were not thermally bonded well enough to the heatsink wouldn't that mean the transistor would get really hot and the heatsink not? Or is the idea the heatsink wont saturate with heat fast enough and stabalize the transistor?
I used to have a heatsink half the size attached to my VAS section but i swapped it out since it was so hot when sitting in an enclosed case i figured it needed more. The smaller one is about the size of mbrennwa's. I'd be very curious to see how your heatsink does when in an enclosed case for an hour.
What is the purpose of this heatsink, is it just for cooling or is it to get the 3 transistors on the same thermal profile?
Read that post mbrennwa's mentioned, OS suggested getting my CCS current to 11ma for the MJE350/340's i have instead of 8ma for better performance of those trannys. Guess that means more heat!
Last edited:
Oh geez, you mean I have to try my spare Clarity MR's at C1? My HB is on the bench for other reasons so it won't be too hard to try them. The race is on again, Ron. 😉
Add some Keratherm or paste and see what it does.
I just tried mine with the lid closed. As far as I could tell by touching the heatsinks with my fingers, the CCS temperature was not different than with the lid open.
Am wondering if someone can verify my wiring before i turn the amp back on.
Linking to an imgur image of a drawing of my wiring below.
I am a) curious how i should ground my input wire shielding? From what i've read from AndrewT's posts i should only ground at one end, but where do i run that to? The case ground point or the 0v on the amp board?
And b) I only have one grounding line coming from my PSU Grnd to the case ground, do i need one from each amp ground as well?
An update on my heatsink issue, i've been running it with the case top off all day and it has been coolish to the touch so i think things are fine.. I'll experiment with the case top on later.
Linking to an imgur image of a drawing of my wiring below.
I am a) curious how i should ground my input wire shielding? From what i've read from AndrewT's posts i should only ground at one end, but where do i run that to? The case ground point or the 0v on the amp board?
And b) I only have one grounding line coming from my PSU Grnd to the case ground, do i need one from each amp ground as well?
An update on my heatsink issue, i've been running it with the case top off all day and it has been coolish to the touch so i think things are fine.. I'll experiment with the case top on later.
An externally hosted image should be here but it was not working when we last tested it.
Doesen't look terribly wrong to me, but here are a few comments:
- You don't have a ground loop breaker (that may be okay, but the breaker may help to reduce ground loop issues in combination with other devices connected to the same safety earth).
- Do you have a fuse between the mains and the PSU in?
- Do you have an inrush limiter?
- I don't know the diyAudio PSU, so can't comment on that.
- The input wire shields look like they are not connected anywhere, so they are useless. In fact, I don't use shielded cable for these connections, just some nicely twisted pair stuff.
- Do you have a DC protection to make sure you don't have DC on your speakers in case of a failure?
With new builds, I like to power up things step by step:
- First check if inrush limiter works (without the PSU connected).
- Then connect the PSU, without the amp boards connected. Don't forget to discharge the capacitors before continuing with anything else!
- Connect one channel of the amp; check if it behaves well. You might even do this without the power devices installed at first (see the Honey Badger build guide).
- Then add the second channel.
Use a bulb tester!!! This helps to keep the magic smoke inside your parts.
Best, at least to my experiences, is a ground star next to the PSU's filter caps. Everything must return to this, power GND of the PCBs, signal GND by their own connections, speaker gnd and a ground breaking network, e.g. consisting of two anti-parallel power diodes and an 100 nF cap, to your earthed chassis. Connect the RCAs by a shielded cable each and mount them isolated. The shield carries signal return and must be connected at both ends.
Edit: Sorry, your drawing is wrong in one point. As said before, speaker GND has to return to the star next to the filter caps. Your drawing scheme applies to single supply amps only with an electrolytic in the output.
No, you don't, see above.
Best regards!
Last edited:
I really doubt I stated that.
I repeatedly refer to 2wire signal connections.
If one uses a coaxial cable for that 2wire connection, then the screen/shield is part of the signal connection and MUST be connected to the signal solder pads at BOTH ends. You must not break the signal pair, by disconnecting one of the pair.
Alternatively one could use a screened twisted pair as the signal connection. The two cores form the 2wire connection. The screen/shield is a dedicated screen and carries no signal information.
This screen should be connected to the chassis at both ends to gets the widest band of interference attenuation. If you connect at one end only, the impedance of the screen impedes the passage of interference and attenuation of the interference is reduced.
Post a link to the post that told you otherwise and I'll read through it to find out why some third alternative could be adopted.
Last edited:
Sorry Andrew, you are correct, for some reason i remembered it wrong. There is a lot of back and forth on that subject in various threads in this forum, think i just got my info mixed up.
I do have mains fuses before the transformer and no ground loop breaker, that is something i'll look in to.
I have the DIYA Soft Start built but have not tested or put it in place yet.
I also have the DIYA Speaker Protection board built, but also not tested or put in place yet.
As far as why my speaker ground and signal ground are going to each amp is because that is how they are labelled on the amp pcb.. i don't know any better 😛Kay Pirinha said:
So what you're saying is i need to move my speaker and signal ground to connect to the star ground at my PSU instead of at each amp that the positive signal comes from/goes to?
Can you look at the linked drawing and tell me if that is what you mean with my grounding?
An externally hosted image should be here but it was not working when we last tested it.
The two Grnd blocks on the PSU drawing are connected together on the PCB and are right next to each other, am i OK with grounding my devices to these two blocks as my star point? Or do they have to be stacked on each other on a bolt or something.
If you want a visual of my PSU, i'll link a pic:
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Can't see any pics, attach them using the Forum's "Attachment" facility.
For a monoblock amplifier there is one PSU and one amplifier and one Main Audio Ground (MAG).
This MAG can be on the PCB, or near the PCB. In the past I have suggest that the group of Input socket, Output socket and Power ground be located fairly near each other and that the MAG be located at the centroid of this group. Peter Daniel seems to do something similar.
A stereo amplifier is different.
When one has a single PSU with multiple amplifiers then the MAG must be off board. Better to locate it at the centroid of that group: input sockets/output sockets/power grounds. Again P.Daniel does this.
If one has 4 secondaries (from two dual secondary transformers, or a quad secondary transformer), then there can be two MAGs, one dedicated to each amplifier.
Both these MAGs need separate safety connections to the Chassis. If these are direct wire connections, then the two channels become looped together. For multi-channel amplifiers I think one should use a Disconnecting Network to make the safety connection from each channel's isolated MAG to the chassis.
For a monoblock amplifier there is one PSU and one amplifier and one Main Audio Ground (MAG).
This MAG can be on the PCB, or near the PCB. In the past I have suggest that the group of Input socket, Output socket and Power ground be located fairly near each other and that the MAG be located at the centroid of this group. Peter Daniel seems to do something similar.
A stereo amplifier is different.
When one has a single PSU with multiple amplifiers then the MAG must be off board. Better to locate it at the centroid of that group: input sockets/output sockets/power grounds. Again P.Daniel does this.
If one has 4 secondaries (from two dual secondary transformers, or a quad secondary transformer), then there can be two MAGs, one dedicated to each amplifier.
Both these MAGs need separate safety connections to the Chassis. If these are direct wire connections, then the two channels become looped together. For multi-channel amplifiers I think one should use a Disconnecting Network to make the safety connection from each channel's isolated MAG to the chassis.
Last edited:
Is one single disconnecting network (ground loop breaker) enough for a multi-channel amp, or do you recommend a dedicated disconnecting network for each channel?
Is this a trick question?
As explained in post2393, there are three situations and each requires a different solution.
Single channel
Multi-channel with 1 PSU
Multi-channel with isolated PSUs.
Each isolated PSU needs it's own MAG to Chassis Safety connection.
If you have a single PSU, then you must have a single MAG.
If you duplicate the Safety connection for each amplifier from MAG to Chassis, when each are already connected together via the PSU Zero Volts, then you are incoporating multiple loops.
Last edited:
No.
My question refers to the last of these situations (multi-channel with isolated PSUs).
Yes, I understand this. But how should this connection be made? I see two ways of doing this:
- Should both MAGs have their own dedicated ground loop breaker to safety earth? The two MAGs would be separated by each other via the two ground loop breakers. This sort of reminds me of the Daniel Joffe approach to avoid ground loops in multi-channel amplifiers using hum breaking resistors at the inputs (by separating the signal-zeros of the different input channels).
- Should both MAGs connect to a single ground loop breaker, which then connects to safety earth? The two MAGs would be connected to each other directly (at their connection to the ground loop breaker unit). I believe this would be okay if the Daniel Joffe approach is used at the amplifier inputs.
I am not sure if understand this correctly. I don't intend to duplicate the connections.
Each power supply unit in an amplifier should use it's own separate ground loop breaker to keep each channel isolated from each other. This type of amplifier is basically two complete monoblock amplifiers housed in a single chassis, so they should be treated as if they were in separate chassis and unable to share common parts to maintain isolation.
Would this imply that we could omit the Daniel Joffe hum breaking resistors at the amplifier inputs, because the main audio grounds of the two channels are separeted from each other already?
(not that I am planning to modify the Honey Badger design, but I am trying to make sure I understand things well.)
You can never be sure what is going on for grounding in the source equipment, so it's usually best to leave any existing hum breakers at the input. If your source were to be chassis grounded the input hum breaker would likely be beneficial. They won't degrade operation at all.
Hi Matthias,
I'm curious on the answers to your question as well.
To my understanding, the hum breaking network (R5||D1||D2 in the HB's v2.4 design) prevents from hum that could be induced into the loop area formed by the two cables running from your preamplifier to the power amp. So, if you strictly separate each channel and their PSUs(the yet mentioned Dual Monoblock design), tying them loosely only by ground breaking networks elsewhere, e.g. next to the ground star of the filter caps (where impedance is lowest), to my understanding these three components on the boards could be omitted, i.e. replaced by a wire jumper.
Edit, after reading jwilhelms last answer: That would mean that we connect the RCA input by a shielded cable (or by a twisted pair) to the PCBs input terminals, and that any other connection from either end of that cable to ground would be contraproductive?
Best regards!
I'm curious on the answers to your question as well.
To my understanding, the hum breaking network (R5||D1||D2 in the HB's v2.4 design) prevents from hum that could be induced into the loop area formed by the two cables running from your preamplifier to the power amp. So, if you strictly separate each channel and their PSUs(the yet mentioned Dual Monoblock design), tying them loosely only by ground breaking networks elsewhere, e.g. next to the ground star of the filter caps (where impedance is lowest), to my understanding these three components on the boards could be omitted, i.e. replaced by a wire jumper.
Edit, after reading jwilhelms last answer: That would mean that we connect the RCA input by a shielded cable (or by a twisted pair) to the PCBs input terminals, and that any other connection from either end of that cable to ground would be contraproductive?
Best regards!
Last edited: