| BrianGT |
Well, with a prototype made with the board layout, I am considering organizing the order based on the current layout.
As for pricing, it looks like 1 pcb set would cost $20, and include:
- 2 power supply boards
- 2 amplifier boards
- 1 bridging board (DRV134)
This price would be for 4 oz copper (for the extra current handling), and the boards would be put together with scoring, as with my current LM4780 boards. (boards to either be red or black)
The amplifier pcb would be a single LM4780 chip hardwired in paralled configuration.
One pcb set would allow for these configurations:
- 2 channel amplifier (one amplifier pcb per channel) with a single or dual power supplies. (dual monoblock setup possible with one pcb set)
- bridged setup producing 150-200w for one channel
Remember that the LM4780 is hardwired in parallel mode, so one chip = 1 channel.
The non-bridged board should put out around 50-100w, depending on the power supply voltages and the load driven.
Kits will also be offered, and pricing will be announced soon. Pricing should be just a bit more than the current LM3875 group order.
I modified the Wiki for a new list to see the interest out there. If the interest is fairly substantial (100+ pcbs being the target to get the price), I will go through with the order, and offer kits. Please sign-up again here:
http://www.diyaudio.com/wiki/index....PCB+Group+Order
Please indicate your initial interest. This will allow for me to determine the kit pricing, and the whether there is enough interest for me to put time into this.
Drop me an e-mail if you have any questions. PCB set pictures will be posted soon.
--
Brian |
|
|
| OddOne |
I'm at the feeler stage for my paralleled version, so it appears we'll have both bases covered. :D
EDIT: Oh, didn't see yours was also a paralleled design. Wowsers, two designs to choose from at once. Hope there's a ton of interest for both. :)
oO |
|
|
| hams |
so to get teh 150w for a single channel, i need 1 complete pcb kit correct? so if i wanted six 150w channels, i'd need 6 kits?
what size toridal transformer is needed to power a single channel to 150w?
thanks,
kevin |
|
|
| bigparsnip |
Brian, if you do go ahead with this, and are offering a component kit, is it likely you will be using the Mills MRB-5 power resistors at all? Just if you are I would also be interested in buying a bunch of these seperately, to use as output resistors for my Aleph-X (preferably 0.33 or 0.22 ohms).
Alternatively, if anyone knows where you can buy these off the shelf I would really appreciate it if you could let me know. |
|
|
| mhennessy |
I'm curious - looking forward to seeing the prototypes. Personally, I like red.
But is there an option to order just the LM4780 PCBs? For my application, I won't need the rectifier boards. I guess the answer to this might be "no", as I guess you get all 5 boards made on one piece of FR4...
BTW, hams, I'm not sure that you'd get 150W from one PCB. To get 150W into 8 ohms, you need an RMS voltage swing of 35V, which is 50V peak. The highest supply rails you can put into an LM4780 are +/-42V, and these will fall under load to perhaps around 35V (this is hard to predict, of course). Allowing for other losses, I reckon you'd get around 30V peak, or 21V RMS, which is 55W. The parallel configuration will reduce some of these losses, so you might get nearer to 65W - I reckon that 60W is a fair guess of the continuous RMS power output of one of these PCB's.
Bridging a pair of these boards will give you double the voltage swing, which is 4 times the power in theory. Unfortunately, losses will reduce this to perhaps 3 times the output, so say 150-180W. But, these numbers are for continuous RMS outputs - dynamic or short-term "peak" outputs will be substantially higher. I reckon that a bridged setup will drive my ATC's quite well ;)
When sizing the torroid, a factor of two is a good starting point. But, it depends on the application - for a PA amplifier, you might want to increase it. For home use, you might get away with a smaller one. Unless you listen to heavily compressed music, you rarely need the full output power - it's there to deal with peaks. For the stereo bridged amplifier amplifier I'm thinking about (using 4 PCB's), I reckon 500VA will be enough ;)
Cheers,
Mark |
|
|
| roibm |
me me me me me
me me me me me
me me me me me
me me me me me
me me me me me
me me me me me
me me me me me
me me me me me
me me me me me
me me me me me
me me me me me
me me me me me
.
.
.
.
me too
and I like black. black is cool. imagine black pcb, black heatsinks, black case, black toroid :D
yeah, and another point... what is a kit?
one channel bridged? or two channels bridged? |
|
|
| BrianGT |
| quote: | Originally posted by bigparsnip
Brian, if you do go ahead with this, and are offering a component kit, is it likely you will be using the Mills MRB-5 power resistors at all? Just if you are I would also be interested in buying a bunch of these seperately, to use as output resistors for my Aleph-X (preferably 0.33 or 0.22 ohms).
Alternatively, if anyone knows where you can buy these off the shelf I would really appreciate it if you could let me know. |
I will keep this in consideration when ordering parts. This shouldn't be a problem if I go for the Mills resistors.
--
Brian |
|
|
| BrianGT |
| quote: | Originally posted by mhennessy
I'm curious - looking forward to seeing the prototypes. Personally, I like red.
But is there an option to order just the LM4780 PCBs? For my application, I won't need the rectifier boards. I guess the answer to this might be "no", as I guess you get all 5 boards made on one piece of FR4...
Bridging a pair of these boards will give you double the voltage swing, which is 4 times the power in theory. Unfortunately, losses will reduce this to perhaps 3 times the output, so say 150-180W. But, these numbers are for continuous RMS outputs - dynamic or short-term "peak" outputs will be substantially higher. I reckon that a bridged setup will drive my ATC's quite well ;)
|
Mark,
You can find the prototype pcb pictures in this thread, where the idea of the layout was created:
http://www.diyaudio.com/forums/show...1330#post391330
The pcb set will be created on a single pcb set, with scoring to allow for the boards to easily be broken apart, just as with my current LM3875 order:
http://www.briangt.com/gallery/nigc-kit
It will consist of 5 boards on one, instead of 3. Having 2 power supply boards will easily allow for building monoblocks from one pcb set. The bridging pcb hasn't been finalized yet.
When bridging with the LM4780 layout, keep in mind that it is actually a bridged-parallel configuration, which seems to be the optimal high-power solution with the chipamps.
Here is a picture of the amplifier board layout. The power supply board will be similar to the last one from the LM3875 group order.
Wiki signup:
http://www.diyaudio.com/wiki/index....PCB+Group+Order
--
Brian |
|
|
| peranders |
Brian, I like what I see BUT the signal ground (red) trace should go more left and directly to the OG pad. Than you have got a perfect pcb!
BTW: Where are the decoupling caps? Can't see any. |
|
|
| Peter Daniel |
| quote: | Originally posted by peranders
Brian, I like what I see BUT the signal ground (red) trace should go more left and directly to the OG pad. Than you have got a perfect pcb!
BTW: Where are the decoupling caps? Can't see any. |
Wouldn't connecting SG and OG pads with bottom layer trace trace take care of that?
There is no coupling caps, as they are not required here. All the amps I build now don't have coupling caps either. I used them before though. |
|
|
| roibm |
what is a kit?
one channel bridged? or two channels bridged? |
|
|
| mhennessy |
Something that's crept in to later versions of this layout - the connection between pin 7, R1 and the output. I'm not happy about the noding here. The connection on the other half of the chip is a perfect textbook example of how to do it, but the "T"ing of the output connection worries me slightly. I can see the space constraints that have caused you to shape the track in the way you have, but could you not run a track on the red (top?) plane from pin 7 to R1, and then take the output from R1? The via shouldn't cause a problem here, as it will have the wire and solder from R1 to "beef up" the connection.
I also agree with peranders about the track linking the earth planes - it doesn't need to be so long, and would be better connected to the middle of the earth plane. I also think that the OG would be better if you moved it to be between OUT and SG - the symmetry would definitely improve performance - easily measured (if not necessarily audible;)). However, this is only an issue if you're not bridging a pair of boards, so I don't feel that strongly about it.
Have you considered moving R6 and R7 around to the right of the IC, behind the cap? It would reduce the length of the input traces, and these are at a higher impedance than the input node, so slightly more susceptible to pickup. You might even be able to make the board slightly smaller as a result. (Although, is the spacing of the caps chosen so that you can easily get to the screws holding the IC to the heatsink?). Also, the trace from R6 looks very close to the R9 output node, and as this is a non-inverting amplifier, the chances of positive feedback are very real.
Another small thing - the spacing of the resistors R3 to R7 looks to be non-uniform - is this the case in reality, or a screen-draw issue? Moving them all closer together so the spacing matches R6 and R7 will make it look a bit neater, and you will be able to tidy up the blue SG trace slightly (and the trace to R3, which looks as if it "misses" the centre of the pad).
Sorry if all of this seems unduly critical, but these are the sorts of details that I ask myself when I'm nearly ready to declare my designs "finished". I normally find that the critical analysis can take longer than the basic layout. But, I'm a perfectionist, which isn't always easy ;)
Finally, is it an option to purchase the resistors along with the PCBs?
Best regards,
Mark ;) |
|
|
| Peter Daniel |
| 2 channel paralleled or one channel bridged. |
|
|
| Peter Daniel |
I would be much more concerned with via than a T. There is no problem running this trace directly between pads, without the need for a T. I wouldn't recommend mixing 2 layers for that.
As to the output ground 200mill won't make a difference in performance, but it will make big difference in how easy it is to run the connection wire. I don't care for perfect symetry. I care much more for convenience.;) |
|
|
| Peter Daniel |
| quote: | Originally posted by mhennessy
Another small thing - the spacing of the resistors R3 to R7 looks to be non-uniform - is this the case in reality, or a screen-draw issue? Moving them all closer together so the spacing matches R6 and R7 will make it look a bit neater, and you will be able to tidy up the blue SG trace slightly (and the trace to R3, which looks as if it "misses" the centre of the pad).
|
The spacing it's done here on purpose. Those are not Krell (or Kristian's boards for that matter) where everything is lined up and spaced perfectly. Looks were not considered to be the goal, but performance. The board accomodates Caddocks, but I have strong feeling that Rikens will be better here, so this extra space allows to mount Rikens as well (in vertcal position). You wouldn't mind putting them this way? After all those extras on the leads are part of 3-D layout, (popularasied by Jonathan Carr);) |
|
|
| Peter Daniel |
| quote: | Originally posted by mhennessy
Have you considered moving R6 and R7 around to the right of the IC, behind the cap? It would reduce the length of the input traces, and these are at a higher impedance than the input node, so slightly more susceptible to pickup. You might even be able to make the board slightly smaller as a result. (Although, is the spacing of the caps chosen so that you can easily get to the screws holding the IC to the heatsink?). Also, the trace from R6 looks very close to the R9 output node, and as this is a non-inverting amplifier, the chances of positive feedback are very real.
|
Even if you place those resistors closer to the chip, you still have to run the connection wires around the board (this makes those wires longer) so I don't see much difference here. And anyway, there is no space to place those resistors there. Again having conveniently placed input access points on the board was major deciding point.
I would rather modify the traces (less square) and this would further reduced their length.
Regarding the "trace from R6 looks very close to the R9", look how close those are on IC pins;) |
|
|
| Peter Daniel |
| quote: | Originally posted by mhennessy
Sorry if all of this seems unduly critical, but these are the sorts of details that I ask myself when I'm nearly ready to declare my designs "finished". I normally find that the critical analysis can take longer than the basic layout. But, I'm a perfectionist, which isn't always easy ;)
|
It took at least 5 days and countless hand drawings to finalize that layout.
I'm still working on a DAC (for almost a month) before committing to final layout.
So be asured that a lot of thought went into those amp boards. Also, we ride on experience gained with a previous version, which tested by hundreds brought only positive feedback;) |
|
|
| BrianGT |
| quote: | Originally posted by roibm
what is a kit?
one channel bridged? or two channels bridged? |
A kit will consist of the components for 2 amplifier boards and 1 power supply board. Additional components will be available for the bridging board, and the extra power supply board at additional cost.
My estimate on the kit price for the basic kit will be $40-45, and the premium kit will be $70-75.
Here is a list of components for the kit to contain:
2 - LM4780 chips
1 - PCB set
6 - 22k resistors
4 - 1k resistors
4 - 680 ohm resistors
4 - 0.22 ohm 3w resistors
2 - SMD resistors for mute
4 - 1500uF 50v Panasonic FC capacitors
2 - 4.7uF capacitors for power supply pcb
8 - MUR860 diodes for power supply pcb
The difference between the premium and basic kits is the choice of components. The premium kit will use Caddock MK132 resistors and 4.7uF BlackGate N capacitors for the power supply board. The 1500uF 50v Panasonic FC capacitors will be used for both kits for now. I am considering offering BG 1000uF STD capacitors in the future, but they are rather expensive costing $12.50 each (with quantity discount), making the price $50 for the capacitors needed fo a kit.
As for the upgrades to the kit, it will cost $5 to add the components for an extra power supply for the basic kit and $10 for the premium kit. I haven't determined the pricing for the bridging board components yet, as the design is being finalized.
I will post more details when everything is finalized.
--
Brian |
|
|
| joensd |
roibm,
would you mind ordering 2 additional sets of PCB´s and send them to me. Just to save a bit of shipping cost.
Your email adress is not available through the forum, can you drop me a mail?
Regards
Jens |
|
|
| roibm |
| quote: | Originally posted by joensd
roibm,
would you mind ordering 2 additional sets of PCB´s and send them to me. Just to save a bit of shipping cost.
Your email adress is not available through the forum, can you drop me a mail? |
Neither is your email address. And this is not normal, as it has worked just fine previously.
I could do just that, even for other people if interested(as long as the quantity is not that high, otherwise we will have to pay customs + our lovely mwst)
If there is someone from romania that wants to order I could take the order for you as well. You will have to send the money to a romanian bank account so this way you will save some money. For more details contact me when the email contact function gets back. |
|
|
| mhennessy |
| quote: | Originally posted by Peter Daniel
I would be much more concerned with via than a T. There is no problem running this trace directly between pads, without the need for a T. I wouldn't recommend mixing 2 layers for that. |
This is a NFB system. The output should be taken from the same point that the FB is sampled from. That point is R1. The current flowing in the T shaped track will cause voltage variations, which equals distortion. The currents will be measured in amps, and the distortion will be easily measured, and perhaps audible. This in itself might not be a big deal, but when it's easy to fix, why not?
The via will not cause problems because it means the output will be taken from R1. Sure, it might add some resistance, but this will be swamped by just the tolerance of R8.
| quote: | Originally posted by Peter Daniel
As to the output ground 200mill won't make a difference in performance, but it will make big difference in how easy it is to run the connection wire. I don't care for perfect symetry. I care much more for convenience.;) |
Sorry - I don't see how moving OG will alter this. I proposed moving it to between SQ and OUT - that it, move it south, then east a bit so that it's directly above the OP pad. This will make wiring simpler because you can group (or twist) the OP wires. It means altering the blue output trace slightly, but that's all. Again, it's easy to do, and will have a positive effect on the performance. Why not?
| quote: | Originally posted by Peter Daniel
The spacing it's done here on purpose. Those are not Krell boards where everything is lined up and spaced perfectly. The board accomodates Caddocks, but I have strong feeling that Rikens will be better here, so this extra space allows to mount Rikens as well (in vertcal position). You wouldn't mind putting them this way? After all those extras on the leads are part of 3-D layout, (popularasied by Jonathan Carr);) |
Sorry - you can't convince me that the uneven spacing was deliberate. This is just how things look after the first "pass" at placing components without using a horizontal grid. If it was deliberate, why is R4 so very close to R8, causing the blue SG trace to almost touch the output node. BTW, this is bad, because the current in the output node could easily induce currents in the SQ, causing yet more distortion (although this partly depends what the IC does with it's GND pins, and I can't comment because I've yet to see an internal schematic).
I wouldn't like to mount resistors vertically - this is universally bad practice on a PCB. What JCarr is quite different - he uses this technique to shorten signal path lengths. I was planning to use 1/8 watt resistors that mount horizontally, bit would consider using those square-body resistors that I've seen (sorry - I'm not up with the brand names)
| quote: | Originally posted by Peter Daniel
Even if you place those resistors closer to the chip, you still have to run the connection wires around the board (this makes those wires longer) so I don't see much difference here. And anyway, there is no space to place those resistors there. Again having conveniently placed input access points on the board was major deciding point. |
If you moved the resistors nearer the IC, the traces between resistor and IC gets shorter. The trace from I/P pad (which doesn't move) becomes one of the two traces that you currently have. You might have mis-understood the point - in terms of suscecpability to RF pickup, the traces between IC and resistors are more sensative because they are at a higher impedance, so these should be "protected" or shortened. The trace from IP pad is at a lower impedance, so is less of a problem. Lookup guard-traces in any precision op-amp datasheet.
| quote: | Originally posted by Peter Daniel
Regarding the "trace from R6 looks very close to the R9", look how close those are on IC pins |
You're missing the point - sure the pins are close together on the IC, but look at how National allocated them - the OP pins are seperated from the IP pins by power lines. Indeed, this is normal practice in most amplifier ICs. Again, how much effort would it take to move that trace south a bit? If there's even the slightest doubt about something, and it's easy to fix, then why not?
| quote: | Originally posted by Peter Daniel
It took at least 5 days and countless hand drawings to finalize that layout.
I'm still working on a DAC (for almost a month) before committing to final layout.
So be asured that a lot of thought went into those amp boards. Also, we ride on experience gained with a previous version, which tested by hundreds brought only positive feedback |
I'm not suggesting that you've not spent a lot of time on this. However, I took quite a lot of time to reply to you, and suggested some genuinely simple things that you missed. I could have suggested lots of other things, but realised that you don't want to re-design the whole PCB at this stage. I thought I was being helpful - the suggestions were based on my experience and understanding of the theory - which is just as valid as your original input, even though you don't seem to think so.
Sorry if that's harsh, but despite the use of smileys, the tone of your replies is dissmisive. In the time it took you to start replying, I can't believe that you even had time to read the message properly, let alone think carefully about a response - you must read very quickly ;)
As I keep saying, if there's even the slightest doubt about something, and it's easy to fix, then why not? It's this attention to detail that makes a product unique and rewarding.
YMMV
Best regards,
Mark
PS What does BrianGT think? |
|
|
| garbage |
any ideas if it would be ok to use a 4ohm speaker for bridged mode?
i understand that each amp(when bridged) will see the speaker as a 2ohm load per channel, but want to know if the lm4780 can take such a load. |
|
|
| cdoggy81 |
| Looks like another good design!!! |
|
|
| BrianGT |
| quote: | Originally posted by mhennessy
PS What does BrianGT think? |
Mark,
Thanks for the your comments on the layout. I spoke with Peter, and I will make some revisions to the design and post a new screen shot this evening. I find your concerns to be valid, and I will see what I can do.
--
Brian |
|
|
| Peter Daniel |
Via is not an option in a well designed, audio board, sorry.
Besides replying on a fast, I aslo took your suggestions seriously, and will try to implement, togethe rwith Brian, to the best degree. But some things are either negligable, or simply not possible.
You mught not like vertical standing resistors from aestchetical point, but if they sound better, I will not loose sleep over that. Those resistors are vertical in order to maximise the width of ground plane. You can always use smaller size regular ones , or Caddocks. |
|
|
| BrianGT |
| quote: | Originally posted by garbage
any ideas if it would be ok to use a 4ohm speaker for bridged mode?
i understand that each amp(when bridged) will see the speaker as a 2ohm load per channel, but want to know if the lm4780 can take such a load. |
The amplifier board uses 2 channels in parallel with the LM4780, so each channel would see 8 ohms on the amplifier pcb if you are just using one pcb. When you bridge two amplifier boards, each amplifier board will see half of the load, but since there are 2 channels in parallel on each board, each channel (4 of them used in a bridged setup), will see a 4 ohm load.
--
Brian |
|
|
| lgreen |
| I'm interested in sufficient boards for 4 or maybe 5 bridge parallel channels. Red boards preferred. What will 4 ch of bridge parallel need? About 8 boards? Probably don't need all those power supply boards for this application. |
|
|
| Peter Daniel |
| quote: | Originally posted by mhennessy
I'm not suggesting that you've not spent a lot of time on this. However, I took quite a lot of time to reply to you, and suggested some genuinely simple things that you missed. |
I didn't miss anything, Brian did the final layout, following my earlier suggestions. If I modified it, it might have look differently, as well. My points are valid, regarding major design decisions, not small details. |
|
|
| Peter Daniel |
| quote: | Originally posted by mhennessy
Sorry - you can't convince me that the uneven spacing was deliberate. This is just how things look after the first "pass" at placing components without using a horizontal grid. If it was deliberate, why is R4 so very close to R8, causing the blue SG trace to almost touch the output node. |
The spacing between those resistor should be according to that scetch. Only than installing Rikens present's no problem. This spacing is deliberate. It seems like Brian didn't do it quite accurately though. |
|
|
| mhennessy |
| quote: | Originally posted by BrianGT
Mark,
Thanks for the your comments on the layout. I spoke with Peter, and I will make some revisions to the design and post a new screen shot this evening. I find your concerns to be valid, and I will see what I can do.
--
Brian |
Thanks for offering to consider my suggestions, Brian. That's all I ask ;)
| quote: | Originally posted by Peter Daniel
Via is not an option in a well designed, audio board, sorry. |
Genuine question - Can I ask why?
| quote: | Originally posted by Peter Daniel
Besides replying on a fast, I aslo took your suggestions seriously, and will try to implement, togethe rwith Brian, to the best degree. But some things are either negligable, or simply not possible. |
I only wanted you to consider them as possible options. Can you at least understand why the speed and tone of the reply didn't suggest this?
| quote: | Originally posted by Peter Daniel
You mught not like vertical standing resistors from aestchetical point, but if they sound better, I will not loose sleep over that. Those resistors are vertical in order to maximise the width of ground plane. You can always use smaller size regular ones , or Caddocks. |
My concerns for this are not aesthetic - I was thinking about the (real) possibility for RF pickup again. I've worked in different dissiplines over the years, and I guess that certain experience makes you worry about certain things...
So, have you found that mounting resistors vertically improves the SQ? Or, are you thinking of comparing different sorts? For example, I quite like 0.1% RN55's, which would have to be mounted vertically on this PCB - if I compared these to standard quality miniature 1/8W resistors (mounted horizontally), would the audible differences be due to the component or mounting method?
Best regards,
Mark |
|
|
| mhennessy |
| quote: | Originally posted by Peter Daniel
The spacing between those resistor should be according to that scetch. Only than installing Rikens present's no problem. This spacing is deliberate. It seems like Brian didn't do it quite accurately though. |
OK - I take your point, although it looks to me like that spacing was to accomodate the blue ground traces - something that isn't an issue with the latest layout.
Also, can I compliment you on your perfect output noding ;)
I'm looking forward to seeing (and eventually hearing) the finished product. Thanks to you and BrianGT for all your work!
Cheers,
Mark |
|
|
| Peter Daniel |
As of now, I'm already working on improving layout based on your suggestions.
Spacing of those resistors is not really desirable (as blue ground trace would be easier to run) but I need that space.
I never seen vias on well implemented audio boards, you can see them in digital a lot though. Aleph X group purchase board was designed with vias, but it was questioned initially, and whwnever I look at that, I think to myself that it could be done differently. To me, via connection is a waek point in a signal path and I never even consider it from pure audiophile POV. Using it is conveneient though.
Regarding vertical resistors, this is not really standard option for that board, Caddocks can be easily used here, but I think that Rikens will sound better (they do on 3875 board) and I wanted to allow placing them here, even at the cost of vertical arrangement. I personally also don't like that way to mount resistor, but inn some cases it is a calculated compromise. In this particualr layout, using Caddocks, leaves more space for running uninterupted ground plane, so I opted for Caddocks (or Vishays) as it looks good and it's a small package. But in my personal amp, I wil be probably using Rikens here. Doen't treat it as inproperly mounted resistor, treated as a resistor with additional trace attached in 3-D space. To me it doesn't make any difference. Running extra trace in air is worth of widening the ground plane.
Regarding the other two resistors (the series input connection), it would be nice to place them closer to the chip, but then the ground returne path is getting longer, (as you can't move all resistors from lack of space) so I don't know if anything is gained by that.
I reply fast, because I already have al lthe answers ready. I asked those questions myself many time when doing that layout.
However I agree that there is a room for improvement here, and whatever possible, it will be done ;)(this ones for real) |
|
|
| mhennessy |
| quote: | Originally posted by Peter Daniel
Regarding the other two resistors (the series input connection), it would be nice to place them closer to the chip, but then the ground returne path is getting longer, (as you can't move all resistors from lack of space) so I don't know if anything is gained by that. |
Thanks for the reply. It had never occurred to me that vias could be a week point, because they are no worse than a standard thermal relief in a ground plane, but I would agree that it's normally good plactice to keep signals on one side of the board...
Anyway, I'm not sure if I explained the two-resistors thing clearly enough, so I've modified the .gif (I don't have access to PCB CAD here) - the image is messy, but you'll hopefully get the idea. Perhaps putting them there causes physical problems (if they're vertically mounted, for example)? Note that I also took the liberty of following this through to the logical conclusion - I moved R4 to occupy the space left by R6/7, improving the earth trace that I mentioned earlier.
Having said all of that, I prefer the aesthetics of 5 little resistors in a neat row ;) |
|
|
| Peter Daniel |
But do you see how input trace extended? So nothing is really gained except for unning shorter 2 traces and longer a single trace. Also the SG node is chanched and I don't think if for better.
I modified resistors spacing a bit and some traces, addressing T connection and trying to simplify input traces (they are shorter now). I still didn't do adjust polygon but connected SG and OG with a single blue trace. |
|
|
| Peter Daniel |
| With regards to have OG centrally positioned, I think it could be swaped with SG pad for more central placement. |
|
|
| mhennessy |
| quote: | Originally posted by Peter Daniel
But do you see how input trace extended? So nothing is really gained except for unning shorter 2 traces and longer a single trace. |
Yes, but the point you're missing is the traces between the resistors and IC are the vulnerable ones! Sure, the input trace is perhaps slightly longer, but remember that this trace is just an extension of the input wiring from the rear panel (or DRV134 PCB). This trace is at low impedance, and less prone to interferrence.
| quote: | Originally posted by Peter Daniel
Also the SG node is chanched and I don't think if for better. |
The earthing is actually better with this, but as it's the signal earth and no currents are flowing here, this is a mute point.
| quote: | Originally posted by Peter Daniel
I modified resistors spacing a bit and some traces, addressing T connection and trying to simplify input traces (they are shorter now). I still didn't do adjust polygon and forgot to connect SG and OG with a single blue trace. |
Yes, this is much better, and addresses most of my concerns.
The biggest problem with my input resistor move is the uncomfortable aesthetic of those two resistors all out on their own (although you did say that you prefered sound to aesthetics ;) )... While I was playing, I couldn't resist improving the signal earthing further (possible alternative route shown in green, although I prefer the blue) |
|
|
| Peter Daniel |
Like this, but I'm not sure if it's really better. I'd rather prefer to have input signal and input ground side by side, and moving OG by 200mil won't make any observable difference, IMO.
Another issue is running the connections wire. Going away from the board I would also prefer to have them spaced, and this arrangement doesn't relly allow it (if you know what I mean). |
|
|
| mhennessy |
| quote: | Originally posted by Peter Daniel
With regards to have OG centrally positioned, I think it could be swaped with SG pad for more central placement. |
Yes, I'd agree with this ;)
Although the two wires from the input signal "straddle" the OG, which might be uncomfortable. But, these amps are capable of passing a lot of current, and every mil is important - so it's a compromise I'd seriously consider. But, I'm using these in bridged mode, so as I said before, I don't personally feel that strongly about it.
Hometime now - look forward to seeing the updates tomorrow morning ;)
Best regards,
Mark |
|
|
| Peter Daniel |
| quote: | Originally posted by mhennessy
although I prefer the blue) |
If you check pin out on the chip, the pin where you connected blue trace is NC. The best connection point for ground path would be your left, green trace.
Regarding long traces from input resistors to the chip, I might be concerned with them, but obviously it wasn't a concern of National design team when they made evaluation board. Those traces are even longer on their board. |
|
|
| joensd |
roibm,
You can now click my profile to drop me a mail.
Speak to you soon.
Regards
Jens |
|
|
| Jean |
| Count me in for a pair of boards :D |
|
|
| motherone |
I'm probably game for some boards as well.
Peter, I'm with you half way on the resistors, as there are some others I would like to try as well. I'm obviously not a savvy PCB layout designer, but it would be cool to have a board that accomodated various sizes of resistors (Caddocks, Vishays, Rikens, Holcos, etc.) the same as with the capacitors. While I gather from your previous posts you're a form-follows-function kind of guy, it does suck having some beautiful PCB's with resistors standing up on their ends! (Which I'm going to probably have to do with my other LM3875 boards since I have a bunch of other resistors I want to try!).
Is there any way to make the board slightly longer to accomodate bigger resistors? I don't know if space is limited on the size of the PCB (since it's probably going to be 1 board scored in 4 spots from the way I interpreted Brian's post), but I'm certainly willing to deal with a slightly larger board for the tweaking possibilities!! Hopefully this could be done without compromising the ground plane.. |
|
|
| Peter Daniel |
| quote: | Originally posted by motherone
I'm probably game for some boards as well.
Peter, I'm with you half way on the resistors, as there are some others I would like to try as well. I'm obviously not a savvy PCB layout designer, but it would be cool to have a board that accomodated various sizes of resistors (Caddocks, Vishays, Rikens, Holcos, etc.) the same as with the capacitors. While I gather from your previous posts you're a form-follows-function kind of guy, it does suck having some beautiful PCB's with resistors standing up on their ends! (Which I'm going to probably have to do with my other LM3875 boards since I have a bunch of other resistors I want to try!).
Is there any way to make the board slightly longer to accomodate bigger resistors? I don't know if space is limited on the size of the PCB (since it's probably going to be 1 board scored in 4 spots from the way I interpreted Brian's post), but I'm certainly willing to deal with a slightly larger board for the tweaking possibilities!! Hopefully this could be done without compromising the ground plane.. |
Just to see 2 horizontally mounted resistors, you would like to compromise the whole design?
Look at this Pass Ono board. They have standing capacitors, so what's wrong with standing resistors?;) |
|
|
| Peter Daniel |
| Now check this tightly packed Levinson board. Although some of those polystyrene caps are mounted horizontally, whenever space was limited, they mounted them in vertical position. I really don't know why people think that this way of mounting components is compromised? It all depends how it fits in a circuit, but I'd rather compromise placement of 2 components than other, more important aspects of the design. I wouldn't go with bigger board just to accomodate 2 resistors (which you propbably won't be using anyway). |
|
|
| kestrel200 |
| So..to have a bridged 150watt RMS Stereo I'd need two kits?? |
|
|
| BrianGT |
| quote: | Originally posted by kestrel200
So..to have a bridged 150watt RMS Stereo I'd need two kits?? |
This is correct.
--
Brian |
|
|
| kestrel200 |
Brian,
You may be responsible for my divorce if you keep coming up with these enticing, irresistable designs.:devilr: |
|
|
| chipco3434 |
Brian:
Will you have documentation like last time? With pictures? This is especially helpful for real Dumas's.
Mr. Dumass has a question. What's the difference between parallelled and bridged designs? Get technical with me.
Pictures work.
By the way, my little billet clone is simply wonderful. I am piping XM through it to my line arrays. It's magic. I am very happy with the whole project and was very impressed by your preparation of the kits and the docs. OUTSTANDING |
|
|
| grege |
My AKSAs have some vertically placed resistors and I have never heard a bad word about them. I guess vertical resistors allows for a more compact and elegant PCB.
I could start a new theory that using vertical and horizontal resistors mean that the electrical and magnetical fields are at 90 degrees to one anther and therefore reduce the chance of cross talk. :D |
|
|
| neutron7 |
| Just a quick question. Earlier in the thread Peter had a diagram with 2 boards facing each other and mentioned the 2 boards would have mirror image power connectors so you could simplify bridging 2 boards like this. Is it going to be done that way in the final version? it would be quite handy. |
|
|
| BrianGT |
Here is a picture of the National Evaluation board layout, for comparison purposes.
Would the output track where Rf2 connects be considered a T track?
--
Brian |
|
|
| sunil |
Hello,
Could somebody please put in plain simple words what kind of real world output is possible using one board (single chip used in stereo) into 8 & 4ohms, and, 2 boards (two chips, bridged-parallel) into 8 & 4 ohms.
Thankyou,
sunil |
|
|
| roibm |
| quote: | Originally posted by Peter Daniel
Now check this tightly packed Levinson board. Although some of those polystyrene caps are mounted horizontally, whenever space was limited, they mounted them in vertical position. I really don't know why people think that this way of mounting components is compromised? It all depends how it fits in a circuit, but I'd rather compromise placement of 2 components than other, more important aspects of the design. I wouldn't go with bigger board just to accomodate 2 resistors (which you propbably won't be using anyway). |
Are there sockets under some caps on Levinson's board? And if so I would they do such a thing? Upgrade? |
|
|
| peranders |
| quote: | Originally posted by mhennessy
I couldn't resist improving the signal earthing further (possible alternative route shown in green, although I prefer the blue) |
Brain, I'm sorry to say but this green trace is doing no good at all. I had also such ideas when I was younger but those traces are just antennas.
Check also my grounding, green. |
|
|
| mhennessy |
| quote: | Originally posted by sunil
Hello,
Could somebody please put in plain simple words what kind of real world output is possible using one board (single chip used in stereo) into 8 & 4ohms, and, 2 boards (two chips, bridged-parallel) into 8 & 4 ohms.
Thankyou,
sunil |
I guess the reason why this hasn't been done is because it's difficult to predict or guarantee a particular output when variables like the power supply transformer are beyond National, Brian and Peters control. That's why, in a previous post, I explained my working and assumptions. Also, the PCB is not configured for stereo operation (you could mod it, but you'd be better off using one PCB for each channel). But, I would guess the following:
Single board:
8ohms - 60W
4ohm - 100W
Two boards bridged:
8 ohms - 150-170W
4 ohms - 200-250W
Technical bit:
Assumptions: Supply rails +/-40V, falling to +/-35V under load. Note that they might fall more, depending on capacitor value and transformer rating - hence the vagueness of the bridged results.
Assumed around 5V dropout in the IC's, meaning peak voltage swing is 30V, which is around 21V RMS (I also assumed this would vary with load as well). Note that in theory, you should get 4 times the output power when you bridge, but the losses compound, and you normally only get around twice the output. The parallel operation of each amplifier helps, but you still won't get better than 3 times in practice.
To get 40V rails, you'd need 28V secondaries. 30V secondaries would probably result in rails that are too high off-load (check the datasheet). Transformers with 25V secondaries are more readily available - but I would expect the outputs reduce to around 40, 75, 140 and 200W respectively.
As I said above, this stuff is hard to predict, and manufacturers will normally have to experiment with different transformers to arrive at the exact result they're after. So, until we start building and measuring, I wouldn't like to be any more specific ;)
Note that these are RMS figures - under transient conditions, peak outputs will be higher. Heatsinking is very important at these power levels! |
|
|
| mhennessy |
| quote: | Originally posted by peranders
Brain, I'm sorry to say but this green trace is doing no good at all. I had also such ideas when I was younger but those traces are just antennas.
Check also my grounding, green. |
Hi Peranders,
I wonder if you thought I was suggesting adding these traces along with the existing blue trace? Just to clarify, I was suggesting alternatives - I only intended Brian to use one path. More than one would create loops, which is bad (as you say). Sorry if it wasn't explicit in the origional post (although re-reading the bit you quoted, I think it was fairly clear)
Cheers,
Mark ;) |
|
|
| mhennessy |
| quote: | Originally posted by Peter Daniel
If you check pin out on the chip, the pin where you connected blue trace is NC. The best connection point for ground path would be your left, green trace. |
Thanks - I hadn't noticed this - but National board connects these NC pins to ground, which is what Brian had done, so there's probably no real need to worry about this ;)
| quote: | Originally posted by Peter Daniel
Regarding long traces from input resistors to the chip, I might be concerned with them, but obviously it wasn't a concern of National design team when they made evaluation board. Those traces are even longer on their board. |
| quote: | Originally posted by BrianGT
Here is a picture of the National Evaluation board layout, for comparison purposes.
Would the output track where Rf2 connects be considered a T track? |
I think that we're all in agreement (from here, and the previous thread about this board) that Nationals board is not an example of an optimised layout ;)
Yours and Brians however, is :cool:
Cheers,
Mark ;) |
|
|
| mhennessy |
| quote: | Originally posted by Peter Daniel
Now check this tightly packed Levinson board. Although some of those polystyrene caps are mounted horizontally, whenever space was limited, they mounted them in vertical position. I really don't know why people think that this way of mounting components is compromised? It all depends how it fits in a circuit, but I'd rather compromise placement of 2 components than other, more important aspects of the design. I wouldn't go with bigger board just to accomodate 2 resistors (which you propbably won't be using anyway). |
Hi Peter,
I think that mounting resistors vertically in your PCB is fine, and a useful compromise because it contributes to the overall small size. But, just for info, here are the objections to them
Increased suscecptability to RF pickup - this can be very serious in some applications. The problem is caused by having a conductor at right-angles to the ground-plane, which forms a good little antenna
Reduced reliability. Sounds unlikely, but think about how stuff gets serviced. When routing around inside a piece of kit, it's easy for these components to be knocked over so that they touch adjacent conductors. To get around this, some manufacturers use special, pre-formed resistors that have the exposed wire covered in the same enamal as the resistor body (you've probably seen these in el-cheapo kit)
Mechanical stability. For larger components that run hot, there's a greater chance of bad solder joints appearing because the mechanical stress on the joints is greater.
There's probably others as well, but it's still early in the morning for me ;)
Just to finish with some (hopefully) sensible advice, when mounting components vertically, you can avoid some of the issues by deciding which way round to place the component. In the ML example above, I wouldn't mind betting that the long exposed wire was connected to ground, or failing that, the lowest impedance (or "earthy") part of the circuit.
So, for any resistor in your PCB, you can work out the optimum way to vertically mount all of the components, by just thinking about circuit impedances. For example, R4 should have the long lead connected to ground. The long wire of R2 should go to in the IC output node, as the input end of that resistor will be very prone to pickup.
Hope this makes sense and helps,
Mark ;) |
|
|
| jamart821 |
| Count me in for one set of boards. I already have the two chips so I would be also interested in two sets of parts only. |
|
|
| BrianGT |
Here is a modified layout, following some of the suggestions. I worked on it with Peter last night. I tried putting the resistors above the capacitor, but it didn't look good. Instead, the R6 and R7 were swapped to equalize the trace lengths a bit.
As for the T, I think that it is unavoidable, and shouldn't be an issue. I increased the width of the output trace from the IC chip, which should help. I had tried using both layers, but came back to this solution.
The heat relief traces are removed as well.
Any comments?
--
Brian |
|
|
| Peter Daniel |
| quote: | Originally posted by mhennessy
I think that we're all in agreement (from here, and the previous thread about this board) that Nationals board is not an example of an optimised layout ;)
|
It might not be optimised (for nice appearance), but it definitely is an example of a proper layout.
It's just that some small issues you might think affect the performance, in real life may have no importance at all;) |
|
|
| peranders |
Now I really like the layout. :nod: Maybe even a certain person will like it also. One thing though, how about the hole size for the speaker cable? ..and what about decoupling, like 100 nF close the the IC? Have you forgot those? I advice you to put in 2 psc. 100 nF 63 V polyester.
http://www.national.com/ds.cgi/LM/LM4780.pdf
The datasheet in page 5 suggests 100 nF+ 10 uF + 1000 uF. I think you at least should have 100 nF.
I think the pcb is real good looking now and far above in looks and probably also in performance compared to many commercial boards (LM3875/3886). |
|
|
| ALW |
| quote: | | I really don't know why people think that this way of mounting components is compromised? |
As with all things it depends totally on the application.
The two end terminations of most resistor types present a thermocouple junction.
Different temperatures at each end of the resistor will generate a voltage, a small voltage in most cases, but a voltage nonetheless.
How much this matters depends upon what function those resistors perform, their power dissipation, the air flows and particularly their construction.
The way to minimise it is to keep the end terminations at the same temperature, and horizontal mounting helps this, from both a self-heating and ambient perspective.
None of this is to say that the effect is important at AC though, since the EMF generated will be inherently LF, since sudden temp changes are unlikely in most app's.
Andy. |
|
|
| mhennessy |
| quote: | Originally posted by Peter Daniel
It might not be optimised (for nice appearance), but it definitely is an example of a proper layout. |
Actually, it's neither. It's an example of a layout done by a draughtperson, with relatively little regard taken to layout or appearance, other than providing a rudimentory star earth. Traces are much longer than they need to be, compromising performance. I thought this was discussed in another thread? Anyway, even your initial paper draft was miles better than this layout in both regards ;)
| quote: | Originally posted by peranders
One thing though, how about the hole size for the speaker cable? ..and what about decoupling, like 100 nF close the the IC? Have you forgot those? I advice you to put in 2 psc. 100 nF 63 V polyester. |
I agree about hole sizes - but having read previous comments about decoupling, I don't think you'll win that one. I believe Peter has found them unneccesary based on listening tests using good quality electrolytics. I will add just one comment - think about when the electrolytic caps start to age - that's when small parallel decoupling capacitors come into their own.
Brian - I like the new version of the layout. I might have some small comments, but I don't think they're worth voicing at this stage. Just one thing that I notice - you're using vias! Doesn't worry me in the slightest, but what does Peter think? ;)
Best regards,
Mark |
|
|
| carlmart |
It's curious to note, as I just did now, that National seems less "demanding" on the parts requirements for paralleling their chip amps on the LM4780 datasheet.
On their first Application Notes for bridging and paralleling, they specify 0.1% parts and servos. On the 4780 parts are not specified and no servos are used.
Were the other Overture amps so different from one another to demand that tolerance on external parts was so tight? Or were they being too cautious?
Carlos |
|
|
| carlmart |
| quote: | Originally posted by peranders
One thing though, how about the hole size for the speaker cable? |
Who will be soldering the speaker cable directly into the board? You need a short cable to the speaker outputs on box outside, so it needn't be that thick. It's that connector that needs to accept a thick gauge wire.
Carlos |
|
|
| carlmart |
| quote: | Originally posted by BrianGT
Any comments?
|
The pcb layout looks OK to me. But there are some issues that maybe were discussed in other threads, but I missed them.
If this is a NIGC, where are the capacitors that should be at the input and feedback? If I am not wrong this was the same philosophy used on the single NIGC boards, but aren't you running a certain risk by not using those caps?
Carlos |
|
|
| leadbelly |
| quote: | Originally posted by carlmart
On their first Application Notes for bridging and paralleling, they specify 0.1% parts and servos. On the 4780 parts are not specified and no servos are used. |
Bottom right of pg 16 of the LM4780 datasheet...they recommend 0.1% resistors. |
|
|
| peranders |
| quote: | Originally posted by mhennessy
Actually, it's neither. It's an example of a layout done by a draughtperson, with relatively little regard taken to layout or appearance, other than providing a rudimentory star earth. Traces are much longer than they need to be, compromising performance. I thought this was discussed in another thread? Anyway, even your initial paper draft was miles better than this layout in both regards ;)
|
My dictionary hasn't got the word draughtperson and I can't really figure out the word :confused:
Brian, when I think of it, you have unnecessary long traces going in long loops. Since you are going to have this pcb professionally made why don't you shorten all signal traces and uses 15 mil traces between the IC pins? I should also go for more groundplane. Consider also that rather much of your copper areas aren't very usefull. Think electrial field inside these areas.
Not having proper decoupling is like not having your hands washed before cooking dinner. I vote definitely for proper high frequency decoupling. You can bet on that some people which are building this amp will encounter RF/EMII trouble. It's easier not to solder in a cap compared to patch it in.
As a comment about mhennessy's remark: I think it's really good of you designing these pcb's but having eagle control of all parameters yourself is nearly an impossible task, even for a pro. It takes years of practice, maybe a lifetime! I wonder why Mr. D hasn't comment this project? |
|
|
| peranders |
| Brian, why don't you skip the symmetry and place the smoothing caps at the left side only. Most of the power are connected the left side if the IC. |
|
|
| BrianGT |
| quote: | Originally posted by mhennessy
Brian - I like the new version of the layout. I might have some small comments, but I don't think they're worth voicing at this stage. Just one thing that I notice - you're using vias! Doesn't worry me in the slightest, but what does Peter think? ;)
|
As for using vias, I assume you are referring to the two traces on the top layer going to the feedback resistors. I actually forgot to change them back to the bottom layer, after playing around with them earlier. I put them back on the bottom layer now. I doubt that it would make much difference, but I would rather have them away from the power traces a bit more.
What small comments do you have? I am curious. You can e-mail them to me if you don't want to post them here ;)
--
Brian |
|
|
| peranders |
| quote: | Originally posted by carlmart
Who will be soldering the speaker cable directly into the board? You need a short cable to the speaker outputs on box outside, so it needn't be that thick. It's that connector that needs to accept a thick gauge wire.
Carlos |
If you have 0.1 ohms resistors it's probably not very important to have thick wires but I imagine that it looks nicer if you can choose such cables if you want to.
Also Brian, my remarks are probably not very important. I gather the amp will work alright but wouldn't it be a kick to have done everything to make it right? |
|
|
| BrianGT |
| quote: | Originally posted by peranders
Brian, when I think of it, you have unnecessary long traces going in long loops. Since you are going to have this pcb professionally made why don't you shorten all signal traces and uses 15 mil traces between the IC pins? I should also go for more groundplane. Consider also that rather much of your copper areas aren't very usefull. Think electrial field inside these areas.
Not having proper decoupling is like not having your hands washed before cooking dinner. I vote definitely for proper high frequency decoupling. You can bet on that some people which are building this amp will encounter RF/EMII trouble. It's easier not to solder in a cap compared to patch it in.
As a comment about mhennessy's remark: I think it's really good of you designing these pcb's but having eagle control of all parameters yourself is nearly an impossible task, even for a pro. It takes years of practice, maybe a lifetime! I wonder why Mr. D hasn't comment this project? |
The pins on the LM4780 are quite a bit closer than the LM3875, so I don't like the clearance issues with running traces in-between them. I really don't see any reason to need to do this. The only long traces are the input traces.
As for decoupling, I was going to just try soldering decoupling capacitors on the bottom of the board, and see if it makes a difference. I don't really see an optimal place to put them, except on the bottom, on the leads coming from the electrolytics, or the extra mounting holes (the unused set)
--
Brian |
|
|
| BrianGT |
| quote: | Originally posted by peranders
Brian, why don't you skip the symmetry and place the smoothing caps at the left side only. Most of the power are connected the left side if the IC. |
I initially tried, but it didn't seem to come out as nicely. I had the power inputs on the far left, and the inputs on the far right, with the output and grounds in the middle. The board size increases a lot also, since the BG 1000uF STD caps have large footprints. After trying this, I decided that the small tradeoffs of the current layout are greatly outweighed by the symmetry and form factor obtained by the current layout.
--
Brian |
|
|
| Fred Dieckmann |
"I wonder why Mr. D hasn't comment this project?" - P/A
How much can one say about yet another chip amp............:sleep:
I think the things that are important to consider for the layout are in outlined pretty well in the data sheet . Don't confuse ground planes and single point grounding, they are not the same thing. The data sheet mentions the desirability of single point grounding several times I believe.:smash: |
|
|
| Flex |
Hi all,
I have just registered on this forum but i've been following this thread for some time now. And i've also read the last thread about the LM3875 amp.
I'm very interested in building a amp like this but there are a couple of questions,
- How much power does this amp put out with a 8 ohm load?
- How hard will it be it to put volume control on it ?
- What kind of power-supply will it need? (how much Va)
This will be my fist attempt at building an amp, but i allready have experience with building speakers and I have read a lot of documentation about different types of amps.
Most of them are a little to expensive for me at the moment and something like this would be ideal for now.
tnx. :) |
|
|
| Peter Daniel |
| quote: | Originally posted by peranders
and what about decoupling, like 100 nF close the the IC? Have you forgot those? I advice you to put in 2 psc. 100 nF 63 V polyester.
|
That size decoupling cap is best soldered underneath, detectly to electrolytic, if you really needed. Providing special spots for those caps on a board would be compromising simplicity of a ground plane. I got used to attaching components to both sides of PCB (whenever more convenient) and it shouldn't even present clearance problem with such small value. I have my protoype working withaout any extra bypass caps and it prforms fine.
| quote: | Originally posted by mhennessy
Actually, it's neither. It's an example of a layout done by a draughtperson, with relatively little regard taken to layout or appearance, other than providing a rudimentory star earth. Traces are much longer than they need to be, compromising performance. I thought this was discussed in another thread? Anyway, even your initial paper draft was miles better than this layout in both regards ;)
|
I looked at National's layout again and I think a lot of thought went into creating it. If tight packing of components wouldn't be my goal here, I don't think I could improve on that (National's) layout. It is very good for what it was intended. Maybe they just wanted to show that the chip performs even when not every trace on PCB is analysed in depth and optimised to perfection?;)
| quote: | Originally posted by carlmart
The pcb layout looks OK to me. But there are some issues that maybe were discussed in other threads, but I missed them.
If this is a NIGC, where are the capacitors that should be at the input and feedback? If I am not wrong this was the same philosophy used on the single NIGC boards, but aren't you running a certain risk by not using those caps?
|
You definitely don't need input caps if your source is protected against excessive offset. The two parallel channels I've built presented much lower offset than what I got with LM3875 (it was about 24mV). As to the feedback capacitor, I never required it with LM3875 chip either. It's hard to predict if its really needed here, that's why we go with a smaller initial run for testing purposes. After that we will know for sure. |
|
|
| Variac |
Hi Brian,
It occured to me that it might be good to have the 2 rows
of diodes in the power supply facing in opposite directions or
some other idea so that they could be all be attached to
heatsinks. MAybe this would only be important if the bridge were used for 6 100w channels- but many people are probably going to make bigger ampps with this kit. Also would be very useful if someone wanted to use just the power board for a Son of Zen or other big class A amp. Then they would be like Big Parsnips diode boards? (but without snubbers) So only an idea if it doesn't compromise something. What is the capacity in amps for the assembled bridge? Would small heatsinks double the rating?
Second- thank you for my latest order. Excellent counting!!! ;)
BUT (I hope this doesn't drive you over the edge) It's missing the 2 small BlackGate caps for the power supply. They are supposed to be part of the premium bridge kit , right?
Mark |
|
|
| carlmart |
| quote: | Originally posted by leadbelly
Bottom right of pg 16 of the LM4780 datasheet...they recommend 0.1% resistors. |
Pity. For a moment I thought they were cured of engineeretis...
Maybe there's servo hidden somewhere too.
Carlos |
|
|
| netgeek |
It looks great! Yes, it could be tweaked to death - but let's let Brian get on with it...!!! That way we can all build something (in THIS millenium) ... and we can always ***** about it later if it doesn't work out..:) :) :)
When can we place orders? BTW, thanks Brian for going through all the trouble - I know alot of people here appreciate it.
And it will NEVER be "perfect". If you want perfect, you can believe the marketing **** from any number of commercial suppliers and - guess what(?!!) - they lied, it isn't perfect.:D
FASTER!!!!.......:)
Regards,
Bill |
|
|
| roibm |
| quote: | Originally posted by netgeek
It looks great! Yes, it could be tweaked to death - but let's let Brian get on with it...!!! That way we can all build something (in THIS millenium) ... and we can always ***** about it later if it doesn't work out..:) :) :)
|
I prefer to wait longer and something better than the other way around. But that is just me and who is that me is ;) |
|
|
| netgeek |
| quote: | Originally posted by roibm
I prefer to wait longer and something better than the other way around. But that is just me and who is that me is ;) |
Waiting - up to a point - is a good thing... But we've reached a point, I think, where minor quibles are slowing progress.
BTW, my comment about "*****ing" got screened? (Reference to a female dog - starting with "B"). Wow, the filters here are really good - and the moderators must have the "sensibility threshold" set to "MAX"...???
:) |
|
|
| BrianGT |
| quote: | Originally posted by Variac
It occured to me that it might be good to have the 2 rows
of diodes in the power supply facing in opposite directions or
some other idea so that they could be all be attached to
heatsinks. MAybe this would only be important if the bridge were used for 6 100w channels- but many people are probably going to make bigger ampps with this kit. Also would be very useful if someone wanted to use just the power board for a Son of Zen or other big class A amp. Then they would be like Big Parsnips diode boards? (but without snubbers) So only an idea if it doesn't compromise something. What is the capacity in amps for the assembled bridge? Would small heatsinks double the rating?
Second- thank you for my latest order. Excellent counting!!! ;)
BUT (I hope this doesn't drive you over the edge) It's missing the 2 small BlackGate caps for the power supply. They are supposed to be part of the premium bridge kit , right?
|
Mark,
Small heatsinks might double the rating. I will think about it, as I am working on the new diode board design now. I will see what Peter thinks about it.
I will mail you out 2 more BG 4.7uF N caps. Sorry for the mistake.
--
Brian |
|
|
| Peter Daniel |
I was reading that too and it seems like a good idea. I noticed that sometimes the diodes warm up quite a lot so having some sort of heatsink bar would help with disippation. This would require fliping diodes in the other row to have the taps on the inside. It shouldn't complcate the traces too much. But I think that one flat bar piece, 1/8 thick and sized approx 1 x 3 should be enough. Don't go with separate heatsinks as this will add unnecessary bulkiness. Of course adding isolation pads would be required.
The board looks really nice now, isn't it?;) |
|
|
| Peter Daniel |
I have a technical question to those who know more about board design. It seems like adding a track from pin 17 south to connect it with V+ plane would increase current supply to that pin. Yet it creates a loop. Is it good or bad?
Also, would trimming both power planes on the sides, like on LM3875 boards, would be recommended? This would somehow decrease the unused area of the plane and help in directing current where we really want it, towards the caps.
This is more or less theoretical, but this thread is pretty much educational regarding board layout design and we didn't touch that issue yet;) |
|
|
| BrianGT |
| quote: | Originally posted by netgeek
It looks great! Yes, it could be tweaked to death - but let's let Brian get on with it...!!! That way we can all build something (in THIS millenium) ... and we can always ***** about it later if it doesn't work out..:) :) :)
When can we place orders? BTW, thanks Brian for going through all the trouble - I know alot of people here appreciate it.
And it will NEVER be "perfect". If you want perfect, you can believe the marketing **** from any number of commercial suppliers and - guess what(?!!) - they lied, it isn't perfect.:D
FASTER!!!!.......:)
|
heh.. I wouldn't want to rush it. A couple extra days won't hurt anyone. After all, I would guess that over 1/2 of the people ordering the LM3875 kits haven't even started their projects yet.
Be patient, and you will be satisified with the end result.
--
Brian |
|
|
| netgeek |
| quote: | Originally posted by BrianGT
heh.. I wouldn't want to rush it. A couple extra days won't hurt anyone. After all, I would guess that over 1/2 of the people ordering the LM3875 kits haven't even started their projects yet.
Be patient, and you will be satisified with the end result.
--
Brian |
Well, I ordered the LM3875 kit just this morning - and I hope to have it working within hours of receipt - then what..
:) :)
I'll need the LM4780 boards very soon after that just to maintain a "fix"...:) |
|
|
| BrianGT |
| quote: | Originally posted by netgeek
Well, I ordered the LM3875 kit just this morning - and I hope to have it working within hours of receipt - then what..
:) :)
I'll need the LM4780 boards very soon after that just to maintain a "fix"...:) |
Well, I will get your LM3875 kits shipped out to you tomorrow, so you can started. :) I will try my best to feed your addiction, as it makes me feel better about having my own.
--
Brian |
|
|
| bg40403 |
fwiw, Panasonic is offering a new electrolytic capacitor, the "FM".
It has lower impedance than the "FC" but passes a higher ripple current. Low impedance is recommended for the capacitors used as the FC is in the "standard" kit version.
What about higher ripple current? Is that a negative characteristic of the FM cap that outweighs its lower impedance?
I guess a listening test may be in order.
Also, it would be necessary to drop to a 35V rating to get anything above 1000 uF.
Thoughts anyone? |
|
|
| Variac |
| quote: | | . This would require fliping diodes in the other row to have the taps on the insideBut I think that one flat bar piece, 1/8 thick and sized approx 1 x 3 should be enough |
I think flipping a row so that they face both sides of a 1/8 or 3/16" bar is a great idea. Its a simple way to impliment what I was suggesting Also-really easy to through bolt , so only 4 hole need be drilled in the bar other than mounting holes for the bar itself. ( I was envisioning the rows facing out w/ a total of 2 heatsinks- too complicated!!)
It doesn't seem like it would take much if any more room on a circuit board, though Peter cleverly points out that insulators are needed- which also have thickness. If you sized it for an 1/8" bar
with alu oxide insulators, people could also use a 3/16" bar and mica insulators if | | | |
