The ZAPpulse manual (page 14) says to connect line ground directly to power ground, which seems fair enough to balance the ZAPpulse module. But am I right to assume that grounding only protects against shorts, as long as the power input is pure? I'm running Supra shielded cables from a Monster HTS 1000 power filter, so it should be pure enough.
That's also one of the reasons why I went for the "backwards" power layout. The only thing that will pass the amps on its way forward is a straight, uncut Supra shielded cable, not even "exposed" by an IEC socket. Also, it will pass on the power side of the ZAP. That way, all low level signals will stay in the small space between the amps, _behind_ the middle channel, far away from mains and control circuits.
A DACT CT2 attenuator sits on the rear wall of the channel, run by a stepper motor inside the channel. The volume wheel on the front panel only sends up/down pulses to the stepper motor control, and the input selector sends 12V to relays in the back.
This design should prevent interference pretty well.
Good thing you told me to use equal length power cables, I actually hadn't thought of that. Just figured the shorter, the better. I'll see if I can find a way to rearrange the caps anyways, though, that might make room for a later transformer upgrade, too.
BTW - what are the Rane notes?
That's also one of the reasons why I went for the "backwards" power layout. The only thing that will pass the amps on its way forward is a straight, uncut Supra shielded cable, not even "exposed" by an IEC socket. Also, it will pass on the power side of the ZAP. That way, all low level signals will stay in the small space between the amps, _behind_ the middle channel, far away from mains and control circuits.
A DACT CT2 attenuator sits on the rear wall of the channel, run by a stepper motor inside the channel. The volume wheel on the front panel only sends up/down pulses to the stepper motor control, and the input selector sends 12V to relays in the back.
This design should prevent interference pretty well.
Good thing you told me to use equal length power cables, I actually hadn't thought of that. Just figured the shorter, the better. I'll see if I can find a way to rearrange the caps anyways, though, that might make room for a later transformer upgrade, too.
BTW - what are the Rane notes?
I think it won't hurt if you try it either way and find out what actually works best.
Clearly you've put some thought into the layout, maybe it would be even better if you used both techniques?
Yeah equal wire length=equal impedance, maybe more important at higher currents to keep things nicely balanced, especially if it turns out the wire gauge matters a little more than it should.
http://www.rane.com/note110.html
Sorry man I have to get running, I'll let you navigate the site to find the other app note I mentioned. Does you a favor anyway, you'll probably find more there that you'll want to read? I think they're great notes!
Cheers
Clearly you've put some thought into the layout, maybe it would be even better if you used both techniques?
Yeah equal wire length=equal impedance, maybe more important at higher currents to keep things nicely balanced, especially if it turns out the wire gauge matters a little more than it should.
http://www.rane.com/note110.html
Sorry man I have to get running, I'll let you navigate the site to find the other app note I mentioned. Does you a favor anyway, you'll probably find more there that you'll want to read? I think they're great notes!
Cheers
Thanks, mate! I've looked through it, and I understand a bit more now. What I think I'll do is connect signal ground to power ground on the ZAP, as the manual says, and get a three position switch so I can turn the whole grounding direct, through the filter or off. Atleast that way I'm sure.
I have a feeling I'll leave the whole thing ungrounded, though, considering both this amp and all signal sources are fully unbalanced.
I liked their final solution, though:
When all else fails, digitize everything, use fiber optic cable and enter a whole new realm of problems.
I think I'll stick to electrical
I have a feeling I'll leave the whole thing ungrounded, though, considering both this amp and all signal sources are fully unbalanced.
I liked their final solution, though:
When all else fails, digitize everything, use fiber optic cable and enter a whole new realm of problems.
I think I'll stick to electrical
Hi Lars Bjerke: Nice layout. Especially the use of a motorised DACT CT2 attenuator, which sounds very interesting and very appealing?
Chris is right, you should definitely mains ground your chassis, with either a cap or wire connection, but, there is no need to directly ground the PS 0V (ground) rail to mains ground (as shown in the diagram).
I think that your layout is fine, as long as you run the mains active and neutral wire through the channel. Yes, if you had been able to lay the caps front to back, rather than side to side, that would have been better. As Chris said make the wires to the amp boards short & heavy gauge, as specified <5cms (if possible), and all should be OK?
Chris is right, you should definitely mains ground your chassis, with either a cap or wire connection, but, there is no need to directly ground the PS 0V (ground) rail to mains ground (as shown in the diagram).
I think that your layout is fine, as long as you run the mains active and neutral wire through the channel. Yes, if you had been able to lay the caps front to back, rather than side to side, that would have been better. As Chris said make the wires to the amp boards short & heavy gauge, as specified <5cms (if possible), and all should be OK?
Well, here's an updated model (click for bigger):
You can see the inlet for the mains cable in the upper right corner, and as you can see from the other side, it will be some cm away from the caps, and - as I said - it's a shielded cable. And, what little interference it may leak will only affect the PSU and speaker wires.
If I run it through the channel, which in it self sounds good, it would have to pass directly under the RCA connectors, relays and attenuator, doing far more damage on signal level. I know I suck at explanations, but I hope you understand what I mean...
I've rearranged the wires on the caps a bit, so the longest stretch will be about 9 cm. That's actually better than any other cap orientation I came up with, since I don't have room to put all eight beside eachother. With 4 mm speaker cable, those few cm really shouldn't matter.
An externally hosted image should be here but it was not working when we last tested it.
You can see the inlet for the mains cable in the upper right corner, and as you can see from the other side, it will be some cm away from the caps, and - as I said - it's a shielded cable. And, what little interference it may leak will only affect the PSU and speaker wires.
If I run it through the channel, which in it self sounds good, it would have to pass directly under the RCA connectors, relays and attenuator, doing far more damage on signal level. I know I suck at explanations, but I hope you understand what I mean...
I've rearranged the wires on the caps a bit, so the longest stretch will be about 9 cm. That's actually better than any other cap orientation I came up with, since I don't have room to put all eight beside eachother. With 4 mm speaker cable, those few cm really shouldn't matter.
novec said:
Ok, Looks to me like you're in the design phase of a custom case, so please dont' mind if I comment.
Judging by the drill spacing and aparent thickness it seems you've done some homework on keeping out RFI. You're already paying alot of money for a very quiet case. That's alright so far.
While shielded cables/wires are great, it's kind of a band-aid. Like when you design an amp, typically you aim for the lowest noise open loop, minimize all sources as much as possible, and then you wrap a little feedback around it to really burry it.
So my point with that is start with the best possible layout you can, and then you'll get your moneys worth out of your case.
I few points before I continue with that.
You should also be considering safety along with your layout, as good design practice. I see what you're going for with the way you've wired the caps, I dislike the way your rail leads cross themselves!! Should it ever heat and the insulation fail, or just through hours of extreme SPL kicking it around and wearing thin..
There's a chance that even the high frequency ripple currents could induce eddy currents in your power wires where they cross each other now, further heating them.. and it won't sound good either.
Why loop the AC all around the case, with every inch you increase the odds of a mean short! Keep that as short as possible, that means have your transformers right there where it enters.
As a bonus you get lower noise, you said yourself, it will only induce noise/ugly hum into your power supply /rails/caps... that's your source from which all water flows, you're polluting the source upstream, and taking a hit on safety.
I find you're throwing away alot of real estate ($$$) in there with that big center division. I really dont' see a need for it?
You could use a T instead to seperate both power supplies while wasting the least space, maybe this will give you enough room to arrange your caps better.
What I think I'd rather do though, is use the T to seperate your modules! Who cares if the supplies sit next to each other. It's better to have wires crossing perpendicularly as well as opposed to a more parallel run, should you have to cross any.
Then bring your rails through the plate, with nice rubber donut type isolators.
I kid you not the ideal would be having your inputs come in at the front of the amp right next to the modules.... but since you're using shielded.... I'll let you come up with how to route them from the back if you find that's too ugly or inconvenient.
Regards,
Chris
Hi Lars Bjerke: I forgot to say that the wires from the PS caps to the amp boards need to be the same length (anyhow, close to the same length is good).
Yes, if you can lay the PS caps front to back, that should solve a couple of the wire lay issues that Chris wrote about.
Lars, are the motorised DACT CT2 attenuators remote controlled?
Yes, if you can lay the PS caps front to back, that should solve a couple of the wire lay issues that Chris wrote about.
Lars, are the motorised DACT CT2 attenuators remote controlled?
Damnmit... As you might have figured, I'm one of those guys who thinks "my solution is best, no matter what people who actually know what they're talking about says". Luckily for me, I'm gonna break that habit right now.
Just as a comment before we start: Cost is no object on this case, simply because there is none. I get the aluminum for next to nothing (or free) at school, and that's where the everyday accessible CNC mill is. The anodizing part will be done at my kitchen table, thanks to this guy.
The reason why I can't put all caps in a row across the case is pretty simple - I want to keep it at 420 mm width to match the rest of my system. Up it a cm or two, and they fit like a foot up my *** - eh, hand in glove, sorry... However, that means the power output from the caps would be 10 cm from either side, which in turn puts the modules 10 cm from either side, and we end up with the same placement problem.
Placing the transformers in the back simply isn't a viable option withouth moving the amp modules to the front, which means everything on signal level will have to cross the power lines. I simply think it's a better tradeoff to have the AC wire interfere on the power cables rather than the signal cables. If the shielding on AC wire is a bad thing (I don't know much about these things, I just thought shielding sounded good ) it can easily be stripped.
A near perfect design would be to have the two transformers in the middle back, with the ZAPs and input selection on either side. Unfortunately, they have to be adjacent to eachother, because they both have to be connected to the same CT2 attenuator. Then we'd be back to the same signal level criss-crossing problem I mentioned earlier.
I've designed the channel to contain and shield the stepper motor, driver circuits, IR receiver and separate 12V power supply, but exactly where it's placed is rather irrelevant. KLe: The DACT is originally manual, but yes, I've built a remote controlled motor on it. However, if you add $225 when you buy the DACT at DIYcable.com, you get a remote controlled combo. It might be available at the Australian distributor, too.
I'm working on some new designs now, I'll post them as soon as they're done.
Just as a comment before we start: Cost is no object on this case, simply because there is none.
I get the aluminum for next to nothing (or free) at school, and that's where the everyday accessible CNC mill is. The anodizing part will be done at my kitchen table, thanks to this guy.The reason why I can't put all caps in a row across the case is pretty simple - I want to keep it at 420 mm width to match the rest of my system. Up it a cm or two, and they fit like a foot up my *** - eh, hand in glove, sorry... However, that means the power output from the caps would be 10 cm from either side, which in turn puts the modules 10 cm from either side, and we end up with the same placement problem.
Placing the transformers in the back simply isn't a viable option withouth moving the amp modules to the front, which means everything on signal level will have to cross the power lines. I simply think it's a better tradeoff to have the AC wire interfere on the power cables rather than the signal cables. If the shielding on AC wire is a bad thing (I don't know much about these things, I just thought shielding sounded good
) it can easily be stripped.A near perfect design would be to have the two transformers in the middle back, with the ZAPs and input selection on either side. Unfortunately, they have to be adjacent to eachother, because they both have to be connected to the same CT2 attenuator. Then we'd be back to the same signal level criss-crossing problem I mentioned earlier.
I've designed the channel to contain and shield the stepper motor, driver circuits, IR receiver and separate 12V power supply, but exactly where it's placed is rather irrelevant. KLe: The DACT is originally manual, but yes, I've built a remote controlled motor on it. However, if you add $225 when you buy the DACT at DIYcable.com, you get a remote controlled combo. It might be available at the Australian distributor, too.
I'm working on some new designs now, I'll post them as soon as they're done.
Hi,
Well I'll let you work out the fine details and the ultimate choice is of course yours to make.
I do agree with the PSU's in the back towards the middle and the zaps in the front.
I didn't say shielding was bad, it's just I think it's all your prior setup would have had to rely on, when your first priority should be a sound layout from the very beginning, to both minimize the noise interference as much as possible and also for safety. Luckily they seem to go hand in hand on this one.
If you have no choice but to add your inputs to the back as well you can easily put your mains input towards the middle and split it off to both toroids, from there to your rectifiers, place your caps facing the transformers if possible and that will let you use the shortest possible wire from the rectifiers. Then towards the sides you can t off the caps from there ... bend the wire up and over them if you want, through your reversed T plate etc.. and go right to your modules with them. So far nothings crossed at all that I can see.
Put your ground lug in the middle.
Then you can bring in your shielded inputs at the back and run them right up to the front well away from your rectifers and mains input, just the transformer but it's smart to shield those too.
I still dont' see a single crossed wire..?
The input wires will pass by your caps but like I said it'll be in a perpendicular fashion, and they'll be shielded.
I could live with something like that anyway.
Cheers,
Chris
Well I'll let you work out the fine details and the ultimate choice is of course yours to make.
I do agree with the PSU's in the back towards the middle and the zaps in the front.
I didn't say shielding was bad, it's just I think it's all your prior setup would have had to rely on, when your first priority should be a sound layout from the very beginning, to both minimize the noise interference as much as possible and also for safety. Luckily they seem to go hand in hand on this one.
If you have no choice but to add your inputs to the back as well you can easily put your mains input towards the middle and split it off to both toroids, from there to your rectifiers, place your caps facing the transformers if possible and that will let you use the shortest possible wire from the rectifiers. Then towards the sides you can t off the caps from there ... bend the wire up and over them if you want, through your reversed T plate etc.. and go right to your modules with them. So far nothings crossed at all that I can see.
Put your ground lug in the middle.
Then you can bring in your shielded inputs at the back and run them right up to the front well away from your rectifers and mains input, just the transformer but it's smart to shield those too.
I still dont' see a single crossed wire..?
The input wires will pass by your caps but like I said it'll be in a perpendicular fashion, and they'll be shielded.
I could live with something like that anyway.
Cheers,
Chris
Last redesign:
Yes, I know I forgot to ground the ZAPs...
Mains wire will go from the rear connector an forward on the very left, on the wall over the left cap, and T to the transformers immediately after.
An externally hosted image should be here but it was not working when we last tested it.
Yes, I know I forgot to ground the ZAPs...
Mains wire will go from the rear connector an forward on the very left, on the wall over the left cap, and T to the transformers immediately after.
Hi Lars Clausen: I was wondering what the input impedance of the 700XE will be?
You produced a low noise version of the 2.3SE with an input impedance of 2K. Is that correct? Will a low noise version of the 700XE be necessary?
Can the BHC 4 pole T-Net caps be connected in parallel. If so, what do you think of this approach (eg. pros and cons)?
You produced a low noise version of the 2.3SE with an input impedance of 2K. Is that correct? Will a low noise version of the 700XE be necessary?
Can the BHC 4 pole T-Net caps be connected in parallel. If so, what do you think of this approach (eg. pros and cons)?
Hi all
I think switching power supply work for class d amp
pls. www.alab-pro.com for switching powersupply
I think switching power supply work for class d amp
pls. www.alab-pro.com for switching powersupply
Thankyou Lars: New girlfriend, sounds good. They do have a nice way of occupying our time, how true?
So ... Can the input impedance of the 700XE be increased, eg. say from 5K to 15K, by simply using a 10K input resistor?
Lars you forgot to answer this question: Can the BHC 4 pole T-Net caps be connected in parallel. If so, what do you think of this approach (eg. pros and cons)?
So ... Can the input impedance of the 700XE be increased, eg. say from 5K to 15K, by simply using a 10K input resistor?
Lars you forgot to answer this question: Can the BHC 4 pole T-Net caps be connected in parallel. If so, what do you think of this approach (eg. pros and cons)?
KLe: I was referring to this article, posted earlier here in this thread by ClassD4sure: http://www.partsconnexion.com/jensen/jensen.html#faq
If you add a 10 k resistor, you will need 3 V rms for 350W, without the 10k resistor you need only 1 V rms for 350W.
Elso: Sure, but still i will not show you my prey .. he he
But i will say that she is a very beautiful Norwegian girl.
If you add a 10 k resistor, you will need 3 V rms for 350W, without the 10k resistor you need only 1 V rms for 350W.
Elso: Sure, but still i will not show you my prey .. he he
But i will say that she is a very beautiful Norwegian girl.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.