Thanks for posting the article, koldby.
A sand filled construction would only work out for me if I didn't have size constraints. But the article does show how one can win a lot by gluing together dissimilar materials. I'll line my cabinets with mass loaded vinyl.
Just have to figure out if I want to pony up for the thicker 8mm stuff instead of the 4mm I have already. Adding the MLV will satisfy my weight goals and dampen the ply. A simple window brace construction will be used to keep volume reduction to a minimum. The 30W moving mass is 135 gram. That's still pretty reasonable. About double the moving mass of one array. While one array weighs about 65 to 70 Kg I'm not planning for the subs to weigh 140 Kg each
.
A sand filled construction would only work out for me if I didn't have size constraints. But the article does show how one can win a lot by gluing together dissimilar materials. I'll line my cabinets with mass loaded vinyl.
Just have to figure out if I want to pony up for the thicker 8mm stuff instead of the 4mm I have already. Adding the MLV will satisfy my weight goals and dampen the ply. A simple window brace construction will be used to keep volume reduction to a minimum. The 30W moving mass is 135 gram. That's still pretty reasonable. About double the moving mass of one array. While one array weighs about 65 to 70 Kg I'm not planning for the subs to weigh 140 Kg each
.
You are welcome.
Actually I still think the sand filled panels is worth considering, but nut 3/8" 3/4" 3/8", in total 1,5" (~37mm), but rather two 9 mm plywood an perhaps 10 mm sand?
This would still be very dead and pretty heavy too. considering the small dimension of the sides of your subs.
Actually I still think the sand filled panels is worth considering, but nut 3/8" 3/4" 3/8", in total 1,5" (~37mm), but rather two 9 mm plywood an perhaps 10 mm sand?
This would still be very dead and pretty heavy too. considering the small dimension of the sides of your subs.
True, though that doesn't make use of the fact I still have 19 mm and 15 mm ply in stock (birch ply B-B quality) . It's a little too expensive to completely start over if I still have quite a bit available.
The enclosure will be build with the 19mm ply, with Isomat TS14 MLV glued to it on the inside. The MLV weighs 14 Kg/m2 and is just ~6 mm thick. Isomat TS info
The 15 mm ply will most likely be used for the window braces and doubling up of the front baffle.
. It's a little too expensive to completely start over if I still have quite a bit available.The enclosure will be build with the 19mm ply, with Isomat TS14 MLV glued to it on the inside. The MLV weighs 14 Kg/m2 and is just ~6 mm thick. Isomat TS info
The 15 mm ply will most likely be used for the window braces and doubling up of the front baffle.
Transition to JRiver 22 is complete. Both Stereo and Home Theatre sound as they should, very pleased with this painless transition. Made some alterations to the ambient and main EQ based on some extended listening periods. Basically I've been listening a lot to a variety of material and am quite pleased with the tonality and imaging.
Only did some minor changes to both ambient and mains (re-balancing) and I'm always surprised how big a difference that can bring. With the reflection free zone working out as good as it did you can actually hear very small tweaks.
Still working out small details on sub construction but I'm almost there. Time will be limited though I'll be spending more time in the workshop soon. Exited to see/hear what this will result in.
I'm even considering giving the sub enclosures the same epoxy treatment/boat paint as the arrays, but this time without the fiberglass matt. Just use it as a filler/structure paint to make them match. No aluminium front on these though. It would simply be too much.
After a talk with fluid on his thread I've come to the conclusion that whatever amp I go with (still thinking Hypex UcD for both sub and ambient housed in my Pioneer A-447) will need a form of analog volume control. Just to set the balance between all of the amps without sacrificing bits from the DAC output. That is what I do with both amplifiers I use now and that works out quite well.
Only did some minor changes to both ambient and mains (re-balancing) and I'm always surprised how big a difference that can bring. With the reflection free zone working out as good as it did you can actually hear very small tweaks.
Still working out small details on sub construction but I'm almost there. Time will be limited though I'll be spending more time in the workshop soon
. Exited to see/hear what this will result in.I'm even considering giving the sub enclosures the same epoxy treatment/boat paint as the arrays, but this time without the fiberglass matt. Just use it as a filler/structure paint to make them match. No aluminium front on these though. It would simply be too much.
After a talk with fluid on his thread I've come to the conclusion that whatever amp I go with (still thinking Hypex UcD for both sub and ambient housed in my Pioneer A-447) will need a form of analog volume control. Just to set the balance between all of the amps without sacrificing bits from the DAC output. That is what I do with both amplifiers I use now and that works out quite well.
+1Still working out small details on sub construction but I'm almost there. Time will be limited though I'll be spending more time in the workshop soon
Hugh Dean has some information on turning a linear cermet type pot into a logarithmic volume control. They are reasonably cheap to buy and are of good performance. As this will be for the sub's and ambient channels which are band limited this should be enough. Much better volume controls get expensive fast.
::: Aspen Amplifiers :::
Thanks fluid, would you still recommend this for a crude gain adjust? I only need to be in the ball park as the rest (fine tuning with regards to other channels) will be done inside JRiver. Basically it's only a crude gain adjust. The reason to do it with a volume knob instead of a fixed attenuation is so I can change my mind about the maximum achievable SPL level.
So even a course (meaning less steps) stepped attenuator would work.
All of the day to day volume adjustments will be done within JRiver.
So even a course (meaning less steps) stepped attenuator would work.
All of the day to day volume adjustments will be done within JRiver.
If you can source the Bourns or Vishay cermet pots in dual gang at 100K or 50K they make a very good volume control for a reasonable price. Hugh does or did do a lot of testing so if he recommends them to me it is worth considering.
For myself I am not sure that it is necessary to even go that far. I have an O2 headphone amp that uses an Alps RK097 pot which costs only a few dollars. I have no issues with the audio quality of that device so to me the pot is not destroying the sound. The Alps Blue Velvet RK27 pots are also pretty nice but cost more.
Some people are convinced of the benefits of different types of volume controls so it is up to you what you decide to use. What I would say is that you are listening to your system through an integrated amp that is no doubt using a fairly ordinary pot and are happy with the sound, for a start you could probably even use the one inside the 447 chassis you were going to re-use.
Well I'm looking for a good technical solution that fits the UcD type amplifier. You've given me some food for thought. I've never build an amplifier yet. I can understand there will be differences between the different type of volume controls. All I need is gain control. I guess any offset in response will be undone by the filtering I will use.
I'm prepaired to choose a good solution. Objectively good if you know what I mean. With several different amps I do need to be able to offset gain differences and I would like to keep high enough digital headroom.
Short of getting a multi channel pré-amp that can balance all inputs a seperate control on each amp seems the best solution right now.
I'm prepaired to choose a good solution. Objectively good if you know what I mean. With several different amps I do need to be able to offset gain differences and I would like to keep high enough digital headroom.
Short of getting a multi channel pré-amp that can balance all inputs a seperate control on each amp seems the best solution right now.
Controlling multiple channels at the same time is tricky and the chip based solutions such as the CS3318 or switched resistor chips do a pretty good job of that. One of the best features the Najda has is the ability to have a selectable volume offset for each individual channel and still have all the channels volume go up and down maintaining those offsets.
In your setup now if you turn your main amp up you will lower the amount of the ambient channels relatively speaking unless you also go over and turn that amp up by exactly the same amount.
If you are satisfied with setting all the amps individually (to account for voltage gain differences or relative loudness offsets) and then using Jriver's volume to give you a range of adjustment a simple pot on the input to each amp is the easiest way to do that.
If you are using single ended signals then almost any volume control will work, I would start with a simple pot and go from there if you are not happy
In your setup now if you turn your main amp up you will lower the amount of the ambient channels relatively speaking unless you also go over and turn that amp up by exactly the same amount.
If you are satisfied with setting all the amps individually (to account for voltage gain differences or relative loudness offsets) and then using Jriver's volume to give you a range of adjustment a simple pot on the input to each amp is the easiest way to do that.
If you are using single ended signals then almost any volume control will work, I would start with a simple pot and go from there if you are not happy
Bruno Putzeys share a nice white paper "The G Word, or how to Get Your Audio off the Ground" at this link https://www.diyclassd.com/p/application-notes-white-papers/, maybe not your exactly solution wesayso but at page 10 Bruno share a nice diy quality volume control and the rest of the paper paper is good reading too. On same page there is other nice white papers and application notes for UcD type amplifier.
Controlling multiple channels at the same time is tricky and the chip based solutions such as the CS3318 or switched resistor chips do a pretty good job of that. One of the best features the Najda has is the ability to have a selectable volume offset for each individual channel and still have all the channels volume go up and down maintaining those offsets.
In your setup now if you turn your main amp up you will lower the amount of the ambient channels relatively speaking unless you also go over and turn that amp up by exactly the same amount.
If you are satisfied with setting all the amps individually (to account for voltage gain differences or relative loudness offsets) and then using Jriver's volume to give you a range of adjustment a simple pot on the input to each amp is the easiest way to do that.
If you are using single ended signals then almost any volume control will work, I would start with a simple pot and go from there if you are not happy
Is that inter channel balance in the digital or analog domain? If it's in the digital domain than it isn't any different from what JRiver has to offer, right?
I can adjust each channel individually and use the overall volume to adjust while keeping the inter channel differences. As I have up till now. Thoug I chose an arbitrary max volume level, made sure to use the dac just below clipping and adjust the analog volume from the amp to set it's max SPL level desired for each pair of channels. Fine tuning and left/right balance happens within JRiver.
On the amp I just set it and forget it
. In other words I never touch those dials afterwards.This is what I meant with a pre-amp solution. Technically I'd need to be able to adjust each channel separately. With an overall volume control to adjust the level for listening level. Way overkill here though, as I only need to hard limit the amp to make the best use of the DAC's available bits. What good is it to be able to go louder and never use it? I do not need to go louder than my hard limit.(I could by adjusting the volume pots on the amps) So I don't reserve bits on the upper end of the DAC, I only reserve those needed to be able to EQ and be able to run full throttle with that. (while still having headroom available in the amps).
Though the UcD type amplifiers are set for an arbitrary gain, which one can set by changing a single resistor.
If I had the funds I'd build amplifiers for each pair of channels, balance them by adjusting that gain resistor (or something similar as I'd probably use a different amp topology for the mains) and set it for a maximum allowable level. Not having enough money to get all amps I need the volume pot on the old amp will have to do. But it might be wise to adjust the max level of the sub amp by changing that resistor, right? I can leave in a bit of headroom as it would be easy enough to dial the top level back a little inside JRiver.
Last edited:
Very good information as always from Bruno Putzeys, the design is for a preamp that is using the pot in the feedback loop of the last opamp. A nice circuit to build as a preamp I think there have been group buys for circuit boards at some point.
Information in there as to why the linear cermet pot is a good one to use as there is very good matching between the tracks. As you are only using it to set the level rather than using it as an actual volume control a linear cermet pot by itself could work.
Figure 6 also shows why a pseudo balanced configuration works.
It can be digital or analogue, analogue is in 1 dB steps so a combination of both may be required to set an exact level. So it is a bit different.Is that inter channel balance in the digital or analog domain? If it's in the digital domain than it isn't any different from what JRiver has to offer, right?
I can adjust each channel individually and use the overall volume to adjust while keeping the inter channel differences. As I have up till now. Thoug I chose an arbitrary max volume level, made sure to use the dac just below clipping and adjust the analog volume from the amp to set it's max SPL level desired for each pair of channels. Fine tuning and left/right balance happens within JRiver.
On the amp I just set it and forget it
This is what the analogue control panel looks like
Attachments
The UcD have about 27dB of voltage gain which is pretty standard for a lot of amplifiers. You can change the voltage gain with a single resistor but it is a pretty small SMD one on the bottom of the board. How easily you can change this resistor is down to your own skill with SMD but I would say that removing SMD parts is a lot harder than soldering them down to begin with
Sometimes it is easier to parallel a through hole resistor to get the value you want rather than trying to replace the SMD part. Like I did hereOne other thing you could consider is a circuit from Linkwitz that can give up to 12dB of gain reduction with a single opamp and a few resistors and a linear pot.
Part 10 here
Active Filters
If you use the Hypex power supplies then they have auxiliary regulated supplies on headers that you could use to power the opamp.
Attachments
Yes, I noticed it being a SMD part. I guess I'll make the amplifier first and check out if and how much attenuation I really need.
The linear cermet pot still sounds like a viable option too as it's a set it and forget it kind of thing. I'll rebalance the line arrays too at that time when I introduce the subs. Maybe it's wise to just build the sub amp first, I've been checking the second hand Pioneer adverts to see if I can find a good case at a reasonable price without having to sacrifice my working ambient amp. That way I can steadily grow into an amp replacement strategy.
The linear cermet pot still sounds like a viable option too as it's a set it and forget it kind of thing. I'll rebalance the line arrays too at that time when I introduce the subs. Maybe it's wise to just build the sub amp first, I've been checking the second hand Pioneer adverts to see if I can find a good case at a reasonable price without having to sacrifice my working ambient amp. That way I can steadily grow into an amp replacement strategy.
I do hope more people are following all the different line array threads around here. I'll share a bit on this thread for those that might have missed it.
Starting at this post here: http://www.diyaudio.com/forums/full-range/303417-full-range-tc9-line-array-cnc-cabinet-35.html#post5084747 there was an interesting discussion about wiring schemes and its influence on the impedance curve.
Up until recently I did expect to see only minor differences that I thought were perhaps negligible. Though I did not get an entire clear picture about the actual differences.
As Fluid posted his impedance curve, speculation on this difference started again.
Especially after seeing the impedance curve from Owen (OPC):
BYRTT took some time to sim it in bwaslo's X-Sim: http://www.diyaudio.com/forums/full...c9-line-array-cnc-cabinet-36.html#post5085100
Even with the only difference being the variation between drivers there are differences to be observed in the total impedance curves of the two wiring schemes.
I've asked for the sims done by BYRTT to take a closer look on a driver level. That looks like this:
If I plot 5 drivers in X-Sim in series I get this:
This is with 10 watt applied to see the graph more clearly.
S1 to S5 are in series and form one of the 5 strings of drivers connected in parallel.
If I do the same plots of Owen's parallel drivers:
Here S1 to S5 are parallel and form one of the strings of drivers in series.
Again 10 watt applied to the complete array.
I think Owen (OPC) might be the brightest one here.
Let's look at the watts for both of these cases:
First in series:
versus in parallel:
This is only a first look but I genuinely think Owen was on to something here.
Lets pump up the watts, plot it at 100 watt into the array!
First the series connected drivers:
Then the parallel connected drivers:
So it's pretty obvious there are differences between both wiring schemes. Owen also suggests to separate the series drivers from the parallel wired ones. In this post some background is given as to why Owen came to this conclusion. I certainly don't want to steal the thunder from the current discussion so if any of you want to join in, please do so on fluid's thread. Unless of course fluid gets tired of us... I'd be willing to host any and all discussion. I'm almost convinced here, if any of you have some valuable insight...
All of this is looking interesting enough for me to plan some surgery on a certain pair of arrays I own.
Starting at this post here: http://www.diyaudio.com/forums/full-range/303417-full-range-tc9-line-array-cnc-cabinet-35.html#post5084747 there was an interesting discussion about wiring schemes and its influence on the impedance curve.
Up until recently I did expect to see only minor differences that I thought were perhaps negligible. Though I did not get an entire clear picture about the actual differences.
As Fluid posted his impedance curve, speculation on this difference started again.
Especially after seeing the impedance curve from Owen (OPC):
BYRTT took some time to sim it in bwaslo's X-Sim: http://www.diyaudio.com/forums/full...c9-line-array-cnc-cabinet-36.html#post5085100
Even with the only difference being the variation between drivers there are differences to be observed in the total impedance curves of the two wiring schemes.
I've asked for the sims done by BYRTT to take a closer look on a driver level. That looks like this:
If I plot 5 drivers in X-Sim in series I get this:
This is with 10 watt applied to see the graph more clearly.
S1 to S5 are in series and form one of the 5 strings of drivers connected in parallel.
If I do the same plots of Owen's parallel drivers:
Here S1 to S5 are parallel and form one of the strings of drivers in series.
Again 10 watt applied to the complete array.
I think Owen (OPC) might be the brightest one here
.Let's look at the watts for both of these cases:
First in series:
versus in parallel:
This is only a first look but I genuinely think Owen was on to something here.
Lets pump up the watts, plot it at 100 watt into the array!
First the series connected drivers:
Then the parallel connected drivers:
So it's pretty obvious there are differences between both wiring schemes. Owen also suggests to separate the series drivers from the parallel wired ones. In this post some background is given as to why Owen came to this conclusion. I certainly don't want to steal the thunder from the current discussion so if any of you want to join in, please do so on fluid's thread. Unless of course fluid gets tired of us... I'd be willing to host any and all discussion. I'm almost convinced here, if any of you have some valuable insight...
All of this is looking interesting enough for me to plan some surgery on a certain pair of arrays I own.
So this seems to be an argument for isolating each speaker as Scaena does in their top of the line array. The observation made by Owen was for all drivers in a single enclosure, more specifically a subwoofer design meant to cancel out box vibration while still pushing max air.
One does wonder how those Scaena's are wired....
If we look back on BYRTT's comment about current flowing from positive to negative, the way we wire an array would also influence it's firing sequence. Though that would be a really small difference in an array with separate enclosures.
I guess all parallel drivers would be the theoretical best solution. It would take custom made drivers though.
Hard to believe the results I have right now can get any better, but I'm willing to try!
If we look back on BYRTT's comment about current flowing from positive to negative, the way we wire an array would also influence it's firing sequence. Though that would be a really small difference in an array with separate enclosures.
I guess all parallel drivers would be the theoretical best solution. It would take custom made drivers though.
Hard to believe the results I have right now can get any better, but I'm willing to try!
Last edited:
