Mini-console based on CHR-70 gen1

[mashaffer]

Mark suggested that I chronicle my mini-console project. It is based (speaker wise) on the first generation CHR-70.

Some background:

It is my plan to make a special gift for each of my eight grand children as close to the high school graduation as possible. The first was a guitar amp for my grandson based loosely on the Marshall 18 Watt with a modified plexi front end. The project I am discussing here is the second project which is a mini-console iPod dock for my grand daughter.

The concept is to take the vacuum tube console stereo concept and shrink it down as small as possible while still retaining full spectrum high quality sound but still have it portable enough to move from room to room. It is not to be a portable in the sense of a ghetto blaster but in the sense of easily transportable. It compares to the traditional iPod dock in much the same way that the original Compaq Luggable compares to a notebook.

The design is still in flux and has been discussed in several other places in one form or another.

Electronics and some amp build here...
Mini-console Schematic

Some Layout possibilities here...
Layout

Laborious process of settling on an approach here...
Serious design effort begins on mini-console project - AudioKarma.org Home Audio Stereo Discussion Forums

In the end I decided on a SE KT-88 amp in UL mode with Schade local feedback. The Power amp has been built and has been used in testing various speaker possibilities.

Now for the speaker part:

I started by putting the drivers in a hacked up Onkyo cabinet which is very small and sealed just to get an idea of the sound and capabilities.



They sounded quite good with exceptional imaging and plenty of volume. The mixer that I am using for testing does not have enough gain to drive the amp to full output but still provides very loud listening levels.



It was lacking in bass so I looked into doing a ported box to get down to my goal of usable output to 40Hz. It looked feasible...





So I built a test box. (To be Continued)...

[mashaffer]

Cobbled together a 1/2 cubic foot ported enclosure from scraps on hand and lined it with poly batting...





It worked quite nicely and extended the response solidly below 50Hz but 4" just seemed like it was working awfully hard to get the output.

MiniConsole Speaker Enclosure Beta Test - YouTube

I discussed it a bit in this thread...

Alpair's aren't woofers

So I decided that since this would be used by a young person that I should provide more margin by using helper woofers. I looked at some possibilities like the 6.5" Shielded Dayton Classic and it looked like it would be helpful but when I looked at the cost of adding the drivers and a passive crossover at 100Hz or so I started to think about going back to my original intention of including an integral powered subwoofer instead of just providing the sub output for an external one.

Other than microphonics my main concern had been the added cost and weight of the integral sub. I went back and looked at an 8" subwoofer driver that I had on hand. It was removed from an AR powered sub that I got over 15 years ago and so I dismissed it out of hand as too old to be used in a gift. I took a look at it today and I am having second thoughts.

I had assumed that the surround would be getting long in the tooth but as I look at it it I see no indication of deterioration and I am tempted to give it a try. Buying a small plate amp would really cost no more than the passive crossover and if I didn't have to buy the driver too... maybe.





By using the powered sub I could use a very small sealed enclosure for the CHR70s thus providing some mechanical protection from over excursion and then roll in the sub at about 100Hz.

I am still looking at some other drivers I have on hand for a passive solution and would sure welcome your thoughts on the subject.

[Henkjan]

you could put the CHR70 in 2 liters seald, and add a series capacitor of 680µF to flatten the response and to protect from low bass. this should give an f3 of 80Hz-ish.

some correction of the baffle step might be usefull (1 single RCL will do the trick), if I look at the cabinet size you have, it appears the same as a design I made, there I needed to lower the output around ~1kHz some 3dB to make it sound more neutral. But in a doc this frequency might be somewhat lower

[mashaffer]

Thanks Henkjan.

One of the problems that I am having is that in most alignments I have simulated there is a fairly broad area in the vicinity of 60 to 120Hz where maximum output is Xmax limited followed by a segment where it is power limited and then at the very bottom of course it is Xmax limited again.

What I was hoping for is an alignment where the Xmax limited region was contiguous for easy efficient filtering. In general it appears that a sealed configuration as you suggest is the only one that gives me that behavior. Unfortunately I start getting Xmax limiting at a rather high frequency (around 120Hz typically) and of course the resulting output level is much lower.

Ported alignments don't seem to be very well behaved for the amount of gain that I get however several TL approaches looked promising. Some time in the next few days I will try to get around to rerunning some of the sims and saving the results to post here. One of the most promising was a TL tuned to around 70Hz IIRC. The output levels shown were much higher than the sealed (and even higher than the ported).

If I am remembering correctly it was approaching 100dB in the pass band at just a couple or three watts. I estimate the amps capability at around 8 watts but again in the bass we are Xmax limited to this level. Of course stuffing and the real world will probably shave off some of that output but that should still be adequate.

Going strictly from memory it was a 120cm long tapered line with the closed end at 150cm^2 and the terminus at 50cm^2 driver located 1/3 of the way from the closed end. After converting to inches it looks to me like a two fold (three segment) line of this length should be small enough to work.

IIRC the response begins rising at around 120Hz till the tuning point of around 65 or 70 Hz where it is about 3 or 4dB higher and the of course drops precipitously. If the filtering were placed correctly I suspect that we could get the response to be essentially flat down to the tuning frequency. I could use passive first order in combination with the second order line level filtering to get ultimately a 3rd order response for additional protection.

I have been able to get nice looking responses with longer TLs down to 50Hz or lower but I don't like running it that low as you simply must have a lot of displacement to get usable volume (thus the subwoofer) and the size gets to be problematic.

I ran some simulations on the subwoofer in a TL (it would be in a separate box I think at this point to make its use optional and to retain portability as well as of course limiting the microphonics issue). I measured the parameters that I could some time ago and then guessed at some based on the factory enclosure and its performance therein. Using those parameters it looks like solid 100dB performance down to 30 or 35Hz should be doable in rooms of the size for which it is intended.

I will still keep the sealed approach in mind if my experiments with other approaches do not pan out.

[mashaffer]

OK, got energetic. Will compare 10 liter sealed with 3:1 TLs of length 98, and 120cm in length. I did sims at 108cm also but forgot to save some of the screen shots. Results were intermediate between these two however.

Will look first at 1 watt response graphs.





Clearly we gain a few dB of efficiency out of the TL and even in the shortest one we have useful output to 60Hz and even 50Hz in the longer line. Below 100Hz the sealed box is rapidly losing ground. It rolls of more slowly of course but to make use of that fact would require large amounts of EQ. which would stress the amplifier and no doubt destroy the drivers.

Next we will compare maximum SPL.

[mashaffer]

Here we look at the maximum SPL with 8 watts and 3mm Xmax available. Again order of screen shots is 10l sealed, 120cm TL and 98cm TL. NOTE: the segments in red are Xmax limited.





Both of the TLs provide higher maximum output all the way down to 60 Hz than the sealed can do at around 100Hz. The longer line holds 100dB all the way down to 50Hz. The shorter line give higher output all the way down to 60Hz than the sealed can do at any frequency given the limited power available.

With the tip up in response that the TLs provide we should be able to provide enhanced performance and acceptable safety margins with proper application of filters.

Right now I tend to favor the shorter line for its higher overall output (reducing the temptation to push its limits) and the greater up turn in response which allows for more aggressive filtering.

[mashaffer]

And just for giggles a look at the subwoofer in a 30Hz line.




Nice smooth output down to 30Hz (good enough for the 16' pedal stops even if we miss the 32s ). Guessing at 8mm Xmax it is holding 112 dB down to the cutoff. I just guessed at Xmax on this driver but I think that is a conservative estimate. Output could go higher if I am underestimating.

Will have to see how it acts in the real world.

[mashaffer]

Oh, I almost forgot. Just for fun I ran sims at 150cm for the MA driver.




It can get down there to 40Hz but it just takes too much out of the little fella and maximum output just goes down too far for anything but easy listening me thinks.

[mashaffer]

In looking at the two possibilities (helper woofer and separate sub) I thought to look into the baffle step issue. As I understand it the baffle width is the dimension that is in question regardless of driver offset so that it is not the first surface that is reached by the sound wave that causes it.

Since both channels are mounted on the same baffle (console) is the entire width of the console considered the baffle width? If so this would be mighty convenient as the nearly 3 feet of baffle would put the critical frequency right at the 100-150Hz range where I want to cross to helper woofers. In that case I would just need to use a woofer that is about 3dB more efficient than the MA driver.

But, it seems to me that since the height is less than the width that the height of the console might be the actual baffle width in this instance.

Some guidance would be appreciated.

[markaudio]

Hi M,
Have you got a pic (or drawing/sketch) for the front baffle or console? I'm sort of trying to picture your design.
Cheers
Mark.