WinISD Advice - TL606 box

[Sharkfinger]

Need some help with WinISD.

Having had such marvelous tonal results with upgrading an Ashdown combo, I wondered what would happen if I put that BMS 15N620 speaker into a slightly bigger (deeper at the back) box.  As good as it sounds, I'm finding I usually have to take my 1x15 extension cab to comfortably get enough volume and bottom end. On its own, the combo has nearly enough of both, but not quite.

My intention is to have the same tone as the combo but have that little bit more output and bottom end, that would allow me to leave the extension cab at home for the majority of gigs, which are usually small to medium sized venues.

I've got a TL606 box lying around (pictured below) and just for sh*tz and giggles, just put the driver in and used it at a gig.  There is a bit more bottom end and volume but that seems to be at the expense of the 'crisp' top end.

I've now downloaded and installed WinISD and found a driver file for the speaker here. Starting a new project with this driver (which says the 'driver data fails integrity check' but lets me carry on), I choose 1 driver with 'normal' placement, 'vented' design and an alignment of 'QB3 Quasi-butterworth', which seems to be default.  

I tweak the box volume down a bit to reflect the internal size of this cab (about 86 liters) and leave the tuning frequency at the recommended 46.24Hz.

Now, it seems to want offer me a single round vent of 10cm with length of 5.27 cm.  Now, this cab just has a 'slot' opening at the bottom, which is just an opening, no 'vent' as such, which is probably why it doesn't sound great.

I can change the shape to be square and tweak the setting to be the same width and height as the opening at the bottom but it suggest a length that is longer than the cab is deep.

No problem, can just seal that easily enough and re-port it.   Does the round vent above sound about right to  you guys?

 

 

[SubsonicSimpleton]

Your cab is already ported - rather than sealing the existing port and cutting a different shaped one, surely it would make more sense to tune the length of the exiating port appropriately. There are both commercial and DIY cabs that use slot ports, here is a link to the download page for the Fearful range of cabs which use slot ports, take a look at the plans and it should become clear how the slot ports are implemented.

http://greenboy.us/fEARful/DL/

 

With a little ingenuity you could make a two piece shelf that could easily pass through the existing speaker opening (I would layer two thinner pieces of ply like this - viewed from the cab front)

AAAAAABBBBBBBBBBBBBBBBBBB

AAAAAAAAAAAAAAAAAABBBBBBB

 ..............S..........................S................

Both pieces (A) glued together beforehand into one piece, likewise with pieces (B), then you have a large mating surface to glue the two pieces together inside the cab. If you size the pieces so the long layer of each piece extends to cover both braces(S), you will be able to screw the port shelf down securely to dry fit and finesse the tuning before committing to fixing the shelf more permanently within the cab or glueing the two halves together.

[Sharkfinger]

4 minutes ago, SubsonicSimpleton said:

Your cab is already ported - rather than sealing the existing port and cutting a different shaped one, surely it would make more sense to tune the length of the exiating port appropriately.

....

I did think of that, which is why I changed the port to be square and then tweaked the dimensions to make it rectangular (last WinISD image). However, it came up with a 'vent length' of 44.54cm, which is longer than the cab is deep.

Could make it shorter of course, but by how much?

[SubsonicSimpleton]

Is the port really 8.2 cm tall internally?

There is an enormous difference in port cross sectional area between the two porting suggestions, you could try plugging in a smaller number for the square port (either just the central portion, or just the two corner portions) bearing in mind that you can reduce the port cross section by adding material under the port shelf fairly easily.

Really this needs some input from the forumites with more loudspeaker design knowledge than myself calling @stevie @Bill Fitzmaurice who can probably shed a more expert view on the port tuning aspect - the circular port suggestion has a much higher resonant frequecy than the square one, so there may be other factors to consider beyond taking the numbers from winIsd at face value.

[Sharkfinger]

That's a good suggestion: might block off shelf at the front and use the two end parts as square ports.  

The opening height i stated might have been a bit ambitious, it's probably only about 6-7 cm high.  Width of each end section is probably only about 12cm, so I've modeled that to much better results. 

Best thing is, it's not a lot of work to find out if it's viable.  Invested a fair amount of work into the Ashdown combo, so only want to trial this to see if it's worth 'tarting it up' to look better. I would also turn it into a combo by sticking a box on top for my amp but won't do any of that until I know it's worth the investment in time.

A bit of high-powered image editing.....

That what it would look like from the inside.  I know: you would never know I used graphical editor.....

Edited August 9, 2019 by Sharkfinger

[paulbuzz]

If, in WinISD's 'Box' tab, you tweak the tuning frequency of the box, it will automatically alter the length of the suggested port.
By trial and error, you could set this to match the length of the existing port of the box, and will then be able to see how much this has affected the properties of your cab - you may find that the results are acceptable without any port retuning being required!

You could also use WinISD to model your existing setup with the Ashdown combo cab; then you would be able to see what differences you can expect from the TL606 box - or maybe you've done this already...?

[paulbuzz]

Ah, looking at your pics again, I see now that it seems the existing port has no 'length' as such - so just the thickness of the baffle... 18mm?

So as SubSimp has suggested, you can create length in the port by fixing a shelf on top of the existing port braces; this could be any length you like, up to the limit imposed by the depth of the cab itself.
(You obviously shouldn't bring it so close to the back of the cab that it restricts the area of the port!)

As I posted above, you can model the results for any given port length by adjusting the box tuning frequency until you match your proposed port length.

[Sharkfinger]

25 minutes ago, paulbuzz said:

As I posted above, you can model the results for any given port length by adjusting the box tuning frequency until you match your proposed port length.

That tuning frequency was pre-populated.  Did try messing about with it but as I didn't really know what I was doing, I reset it. 

I'll make sure I've got the 'slot' dimensions right, then play with the tuning frequency until I get to a 'vent length' that looks reasonable.  

Edit: Hmm, not sure about that: with a vent length of 27cm , it will be tuned to about 51Hz.  Seems a little high to me.

Edited August 9, 2019 by Sharkfinger

[paulbuzz]

Yep, when initially setting up your WinISD project, the tuning frequency gets pre-set to create the default QB3 Quasi-butterworth alignment you mentioned in your original post. This represents tuning the cab to the lowest possible frequency with a smooth response before it starts rolling off.

This is often a desirable cab tuning, but is certainly not the only possible choice - you might choose a different tuning for a different-but-lumpier bass response, or for increased power handling, or for other reasons - such as because you can't fit the optimal port into your cab...!

[Sharkfinger]

Interesting! 

Well, think I'll settle on 50Hz, as that's a vent length of 28.4cm.

Should I subtract the depth of front panel, the 18mm ply?  Should I also subtract the width of the supports?  2x18mm.

Edited August 9, 2019 by Sharkfinger

[Sharkfinger]

Interestingly, if I just have a 'centre'  vent and block the side ones, I can have  much shorter vent length and tune it lower, it would seem.

Edited August 9, 2019 by Sharkfinger

[paulbuzz]

Your port will effectively start from the front of the baffle, so the thickness of the baffle is included in the overall port length.
So for calculating the depth of the 'shelf' you'll add to give length to the port, yes, you would want to subtract the baffle thickness.

If you reduce the cross-sectional area of the port, this will increase the velocity of the air moving in the port.
If the port air velocity gets too high, it creates an audible effect known as 'chuffing'.
The port air speed is modelled in WinISD, which you can see by going to the item 'Rear port - air velocity' in the pulldown menu. (The menu item that you currently have set to 'transfer function magnitude').
[ Since this only applies at higher amplifier powers, to see the correct port velocity you will need to set 'system input power' in the 'signal' section of WinISD to something like the max output power of your amp. ]

Unhelpfully, opinions differ (as ever!) as to what a maximum acceptable port velocity is.
Unfortunately I can't currently remember what sort of figures are regarded as ok! You might need to get input from someone more well-versed in these matters - eg Bill Fitzmaurice or Balcro or Phil Starr... fortunately this forum is filled with people more knowledgable and less gormless than me...! 😁
 

Edited August 9, 2019 by paulbuzz

[paulbuzz]

With your proposed smaller port area and lower tuning, I see that the port velocity becomes very high at low frequencies.

Also, with your lower tuning, if you go to the 'maximum power' item on the pulldown, you can see that there's a big dip in the power handling centred at 67Hz.
Raising the cab tuning reduces this dip, whilst also reducing the port length required, but also increasingly producing a hump in the frequency response just above the cutoff frequency.

So it's all a question of compromise!

I'd be inclined to stick to the full width port (in order to keep the port velocity as low as possible).
I'd probably start by initially making the port length nearly as long as you can before it starts reducing the port area  by getting too close to the back of the cab. (So maybe reaching up to about 80mm from the inner rear wall of the cab?)
Then tweak the tuning (port length) to get a satisfactory combination of frequency response (transfer function), port velocity, and maximum spl/power.

And create a WinISD model of your existing Ashdown-combo-based setup for side-by-side performance comparison!

[Sharkfinger]

Manage to understand couple of things out of that.

Using the full width of the opening (45cm) and the actual height of 5cm, along with an 'acceptable' tuning frequency of 46Hz, gives me a vent length of 25cm, from which I'll subtract the depth of the baffle.  I'll see where I go from there.  Don't have a gig for a couple of weeks so have a while to do it.

Don't have the original model for the ashdown, as Stevie did it for me.  Don't know for sure what it's tuned to.  But I do notice that to get the port length to 15cm, as advised, the tuning needs to be 56Hz, so that must be what it's tuned to.

Interestingly, with all the same porting, etc the TL606 has more 'bottom' but the ashdown has a smaller 'dip' in the maximum power. 

[Sharkfinger]

Interestingly, making the porting the same as the ashdown, then changing back to the long slot makes no difference to the 'transfer function' or 'maximum power', as long as the tuning frequency stays te same.  So looks like there's no point in changing the port type. That's saved me a lot of work (which I wasn't going to put in anyway).

[paulbuzz]

Sorry if I've been confusing or insufficiently clear!

Yep, I think you have made several correct observations:

1) The box volume and port size (area and length) define the box tuning, and correspond directly to the frequency response ("transfer function") and power handling.

2) Deciding on a box tuning is a tradeoff between:
     a) Lower tuning ---> More extended (deeper) bass response with a more gradual rolloff;
                                        An increasingly large dip in the power handling capability just above the box tuning frequency

     b) Higher tuning ---> An increasingly large hump in the bass response just above the cutoff frequency, with a sharper rolloff below it;
                                         Better power handling capability at the weak spot just above the box tuning frequency

    You've noticed these differences between the models for your Ashdown combo cab and the TL606 cab!

3) Having defined the size of the ports(s) in terms of area and length, the exact details (in terms of shape and number) are irrelevant to the tuning (and therefore to the frequency response and power handling).
    These details do, however, affect the port air velocity.

 

As you may have discovered, WinISD allows you to load multiple projects simultaneously. So if you load up one project with your Ashdown-based cab (you know the box and port dimensions, so you can easily create this), and another project with the TL606 box  ( volume = 85.5 litres; tuning still to be finalised but maybe somewhere around 50Hz  ), you can then see what differences you get.

Since you're aiming for a bit more bottom and volume, the most vital graphs are those for transfer function and  maximum SPL, though don't forget to also check that the port air velocity isn't too wildly off the scale!

Hope this makes some kind of sense!

[Sharkfinger]

Perfect sense, thanks. I'll let you know how I get on.

[Balcro]

Hello Sharkfinger,

In reply to your first question, the size of the round port, the answer I get is no. I set the box volume to 86L and tuning to 46.24. The result for me was a round port of 13cm diameter and a length of 12.08cm.

I'm not sure about the mix of T/S parameters and how they might mix things up, but worthwhile having a double check on them. The one that gets me is usually Cms. I get the millimeters & micrometers mixed up. How much that has a bearing on your overall results and the dreaded "driver fails integrity check" I couldn't say.

When I tried the assessment for a square / slot port using your dimensions of 45cm x 8.3cm, I ended up with winISD quoting a length of 44.81cm. Obviously too long, but there was a lot scope to reduce the length without a bad efffect on the "rear port air velocity".

 

 

[stevie]

I modelled the Ashdown and EV cabs based on their interior volumes and don't think there's much to be gained from going to a bigger cab. The extra volume will give you an extra 1dB at 50Hz and 40Hz, assuming the same tuning frequency. So that others can do the comparison, the Ashdown is 68 litres gross with two 100 dia. x 150mm ports (tuned to around 51Hz).

You'd get more or less the same result by lowering the tuning frequency of the Ashdown cab, that is, extending the length of the ports to 180mm.

Also, check very carefully for leaks, as they can cripple the low-end response. Make sure everything is gasketed and completely sealed and make sure there is no wadding within six inches of the port.

 

[Sharkfinger]

3 hours ago, stevie said:

You'd get more or less the same result by lowering the tuning frequency of the Ashdown cab, that is, extending the length of the ports to 180mm.

Also, check very carefully for leaks, as they can cripple the low-end response. Make sure everything is gasketed and completely sealed and make sure there is no wadding within six inches of the port.

Didn't hold up much hope of a great benefit with this TL 606 box, so I didn't invest a great deal of time thankfully.

doubt the ashdown leaks, as it's vinyl covered but I can check the ports.

I'll give extending the Ashdown ports a try, as I love the way that sounds. Not sure how I'm going to do that, though.

As for the wadding, do you mean at the base of the port (on the inside)?

[stevie]

Even a small leak can make a difference. They can happen if there's no foam gasket around the driver or, in a combo, if the hole that the wires pass through to the amp isn't sealed.

It's not that easy to see, but the wadding is probably OK. If there's some at the side of the ports (out of view in the photo) you could remove it. You don't want anything close to the end of the port.

To extend the ports, wrap some thick cardboard round the plastic ports and tape it to give you the desired total length. If it works, you'll have to find a more permanent solution using something like Evostik and gaffer tape to make sure the cardboard doesn't vibrate in use.

 

[Sharkfinger]

Great advice, I'll do that.  I meant to buy some of that foam strip, so I'll try that.

BTW, did you download the driver file from the same place I did or did you put the values in yourself?  In which case, any chance you could share your file?

Edited August 15, 2019 by Sharkfinger

[stevie]

I used the parameters published by BMS. I have one of these drivers (which is why I know how good they are) but it's a 4-ohm version. I measured the TS parameters and they were reasonably close to the BMS figures. If you check that what you have is similar to what's in the BMS spec sheet and you should be all right. Your WinISD results look fine to me.

Another thing worth mentioning, because not many people are aware of it, is that the low frequencies from a large, rear-ported cab don't develop fully until you're well in front of the cab, say 10 to 12 feet. So try listening from that distance and you'll hear the lows from the port becoming louder.

[Sharkfinger]

I extended the shelf of the TL606 box by the amount specified by WinISD and as if my magic, it now almost sounds like the Ashdown.

Gigged it this weekend and indeed it does have a tiny bit more bottom end and volume.  I still plugged in my extension cab, though, as that gig just needed more 'oomph' as we didn't have the full PA that night.  There really is nothing like a 2x15 to my mind.

Is there no cure for a bigger box and more speakers?  I might experiment with making with a compact 2x15 cab....  and maybe make it into a combo.

[stevie]

You might like to try the cab with a bigger amp. Your BMS driver is rated at 500 watts. I don't know what amp you're using, but that might be the limiting factor rather than the speaker. Try something that produces 500W into 8 ohms and see what happens.