Combo lockdown project

[Mottlefeeder]

My usual rig is a pair of 10s  powered by an Ashdown MiBass 500W amp, but I can also use one of the 10s with a small battery amp for acoustic nights and small gigs. Unfortunately, at 300 * 300 * 450 mm, even that is a little bit too big to be unobtrusive, so I started thinking about something as loud as my 10 inch battery rig, but somewhere between a Phil Jones briefcase and a Roland Bass Micro Cube in size.

Obvious first question, why not just buy one of those? Answer, the PJB costs more than I wanted to spend for something that would be used occasionally, and I haven't been impressed with the Bass Micro Cubes I have tried, possibly because I play low-B 5 string basses.

My small battery rig is a class D stereo amp producing 12 W @ 8 ohms / 20 W @ 4 ohms per channel, and according to WinISD, will produce 106 dB from 100 Hz upwards at max volume using one 8 ohm Eminence Basslite S2010. The -3dB point is 66 Hz. That is what I want to achieve in a smaller box.

After a bit of searching, I found a 5 inch Faital Pro driver, available in a 4 ohm version, and two of those in a cab 350 * 350 * 170 mm will produce the same frequency response and the same volume. However, Hoffman's Iron law kicks in here, in that a speaker can go low, go loud, or be efficient, but only two of those at a time. These smaller speakers take 4 times the power of the single 10 inch speaker to reach the same volume. However, for a battery-powered pub session of 3-4 hours,  I can do that on one battery charge.

A few pictures of the build - 

1 The basic shape - speakers at the front, vent in between them, electronics compartment recessed into the back face

2  The cab is sized to include the battery, which sits on a sledge which comes in through an airtight hatch in the back wall.

3  I'm not sure if this is 'you can never have too many clamps' or 'you can never have too much bracing' Bill Fitzmaurice seems to have standardised on 15 mm ply and a shelf brace every 200 mm, and I'm using 9 mm ply, so I'm bracing every 150 mm.

4  Have I missed anything before I glue the second side on?

More to follow

David

Edited January 20, 2021 by Mottlefeeder
spelling

[Merton]

Interesting little project! Will enjoy watching this develop

[Mottlefeeder]

A little more progress -

Edges rounded off, front baffle painted with blackboard-paint, fitted with grill stand-offs, and off-cut of grill cut to size. I considered whether to stain and varnish the cab, but based on the tide marks I got staining a bass body, I decided to go with boring Tuff Cab black - now on order.

Electronics space painted, component placement worked out and sheet aluminium on order -

• Near edge: battery fuse, amplifier fuse and external power/charger pigtail - likely to last longer than a fixed fragile plastic socket.
• Centre: 40 W Class D amplifier. Allegedly a TPA3116D2 containing two paralleled pairs of bridged amplifiers, capable of driving 100 W into 2 ohms. I doubt that this design would, and I'm not going anywhere near that.
• Blank area: EQ - most of my basses have active EQ, and for those that don't, I have an EQ /DI pedal. The function I use most is a variable HPF, so the space will have a home-built FDeck HPF pre, and possibly a gain stage to drive the amp.
• Top edge: instrument in, volume, HPF frequency, on/off switch and charge/external power switch.

David

 

[Matt P]

very interesting build, what kind of heat will the amp be putting out? have you got a plan to keep it from getting too toasty? 

 

Matt

[Mottlefeeder]

5 hours ago, Matt P said:

very interesting build, what kind of heat will the amp be putting out? have you got a plan to keep it from getting too toasty? 

 

Matt

The manufacturer's data sheet for the chip states

"The high efficiency of the TPA3130D2 allows it to do 2 × 15 W without external heat sink on a single layer PCB. The TPA3118D2 can even run 2 × 30 W / 8 Ω without heat sink on a dual layer PCB."

Also, although the amp is rated as 40 W into 2 ohms with a supply of 13 volts, playing bass through it, I'm measuring a supply current peaking at 0.9 Amps - so probably averaging about 0.5 Amps. Assuming the amp is taking 13V * 0.5 A = 6.5W and it is 80-90% efficient, taking the worse case, 20% of 6.5W is slightly under 1.5 W.

The amp will be vertical on the back of the cab,  a 15 mm thick module mounted in the 35mm deep void between the cab back and the aluminium electronics cover, so I'm assuming that ventilation holes above and below the amp will provide enough airflow to keep it cool, Testing it flat on the bench, it might have been a few degrees above ambient, but not warm to the touch.

David

[Matt P]

@Mottlefeeder it sounds like you have it all under control! Looking forward to the finished amp.

 

Matt 

[Mottlefeeder]

3 hours ago, Matt P said:

@Mottlefeeder it sounds like you have it all under control! Looking forward to the finished amp.

 

Matt 

Sadly, being able to fix the problems I have found so far does not equate to having it all under control - but thanks for that vote of confidence.

David

[Phil Starr]

Well done, I was looking at one of those little amp modules myself to beef up my lockdown combo. I was looking to cheat even more and use one of the ready made pre-amps.

Following with interest.

[stevie]

Also following (although nothing to contribute at the moment). Don't you love a good build thread?😊

[Chienmortbb]

Just to add, those amps are incredibly efficient and shoiuld be fine as used here.

[Mottlefeeder]

Some progress, but mainly cosmetic

3 coats of Tuffcab matt black, and an aluminium back panel - to be sprayed black when completed.

David

[Stub Mandrel]

Looks good, unfortunately my TLRTs have run out for today 😞

[Mottlefeeder]

2 hours ago, Stub Mandrel said:

Looks good, unfortunately my TLRTs have run out for today 😞

That's a new one for me - what's a  TLRT?

David

[Stub Mandrel]

11 hours ago, Mottlefeeder said:

That's a new one for me - what's a  TLRT?

David

"Those Little Reaction Thingies" (c) ^TM @Teebs ^h^h^h^h^h @Skinnyman

Edited January 30, 2021 by Stub Mandrel

[Chienmortbb]

On 26/01/2021 at 18:49, Mottlefeeder said:

The manufacturer's data sheet for the chip states

"The high efficiency of the TPA3130D2 allows it to do 2 × 15 W without external heat sink on a single layer PCB. The TPA3118D2 can even run 2 × 30 W / 8 Ω without heat sink on a dual layer PCB."

Also, although the amp is rated as 40 W into 2 ohms with a supply of 13 volts, playing bass through it, I'm measuring a supply current peaking at 0.9 Amps - so probably averaging about 0.5 Amps. Assuming the amp is taking 13V * 0.5 A = 6.5W and it is 80-90% efficient, taking the worse case, 20% of 6.5W is slightly under 1.5 W.

The amp will be vertical on the back of the cab,  a 15 mm thick module mounted in the 35mm deep void between the cab back and the aluminium electronics cover, so I'm assuming that ventilation holes above and below the amp will provide enough airflow to keep it cool, Testing it flat on the bench, it might have been a few degrees above ambient, but not warm to the touch.

David

I am just wondering what motherboard and processor you will be using...

Seriousley though, the amp itself will be 90% plus efficient but the biggest issue with those amps is the inductors. They often use the wrong core material and many people replace them with improved inductors. You will know if you run it at even moderate levels, if the inductors get hot you may need to change them. For those that are wondering, the inductors are the black squares.

Edited January 30, 2021 by Chienmortbb

[Mottlefeeder]

1 hour ago, Chienmortbb said:

...the biggest issue with those amps is the inductors. They often use the wrong core material and many people replace them with improved inductors. You will know if you run it at even moderate levels, if the inductors get hot you may need to change them. For those that are wondering, the inductors are the black squares.

One of the first things I noticed about your class D amp was the size of the inductors, which led me to the conclusion that this pcb was not going to deliver the power that the chip could deliver. 

So far I have not felt the cores getting warm.

David 

[Chienmortbb]

2 hours ago, Mottlefeeder said:

One of the first things I noticed about your class D amp was the size of the inductors, which led me to the conclusion that this pcb was not going to deliver the power that the chip could deliver. 

So far I have not felt the cores getting warm.

David 

Sounds like you should be OK then.  

[Mottlefeeder]

1 hour ago, Chienmortbb said:

Sounds like you should be OK then.  

Hopefully - upgrading components on double sided PCBs was not in the original project spec. 

David

[Mottlefeeder]

The longer term plan was to use an FDeck clone HPF giving a fixed 12dB/Octave at 35 Hz and a further 12dB/Octave filter, variable between 35 Hz and 140 Hz. Some form of limiter/soft clipping was also envisaged, although I have yet to hear a simple compressor circuit that I could live with. I also had plans for some fancy power switching to allow one socket on the back to be used for power-in or battery-charge, or not in circuit when the internal battery was in use. 

I've decided to start with some simpler options, and swap out bits if I need to: first on the list was the HPF. 

The first Win ISD graph shows the response with no filter (red), 35Hz 12 dB/Octave filter (Blue) and 40Hz 12 dB/octave filter (Green). As you would expect, neither filter has much effect on the response of this speaker enclosure, with its relatively high tuning.

Having said that, the battery sits within the speaker enclosure, and I had concerns that the enclosure response would vary depending on whether you fit no / a small / a large battery. In reality, the difference between the larger battery and no battery increases the volume by 7% which reduces the tuning by 2Hz, which results in response changes of less than 1dB, so it is not an issue. 

The second WinISD graph shows the cone movement at maximum power, using the same colours, and this shows that an HPF is definitely having an effect. The horizontal red line is the limit for controlled cone movement, not the limit for coil damage, so I probably don't need an HPF for speaker protection, but using an HPF should give me better cone control and a tighter sound. An added advantage is that by suppressing the lowest frequencies, the battery will last longer.

I already had a preamp from a previous combo amp design, which included a variable HPF, so I used that as the basis for this interim preamp. It contains three op-amps, and I configured them as a Hi-Z input buffer, a variable HPF, and a variable gain stage to provide both volume control and compensation for differing outputs from basses. The last image shows the 'finished' (cobbled together) preamp and a basic schematic.

David  

Edited February 10, 2021 by Mottlefeeder
Clarity

[Chienmortbb]

Nice work.

[Phil Starr]

Nice work.

You probably already know this but others may not. If your speaker is following the red line for excursion then the coil has moved out of the magnetic field for part of the time, so motion is going to be non-linear for all the frequencies not just the bass. This means that there will be increased distortion across the board. Keeping the coil in the magnet gap also helps with heat dispersal. You get a lot of gains from that HPF

[Mottlefeeder]

1 hour ago, Phil Starr said:

Nice work.

You probably already know this but others may not. If your speaker is following the red line for excursion then the coil has moved out of the magnetic field for part of the time, so motion is going to be non-linear for all the frequencies not just the bass. This means that there will be increased distortion across the board. Keeping the coil in the magnet gap also helps with heat dispersal. You get a lot of gains from that HPF

It makes perfect sense, but I hadn't thought it through other than for the bass frequencies.

Thanks

David

[Mottlefeeder]

A bit more progress - the garage is about 3-4 degrees C at the moment, so metal bashing/drilling/filing occurs in short sessions.

David

[Mottlefeeder]

Three steps forward, one step back...

I was so engrossed in feeding battery and speaker cable through a small hole from the cab to the electronics that I forgot the speakers should be in parallel, and I wired them in series. Having fixed the speakers (8 screws) and the grill (6 screws), it all has to come  off again to be rewired. On the bright side, I was able to use a 'normal 'amplifier to drive 10Hz though the 8 ohm load to run in the speakers.

I've also fitted some sound absorbing polyester wadding in the front half of the speaker i.e. between the baffle and the brace, and also on the back wall. I can't see much reflection coming back from anywhere else.

David

Quote

 

 

[Stub Mandrel]

14 hours ago, Mottlefeeder said:

I've also fitted some sound absorbing polyester wadding in the front half of the speaker i.e. between the baffle and the brace, and also on the back wall. I can't see much reflection coming back from anywhere else.

I once made a little combo with four 4" speakers. Last year (or the year before...) I put a small port in it, which improved the bass response. On a hunch I removed most of the wadding inside and it improved the sound and volume - I think I had over damped it. It's worth experimenting, you can alwasy put it back in.