dublin78 - Yes sadly the JAB boards look great on paper but are lacklustre in reality. I have wasted a lot of money on them.
SigmaStudio is fantastic. You don't have to look at a single line of code to start programming advanced DSP. To give you an idea of what is involved look at dzwer's link and also:
https://www.analog.com/media/cn/technical-documentation/evaluation-documentation/EVAL-ADAU1701MINIZ.pdf
SigmaStudio is fantastic. You don't have to look at a single line of code to start programming advanced DSP. To give you an idea of what is involved look at dzwer's link and also:
https://www.analog.com/media/cn/technical-documentation/evaluation-documentation/EVAL-ADAU1701MINIZ.pdf
Have any of you had any experience with toroidal inductors as the output filters for class D amps? I'm considering them to keep the board width low while supporting 4 channels.
Alternatively dual package inductors designed for class D amps are available but are (relatively) expensive, i.e: Coilcraft HA4158-EL, JA4575-BL Class-D Dual Inductor
Alternatively dual package inductors designed for class D amps are available but are (relatively) expensive, i.e: Coilcraft HA4158-EL, JA4575-BL Class-D Dual Inductor
I have had problems with the amplifier (TAS5754M) shutting down on maximum volume using those coilcraft inductors. Cant say for sure if its the inductors fault, because on paper they should be able to handle it, but after changing the output filter the problem was gone.
I now use Abracon ASPI-0630HI-100M-T15 and they work great, dont cost much and are pretty small. Only 6.65mm width and 0,48€ in quantities of 100.
I now use Abracon ASPI-0630HI-100M-T15 and they work great, dont cost much and are pretty small. Only 6.65mm width and 0,48€ in quantities of 100.
Very odd... I didn't think an inductor that is within spec could cause an amp to go into protection. Worst i've had is my inductors sitting at a couple of hundred degrees at 'idle'...!
Thanks for the suggestion, I will check them out.
The battery management boards are now with the fab house but will be a few weeks due to Chinese new year.
Update! The battery management and charge controller has been assembled and is working really well, i'm very pleased with the performance. The board happily handles 15A continuous discharge, balancing works well and the charge controller is perfect. In the picture below I am testing the simultaneous charge and discharge performance using a programmable DC load I designed.
The amp system is going through final checks on the PCB design and then designs will be sent off to the fab house, hopefully within a week.
An externally hosted image should be here but it was not working when we last tested it.
The amp system is going through final checks on the PCB design and then designs will be sent off to the fab house, hopefully within a week.
4 x IXFH60N50P3. They are bolted to two CPU coolers at the rear and will happily soak up 300W continuous, a lot more peak. I'll post some photos of the internals if you like.
835D is incredible for the price. The internal PSU is great when I just need some quick dirty power for testing and is surprisingly powerful. It is not accurate though, I use another supply for anything more precise. The iron and air gun feel cheap but work fine, just get some new tips for the iron.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Latest version is working really well, just a couple of things to change then its ready. Its currently sounding fantastic running a prototype of my latest portable speaker (in 2.1 mode).
Problems/changes:
No turn on delay implemented for the CSRA64215 bluetooth, sometimes causing intermittent startup of the internal voltage regulator.
I2S level shifter is stopping the bluetooth in-circuit programmer from working (de-soldering the chip each time to program currently!), add a way to put it into standby while programming
Adjust the silkscreen.
Do you have any good reading on core material vs performance in class D applications? I am mainly going on datasheet specification and things are working well so far.
In the photo are Sunlord inductors I am trialling. Its difficult to find inductors small enough for all 4 channels on the board size I would like so i'm trying a few out. I haven't taken any measurements but this board sounds better than the last model, which used larger inductors with a lower dcr - but the EMI snubbers have been removed so this could also be playing a part.
The next inductors to try are these (recommended by lutkeveld):
https://www.bourns.com/docs/Product-Datasheets/SRP6060FA.pdf
Edit: I was getting mixed up in the above message - l think lutkeveld recommended Abracron ASPI inductors and not the Bourns ones above.
All of these mentioned are the pressed powdered metal type. I understand these are 'soft-saturating' but as long as I keep the design well below Isat then presumably this does not matter? Or at least in the region on the current/inductance graph that is relatively linear?
All of these mentioned are the pressed powdered metal type. I understand these are 'soft-saturating' but as long as I keep the design well below Isat then presumably this does not matter? Or at least in the region on the current/inductance graph that is relatively linear?
I use the ASPI inductors in an my design, where my -3dB point is around 90Khz with Fsw=768Khz. This way any fluctuation in inductance does not affect the audio response that much (<0.5dB) while still providing 35dB suppression at Fsw.
A lot of linear inductors are bigger and more expensive. If it has significant negative impact on sound equality I will consider switching them out. Would also like to see any papers on this issue.
A lot of linear inductors are bigger and more expensive. If it has significant negative impact on sound equality I will consider switching them out. Would also like to see any papers on this issue.
Soft saturation cores tend to add hormonics rights from the begining, even at low currents. From my personal measurements, it is mainly k2 - which may "richen" the sound, as it is the same harmonics triodes add to the sound.
The best inductors regarding THD i had used in small size so far are "Codaca CSD0910B":
CSD0910B
These are ferrite-based.
Unfortunately sourcing these parts is difficult, as Codaca don't have distribution channels outside China, you'll have to oder from Codaca directly with a minimum of 50+pcs. They also sample smaller batches but shipping is expensive.
Beside Codaca, Coilcraft offers a variety of dual inductors, all ferrite-based.
Like:
UA8013-AL, UA8014 Class-D Dual Inductor | Coilcraft
Coilcraft HA4158-EL, JA4575-BL Class-D Dual Inductor
Coilcraft GA3416-CL Class-D Dual Inductor
Maybe these fit your board, HA4158-EL are the smallest footprint.
I use carbonyl dual inductors with TPA3128 knowing they will colour the sound with K2 so it is a question of personal liking.
I use Vishay, but Bourns offers exactly the same:
https://www.bourns.com/docs/product-datasheets/srf1010da.pdf?sfvrsn=d6027df6_2
https://www.vishay.com/docs/34381/ld40kb5a.pdf
The core material for the SRF1010 listed on the screenshot above is wrong, they're both same iron powder cores.
Carbonyl: add K2 harmonics, colours the sound like triodes
Ferrite: being higly linear until saturation current is reached. Keep temperature derating in mind, as saturation point shows to lower current when hot.
The best inductors regarding THD i had used in small size so far are "Codaca CSD0910B":
CSD0910B
These are ferrite-based.
Unfortunately sourcing these parts is difficult, as Codaca don't have distribution channels outside China, you'll have to oder from Codaca directly with a minimum of 50+pcs. They also sample smaller batches but shipping is expensive.
Beside Codaca, Coilcraft offers a variety of dual inductors, all ferrite-based.
Like:
UA8013-AL, UA8014 Class-D Dual Inductor | Coilcraft
Coilcraft HA4158-EL, JA4575-BL Class-D Dual Inductor
Coilcraft GA3416-CL Class-D Dual Inductor
Maybe these fit your board, HA4158-EL are the smallest footprint.
I use carbonyl dual inductors with TPA3128 knowing they will colour the sound with K2 so it is a question of personal liking.
I use Vishay, but Bourns offers exactly the same:
https://www.bourns.com/docs/product-datasheets/srf1010da.pdf?sfvrsn=d6027df6_2
https://www.vishay.com/docs/34381/ld40kb5a.pdf
The core material for the SRF1010 listed on the screenshot above is wrong, they're both same iron powder cores.
Carbonyl: add K2 harmonics, colours the sound like triodes
Ferrite: being higly linear until saturation current is reached. Keep temperature derating in mind, as saturation point shows to lower current when hot.
Attachments
Attachments
Last edited:
Hi, are you from German, speaking German ?
If you like please send me your Tel / Handy Number per PM
have some technical questions
Sorry for the radio silence here, I've been very busy developing a set of speakers on commission (which will have this amp system installed inside).
The plan is to get a short run of boards assembled (100-500 depending on economies of scale) when I have the money to do so. I need to be sure that the design is completely watertight as well so by implementing hand built versions into current designs I'm getting the system properly field tested.
I am trying to keep the costs as low as possible but the issue is the expensive ICs/modules, like the bluetooth and DSP. I am hoping to be able to sell a board for around £50-£60 GBP / $60-$70 USD. Because this board is an all in one solution that price is good considering the modules you would have to use to replace it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.