TAS5611

Folsom · « **Reply #20 on:** 26 Jan 2014, 07:36 pm »

Am I suppose to be that excited over PCB?

happyrabbit · « **Reply #21 on:** 29 Jan 2014, 12:30 am »

maybe...just a bit

happyrabbit · « **Reply #22 on:** 3 Feb 2014, 09:50 pm »

PCB Layout to support Mundorf M-Lytic 25mm x 30mm + Terminal Blocks for the Speakers and Inputs.

Folsom · « **Reply #23 on:** 4 Feb 2014, 01:05 am »

You are addicted!

happyrabbit · « **Reply #24 on:** 8 Feb 2014, 04:57 am »

happyrabbit · « **Reply #25 on:** 16 Feb 2014, 03:00 am »

Salis !

Digital Love Ver 3.... Can you say Linear Regulator ?

http://cds.linear.com/docs/en/datasheet/3083fa.pdf

I feel I am getting close...

Dwight

wushuliu · « **Reply #26 on:** 16 Feb 2014, 03:13 am »

Definitely in awe of your prowess with these boards. Still, sound quality is important and I'd be curious to know how the TAS compares to the TPA or even the IR2092 for that matter.

Folsom · « **Reply #27 on:** 16 Feb 2014, 05:05 am »

That does look like a nice regulator. I don't think the average bleed regulator is worth snot.

I think it makes absolute sense to use one for the chip, but I'd want to test it on and off the gates. I know it's a pain, but it might be more lively without it on the gates.

I'm getting near prototype PSU...

happyrabbit · « **Reply #28 on:** 18 Feb 2014, 03:17 am »

Wushuliu,

Thanks ! This is my 3rd attempt at a digital amp. If I am not impressed, I will not continue. To be honest, I am looking forward to owning a 1st WATT M2

the TAS and TPA are similar. The TPA design is directed towards BOM costs (aka simple layout and inexpensive parts). The TAS chip requires a more extensive layout and separates the gate, chip and bridge supplies from each other. The TAS has lower resistance mosfets, a bigger chip for lower thermal resistance and and a higher operating voltage. If you like the TPA3116, you will enjoy the TAS5611 even more. YMMV.

SALIS,

Version 2 has a pi filter for each of the half bridges. I am looking forward to hearing the linear regulator design against the pi filter. I can add a 2220 SMD FOOTPRINT to bypass the LDO reg. looking forward to your latest creation !

Dwight

Folsom · « **Reply #29 on:** 18 Feb 2014, 03:37 am »

Good, but I got something else cooking that might make you like class D if the first does not.

happyrabbit · « **Reply #30 on:** 4 Mar 2014, 03:09 am »

Latest Version.... Increased Caps on the LDO REG Input Side and MUNDORF 4-pole 35mm on the output. I increased the LDO REG input caps to limit any possible oscillation issues.
I will be using the Mundorf's as a 3-POLE. the GND's will be tied together to form the 3rd pole. I also switched from a 2220 Ferrite Beads to a Wurth XXL 10uh Inductor on the LDO REG INPUT LC FILTER.

140mm x 155mm

Dwight

Folsom · « **Reply #31 on:** 4 Mar 2014, 04:08 am »

Excellent choice abandoning ferrites. I've never been able to add them in a single place and not notice a decline in overall quality. I mean powercords even. I can't figure out how they could ever be good, for whatever reason...

Mundorf 4 poles? Good grief, no expenses spared here... Where have you looked into using them in this manner? I have yet to play with them, particularly because I don't do tubes as of yet.

I'm still wondering if what you're looking for won't be found with battery. What in the sound are you still trying to achieve?

happyrabbit · « **Reply #32 on:** 6 Mar 2014, 02:08 am »

Hi Sallis !!

Since I had to expand the PCB Board to support the 35mm Mundorf 4-poles, I have plenty of room for a REAL inductor. Ferrite Beads effectiveness drops off quickly with current. I was asking too much from my 2220 FB.

The 4 Pole will lower the HF noise from the LT3083 -> TAS5611 and TAS5611 -> LT3083. I really don't want the switching noise making its way back to the Linear Regulator. ripple rejection = transient response. The TAS5611 switches @ 400KHz. The LT3083 ripple rejection drops off @ 100KHz. I think I am safe. I believe Ver 3 will properly bypass the chips supply with minimal ripple present. I want the LT3083 to track the audio signal and ignore the other stuff... I feel isolating the half bridges from each other is a good thing.... just not very efficient nor practical. The linear regulators will limit the TAS5611 output capability. I only need a couple watts so it is acceptable.

http://www.partsconnexion.com/capacitor_ele_mundorf_ag+.html

I will use the 2200uF / 100VDC. They are the cheapest and plenty uf for my application.

I have decided not to explore batteries until I finalize the design.

I have settled on this power supply for my designs :

http://media.digikey.com/pdf/Data%20Sheets/Elpac%20PDFs/MWA120_Series.pdf

Dwight

Folsom · « **Reply #33 on:** 6 Mar 2014, 02:33 am »

I guess we'll see how you think it sounds. I'm starting to think regulators have a flavor.

Personally, without a regulator, the means of dealing with switching noise may be much easier. But with a SMPS power supply you are forced to try and mechanically couple it more, with inductors. Those 4 pole capacitors will help, but frankly DC capacitance has serious limitations; hence the capacitors on the output of the amp, AC.

The good news is I'd be a bit less worried about ripple rejection than you probably are. Ripple seems often to be an indicator of issues, not the issue itself, a lot of the time.

If you do put that SMPS as your PSU, you should give putting a CMC inline with the power cord, no capacitors following, a shot. It's not ideal, but it might help your attempt at ripple control. (I'm not 100% certain, the why's and how's of that condition are beyond my intellect at this hour)

happyrabbit · « **Reply #34 on:** 2 Apr 2014, 01:51 am »

Round 3 ! I just sent for this PCB board... My final attempt @ a Class D amp.

Mundorf 2-pole, Wima PPS, Rubycon ST , Panasonic OS-CON, for Power Caps
Mundorf EVO OIL for input Caps
Bourns 2200 Inductors
LT3083 for the Linear Reg
TAS5611 for the Class D Driver

happyrabbit · « **Reply #35 on:** 23 Apr 2014, 08:25 pm »

I should have this board built up within the next couple weeks.

wushuliu & salis audio I will have 2 PCB's left over....

Dwight

Folsom · « **Reply #36 on:** 23 Apr 2014, 09:46 pm »

Approaching satiation?

grubyhalo · « **Reply #37 on:** 25 Apr 2014, 12:26 pm »

Hello

Apologies if I'm missing this piece, but is there any plans to offer this on sale?

happyrabbit · « **Reply #38 on:** 26 Apr 2014, 11:27 am »

Halo,

Hopefully.... I am curious about the sound. I should know fairly quickly once operational.

Dwight.

grubyhalo · « **Reply #39 on:** 26 Apr 2014, 12:42 pm »

Thank you, makes sense...