Dayton DTA-2 Class "Caps Only" Bass Mod?

[BrendaEM]

Hi,

I have a Dayton DTA-2 module installed in a suitable box, and run it from a valve regulated lead acid battery. I also added a 3 amp diode to cause a .7 volt drop because a 12v battery can go over 13.5 volts. The reverse protection is nice too.

Except for bass, I am pretty impressed with this amp, but...

I think that in the version that I have, the bass still rolls off pretty high, so I would like to do the bass mod, but the board is laid out different, so I think the cap legends on the board, such as C3 and C4 will be different as well.

Has anyone done this mod on the current board?
http://www.michael.mardis.com/sonic/inputmods.htm

I might add a larger stiffening/reserve cap too, but I think I know which cap it is : )

(I realize that this will void my warranty.)

Thanks,
BrendaEM

[Wolf]

I tried version 3 with the original SI T-amp, and the 'bridge' is harder to do than you think without lifting the solder-pads. Make sure your iron is hot, and be as brief as possible. I just placed the caps in the wire from the harness.

Later,
Wolf

[BrendaEM]

Has anyone done the mod?

[BrendaEM]

Perhaps I will photograph both sides of the amp and post them.

[neildavis]

Perhaps I will photograph both sides of the amp and post them.

That would help, but if you spend about 15 minutes with an ohmmeter you could trace out the input circuit and know for certain which parts are the input caps.

The simplest mod is to bridge the existing caps with some 1uF stacked metallized poly caps. These caps are physically small enough to tack onto the board without the risk of breaking the board traces. Carefully shape the leads with needle-nosed pliers and use a good temperature-controlled soldering iron to prevent damage to the board. Once soldered, secure the parts with silicon glue or hot-melt glue to protect the solder joint from mechanical stress. 1uF will probably extend the bass response by enough--and any larger value will be a lot bigger and heavier (at least for poly caps). If you need the caps and don't want to deal with Digikey (part number P4675-ND), PM me and I will send you a pair. However, Digikey is a lot faster at filling orders and shipping than I am.

http://media.digikey.com/photos/Panasonic%20Photos/V%20SERIES%2011.0H,7.3L%20BULK_sml.jpg

[BrendaEM]

This is what I think is going on, but I am not sure.
Does that seem right?
http://www.brendamake.com/temp/t-amp-top.jpg

From Specs:

"RI Inverting Input Resistance to provide AC gain in conjunction with RF. This input is
biased at the BIASCAP voltage (approximately 2.4VDC).
RF Feedback resistor to set AC gain in conjunction with RI; AV 12(RF /RI ) = . Please refer
to the Amplifier Gain paragraph in the Application Information section.
CI AC input coupling capacitor which, in conjunction with RI, forms a highpass filter at
f 1 (2 RC )"



I believed the X's are slugs/no op parts, but I was wrong, as they are inductors, (thanks neildavis.)

I also see a few caps near the input jack, but they have no resister near them, so I suspect they are low pass to protect the amp from RF.

( Thanks for the info neildavis : )

[Beer man]

This pic is from the audio-magus.com tweak kit.

What I have done is left the resistors and just bypass/add a madisound 2.75richey surplus cap.
They have from 2-2.75uF and do use thin wire 30awg to connect the cap to the delicate chip cap or the traces.

Or goldmine-elec.com has mylar caps surface mount caps and stiffycaps.
I also put in a bigger 3300cap but it had to go in the backside!

[neildavis]

This is what I think is going on, but I am not sure.
Does that seem right?
http://www.brendamake.com/temp/t-amp-top.jpg

I believe the X's are slugs/no op parts.

I also see a few caps near the input jack, but they have no resister near them, so I suspect they are low pass to protect the amp from RF.

( Thanks for the info neildavis : )

The two parts you have labelled C1 are the input caps. They should be easy to bridge if you have the right tools.

It looks like the X's are the L1 and L2 inductors--they help with RF rejection.

Regarding your first post: If you are concerned about exceeding the 12V supply, you can actually run the chip from 14.6V if you use additional protection diodes on the output. The application circuit for the TA2021B shows the additional 4 diodes (the TA2021B is the "slug-up" version of the TA2024). Without the diodes, the supply should be less than 13.2V.

[BrendaEM]

neildavis, Thanks for the input.

Looking at the board I can start to see that L1 and L2 are marked, but whomever did the screening, needs to find something else to do for gainful employment.

So if I stacked some 1uF caps, would it be fairly flat down until 20hz, or should I go a little larger?
(I don't know what the original values are, either. My $20 DMM seems only to check caps by plugging them into a socket on it.)

(Thanks, beerman too : )

The abbreviations I used were from the pdf:
http://www.datasheetcatalog.org/datasheet/Tripath/mXyzxwwt.pdf

[neildavis]

So if I stacked some 1uF caps, would it be fairly flat down until 20hz, or should I go a little larger?
(I don't know what the original values are, either. My $20 DMM seems only to check caps by plugging them into a socket on it.)


The resistors match the values shown in the diagram on Michael's site (http://www.michael.mardis.com/sonic/images/input-new.gif). Since the resistor values are the same, I'm assuming the capacitor values for input capacitors C3/C4 are the same. The data sheet explains that there is a high-pass filter formed by the input cap in series with the input resistance. The input resistance includes the resistor you have labeled iiR plus the volume pot in parallel with the resistor labeled R01 on Michael's schematic. If the input cap is .33uF, the cutoff frequency (-3db point) is 24Hz (1/(2*Pi*Ri*Ci). Decreasing the volume will lower the cutoff frequency. That's not bad...probably not worth modding. The frequency response on this page uses an expanded scale so it looks somewhat bad, but it is really fine for most speakers: http://www.michael.mardis.com/sonic/measure/5066-freq.html

Changing the input caps from .33uF to 1uF lowers the input cutoff frequency to 8Hz. Paralleling the existing caps with 1uF caps (total capacitance = 1.33uF) will lower the cutoff frequency to 6Hz. Either solution should be adequate--no good reason to use a larger cap.

I recommended the 50V stacked metallized poly caps because they are high quality and aren't anywhere as large as the 200V or higher voltage caps that other people have used. Another good solution would be an electrolytic cap, which would be even smaller and easier to fit on the board. Electrolytics are actually fairly low distortion at frequencies where their impedance is low, so a 10uF 6.3V cap would work just fine in this circuit, and provide response down to 1Hz. However, electrolytics are polarized, so you need to observe the polarity--see Michael's site for guidance.

[onlylsleeping]

The simplest mod is to bridge the existing caps with some 1uF stacked metallized poly caps. These caps are physically small enough to tack onto the board without the risk of breaking the board traces.


Does this require removing the existing caps? How is the bridge made with the 1uF caps? Do you have pics or a guide I can follow?

[thekorvers]

Does this require removing the existing caps? How is the bridge made with the 1uF caps? Do you have pics or a guide I can follow?

No, the existing caps are left in place. The 1 uF caps are placed in parallel. This is done by carefully soldering the additional caps to the same spot where the originals are attached. You need a small temperature controlled soldering iron and good soldering skills. You need to be quick about it. The total capacitance will be 1.33 uF, lowering the cutoff frequency from 24 to 6 Hz as neildavis explained. The 1uF metalized poly cap suggested by Neil is a good choice as it is small, but any physically small cap will do. If you use an electrolytic cap, pay attention to the polarity when you solder them in.

[rogoll]

Here's another mod that I think is worth the effort ---> http://techtalk.parts-express.com/showthread.php?t=221983&highlight=heat+sink

[rogoll]

I just purchased a SMD "tweezer-style" soldering iron from MCM (use a catalog code for lower price than displayed on web site).

http://www.mcmelectronics.com/product/TENMA-21-8230-/21-8230

I think these are quite handy for working on items like the DTA-2 amp.