Welcome to Solid State Guitar Amp Forum | DIY Guitar Amplifiers. Please login or sign up.

March 28, 2024, 08:48:07 AM

Login with username, password and session length

Recent Posts

 

single power rail 40 watt amp using the NTE equivelant of the TDA2030?

Started by darwindeathcat, February 06, 2008, 11:53:36 AM

Previous topic - Next topic

darwindeathcat

Hi All,

    I've been looking for an amp build that was: 1)easy, 2)in the 30-40 watt range, and 3)used a single rail power supply so that it was 4)cheap. After untold hours of searching, I seem to have found the solution in two schematics that use the TDA2030 in conjunction with two power transistors. The first schematic comes from a kit offered by quasar electronics (UK) (http://www.electronic-kits-and-projects.com/1109.htm) (schematic is a pdf listed at the bottom of the description, and the other is listed on elecfree.com (http://www.elecfree.com/circuit/power-amplifier/30w-amp-otl-by-tda2030-transistor/
Power Amplifier : Power Amplifier OTL 30W by IC TDA2030 +Transistor).
   My questions for the list are as follows:
   1) I would prefer to build from the schematic from the quasar kit because I know it is proven, and it puts out more power. However, they (of course) do not list any component values other than the IC and Transistor types, and the input voltages. I'm more of a "follow instructions guy" when it comes to amp building, so I have no idea how to calculate these values. Can I use similar values from the other schematic (the elecfree one)? That schematic is very similar except that it uses different transistors and has a lower input voltage (and is missing the equivalent of C2, R1, and R3 in the quasar schematic).
    2) As the TDA2030 seems to be hard to find here in the states, I have looked up the NTE replacement for it, which is the NTE1380. Looking over the datasheet, it *seems* to be an exact duplicate. Is anyone aware of any deviations from the TDA2030 pinout?
   3) As far as the transistor types: I was unable to find the exact types from the schematics. Can I uses some sort of generic equivalents? Any suggestions on which ones?
   4) Any other suggestions about these schematics?

Any help would be greatly appreciated!!!
__ ------------------- __
__ | | | | | | | | | | | __
    --------------------

teemuk

Single supply doesn't equal cheapness – pretty much contrary: To filter an unipolar supply equally as good as a dual supply you need about double the amount of capacitance. Since voltage is higher the capacitors will cost more and they will be a lot bigger and harder to find too. Unipolar also means that you need an output coupling cap, which also costs money.

1) Schematics are, as far as I see, exactly identical if component values are excluded. Values of resistors R6 and R7 depend on the used transistors and driver chip so if you go on changing things you likely need to alter them.
2) No comment.
3) That circuit is very simple so generic equivalents work – assuming you adjust the current sampling resistors R6 and R7 accordingly. The circuit is not too picky about most component values. You just need to know how they interact and work in this circuit, though.
4) Quite frankly, this design doesn't look too good. There is no thermal compensation for the output devices nor a short circuit protection. Aside the extremely high Zobel cap value (the cap in the output in series with 1-ohm resistor) I see no compensation against instability. Relying only on Zobel to prevent oscillation is poor engineering practice. I would convert to dual supply to get rid of the output capacitor and to have a cheaper power supply circuit and likely less ripple in the output. The chips power pins should be bypassed.

I would pick a different design if I were you. As a matter of fact, for this low output power I would just build a chip amp instead of even trying something discrete or "semi discrete". It would be much simpler, so much more reliable and perform a lot better. This would fill your clauses 1 and 2 and since clause 3 (along with four) was skewed anyway something like this would be quite perfect for you, I think. For example, take a look at this: http://home.eunet.cz/rysanek/pdf/tda2050.pdf

iTzALLgoOD

Quote2) No comment.

Is there an issue with NTE that we should be aware of.  After reading this post I checked at Mouser and it says "99 week factory lead time" for the TDA2050.  They do however have the NTE replacement for 3x the price. 

Other than cost should we be worried?  2 years seems like a long time to wait for components.

darwindeathcat

Thanks for your reply! I got some good advice about this from another forum also, and I think I have been thoroughly convinced that a chipamp is the way to go. My choices are the lm3875, lm3886, and the TDA2050 (as you suggest). I read at the "decibledungeon" website about using a couple of old computer power supplies (to get +-24v) with a lm3875 gainclone, and that's one idea I might take. Otherwise, I guess I can head to the thrift store and buy some cheap old HiFi equipment, and hope I get a good transformer and/or heatsink out of it... Hopefully, with all the stuff in my parts bin, the chip and other components that I have to order, and the "salvaged" transformer, I can get one of these chipamps together for less than $20. I've got several preamps builds lying around, but I also thought of building in one of those runoffgroove pedal designs, like the english channel or professor tweed, into this thing as well... That will add a bit to the cost, but certainly not more than $10 more... A 40-60 watt Harp amp, for under $40 total? If it sounds good it will be an amazing deal! I'll keep you advised of my progress!
__ ------------------- __
__ | | | | | | | | | | | __
    --------------------