New Guy with an interesting design idea

[Bajaguy]

So I decided to purse the TDA2005 chip in bridged mode, which should give me about 12 watts into an 8ohm load. I found this schematic (which is right out of the datasheet) that looks like it will work. It even has a nifty veroboard layout which should help with the prototyping:


I'm looking to pair it up with the umble schematic from runoffgrove. It should give me a great little preamp with good adjustability and I hope play well with the input impedance of the TDA2005.



I'm not seeing any giant issues, but I'm not a designer by trade. I figured I'd post up the two and see if the designers of the group had any comments before I started running it all together into a single schematic.

Thanks for looking!

Baja

[mickmad]

Hey there, nice to see that you finally found something that could suit your needs
I'm kinda new to preamp designing, but I think that this setup should work; the things I'd do is to check that the preamp signal that goes into the final amp doesn't exceed the maximum input values of the TDA2005, and lay down the two sections on separate board so that you could plug them together later, just to rest assured that if they do not fit well together, like you are not happy with the way they sound together, you can easily separate them, and maybe reuse 'em; like, you could keep the preamp if you like it, and plug it on a more powerful final amp, for example.

[Bajaguy]

Yea, the umble circuit is really cool, but the big question is do I run a LM7809 to bring the 12V down to 9VDC, or refigure all of the resistances so I don't feed the J201's more than they can handle? From the protoyping end, I'd be better off using the regulator in case I don't end up using the two togther, but I can't see leaving it that way in the end.

Baja

[J M Fahey]

Run the Umble from 12V, it will work much better.
FWIWit would work even better with around +24V power, go figure.
JFETS don't like low voltages, such as 9V.

Only precaution would be to separate the Umble power from the TDA2005 (to avoid motorboating instability and such) with an rcrc filter, with, say, 2 220 ohms resistors and 2 470 to 1000uF x16V (or 25V) caps.
So you'll end with somewhat less than 12V, no problem.
But your proposed combination looks fine. :dbtu:

[Roly]

Now why do I look at that preamp with all those bloody trimpots and have a strange feeling of deja vu?   

[Bajaguy]

Now why do I look at that preamp with all those bloody trimpots and have a strange feeling of deja vu?   

Let me guess, something you've worked on?

[Roly]

Nah, something we've previously discussed here as a "design fail".

When I see every Drain load made adjustable between zero and 100k, accounting for the fact that this changes both the DC and AC conditions at the same time, then I suspect that the "designer" didn't actually know what he was doing.  These will not only adjust the Drain voltages to half the supply, they will also change the gain of each stage.

Just now I have spotted that four inverting stages in a row run off a single power supply rail with no decoupling between stages having the same phase, a really basic error.

At the time I was moved to sketch out something;


Look Ma, no trimpots at all!

Running the Umble from more than 9V should have no issues as JM says, but the power supply decoupling certainly needs to be fixed up.

[balaboo]

Congratulations, GRoper !!
This is exactly how my own preamps are designed - and they run on +24V.

[Bajaguy]

That's cool! I just have one question, do the fets have to be the MPF102 or can they be the J201?

Baja

[J M Fahey]

You'll actually have to try them, there's a wide variation in FET parameters, even is same brand and model ones.
That's why poor designs need trimpots everywhere, by the way

[Bajaguy]

Well I got the schematic all rolled into Eagle this morning, and I thought I'd run it by everyone before I went any further. I stuck in the J201's hoping they will work in the circuit, but I'm going to run this on a protoboard before I PCB it anyways.



Anyone see anything I missed?

Baja

[phatt]

I'll have a guess,  Maybe try Power feed at other end. 
Phil.

[Roly]

Congratulations, GRoper !!
This is exactly how my own preamps are designed - and they run on +24V.

Original, but quite conventional.  The idea was to create a gain module that has very high input impedance and low output impedance, and was hopefully tolerant of devices and spreads.

{I'm just trying to remember why the 10M's from Gate to +ve rail ... I'm sure I had a reason at the time but it just plumb evades me ATM}.


FET's generally have a fair bit of spread, but I've been using quite a few MPF102's lately and it hasn't been a problem (yet).  I imagine J201's should work okay, but if you do need to tweek anything it should be the value of the Source resistor, not the Drain as in the Umble.  As long as the Drain voltages wind up somewhere arounf half the supply it should be okay.  Need more gain?  Just replicate the stage again after the Master volume.

Maybe try Power feed at other end.

Phil beat me to it.

[Bajaguy]

I'll have a guess,  Maybe try Power feed at other end. 
Phil.

Ok, I'm lost with this one....

On a good note, I found a source for the MPF102's :dbtu:

I'm going to finish up the schematic for the amplifier portion today and get my parts order squared up. I'd love to get this thing runing next week, the weather is starting to get nice out here and we've been having a ball playing outside lately.

@ Roly - Thanks so much for the help on the preamp! I've spent a few hours with the old elctronics book from college and the calculator out trying to figure it all out. I did have one question, is the third leg of the preamp gain pot ok just hanging in the breeze, or should it be attached to something?

Baja

[Roly]

Power feed: +12V, R1, C2, Collector of Q4, then otherwise how it's drawn.  That way the power to the most sensitive stage (Q1 & Q2) gets the most power filtering.

What would you like to connect the free end to?  You could try your wife/girlfriend but she might have a bit too much hum.  If you really want to prevent electrons jumping off the end like Lemmings you could connect the free end to the pot wiper.  'course if you do that then some of them will be able to go around in a circle and might get dizzy and confused, so I guess it's a choice between some electrons freelancing or going around in circles - your call really.   8|