More Mods & Tests on the Dual TDA7293

[@Douglas_Blake_579]

Okay, I'm impressed. This one is on my list for sure. One point about circuit architecture ... I've always found that multiple stages with lower gain does a better job than one stage with all the gain, despite the increase in parts counts (within reason, of course).

[@MichaelBeeny]

Well Douglas, you might be able to suggest a few more possible mods!!

[@ronschauer839]

Yes indeed. That is one of the reasons to reduce the gain of the 7293 boards a bit and to add a preamp/buffer in front. In addition, the low end response is significantly improved by changing the indicated resistor. If you look at a normalized overlay of the four resistance value results the differences are quite dramatic.

[@Douglas_Blake_579]

@@MichaelBeeny I may, once I get into it. Hopefully in a couple of weeks. Right now I'm still finishing up my deep dive into the TPA3255 boards ... That chip has so much potential but it seems that every product using it has at least one "day 1" issue that rules it out. All I need is one good board that either is or can be made serviceable and I'm off to the races. Then we play with the TDA7293 ...

[@Douglas_Blake_579]

@@ronschauer839 I saw the graphs online ... it is a decent improvement for sure. Have you considered increasing the input capacitors? More than once I've seen low end roll off above about 20hz just get up and vanish when you bump up the coupling capacitors. For example, on the Aiyima A07 going from 10 uf to 22uf on the inputs brought the low end cut off down to about 4hz.

[@ronschauer839]

@@Douglas_Blake_579 Thanks, I found the graphs to be quite informative. The signal input paths on these particular boards are DC-coupled (there are no coupling capacitors). At very low frequencies C1 has no effect at all, even if left at its original 2200pF. However, if you are referring to what I labeled C2,3,4, then yes I do think it would help in this regard to increase them (with a caveat or two). The RC "frequency response" of the combination of those capacitors and R4 (assuming that R4 is changed to 470 ohms as Mike did) is around 2.5hz. This reveals itself as a drooping of the waveform in lower frequency square waves, which is exactly what he showed. At even lower frequencies these capacitors become more or less open circuits just like C1 on the input, further reducing gain and increasing the drooping effect. With a sine wave however you'd mostly see the effect as a reduced output amplitude. But it would be with some accompanying distortion (eventually becoming more triangular or "sawtoothed" at ultra low sub-sonic frequencies). Unfortunately, if you doubled or tripled the value of the C2,3,4 combination you *could* adversely affect the operation of the DC servo circuit (auto-null). So it is probably best to leave them alone. 🙂 On the other hand, increasing the value of R4 lowers the "frequency" of the C2,3,4/R4 combination, pushing the point of that transition to triangular even lower. That's why he said that changing the R4 value higher will also have a good side benefit of reducing LF distortion despite reducing the overall gain. It *may* also help to further increase the value of C12 (bootstrap capacitor). But the effect of doing that would likely be seen only as a further reduction in LF distortion, which is not a bad thing..!

[@Atelierul29]

Good to see that you're still active!

[@MichaelBeeny]

I'm not too sure about ACTIVE lol

[@KP-nd3bx]

Glad to see you are in good health 😉 think you got me hooked for this amplifier. Looking forward to seeing the next edition. Thank you for sharing 🇩🇰

[@MichaelBeeny]

Not a bad subject to get hooked on, I'm sure.

[@carlosah2598]

Thank you Michael as always perfect analisys

[@MichaelBeeny]

Thankyou Carlos, most appreciated.

[@billkalina3089]

Bravo Mike, and thanks again for your research. We are the benefactors. Frankly I am amazed there isn't greater acclamation for this amplifier. It truly kicks ass! I can't hold back anymore. I have cobbled the pennies together, and I have ordered them!

[@MichaelBeeny]

I cannot hardly wait to hear what you think of them. I'm sure I know what you will say. I just wish I was on commission lol. This board was loosely based on the LINN LK140 apparently.

[@billkalina3089]

@@MichaelBeeny yes, I will let you know Mike. You know what would be great: to deliver the amp to say Steve Guttenberg, the Audiophiliac. I wouldn't tell him anything about what's inside, and invite him to put it through its paces, and to see if he could guess what kind of an amp it is, and how much it cost to build. I'm certain he would be quite surprised!

[@ronschauer839]

@@billkalina3089 If you gold plated the cabinet and used anti-tamper screws to prevent peeking inside, that might indeed be an interesting experiment. 🙂🤣

[@MichaelBeeny]

I don't think that would work Ron. If you could prove these gold-plated amplifiers did not sound any better, lots of high-end manufactures would go out of business.@@ronschauer839

[@justinvain377]

I can't wait to see your next vids, Michael. Thank you and wish you all the best.

[@MichaelBeeny]

Thankyou Justin. I can hardly wait myself lol

[@el_arte]

Amazing square waves for sure!

[@MichaelBeeny]

Yes, even I was amazed.

[@Bengal20]

Thank you very much, Mr Beeny, for this great video! It would be very interesting if you'll make video for buffer 😁👍

[@MichaelBeeny]

I'm working on it right now. Just call me Mike, we are all likeminded friends here.

[@pistollero]

Greetings from Portugal 😊. I bought those amps. I hope to test them soon when i have the transformers.

[@ronschauer839]

True, the improvements are not "chalk and cheese" (just as long as it is driven by a low impedance source). However, if it can be made just that much better with such simple and inexpensive modifications, why not do them? 👍👍 Thanks!!!

[@MichaelBeeny]

That low impedance buffer circuit will probably be my next video.

[@Bengal20]

@@MichaelBeeny can't wait to see this video with buffer 😁👍👍👍

[@boxinggold]

Thank you for sharing. Looks like my next project will be tweaking it 😂

[@hackedoff736]

Another excellent video Michael. I’m inspired to build one with your mods. Do you know what the gain is on the finished amplifier?

[@zoelong6021]

So good to see videos again....and you even wet our appetite for more on the way...just like the good old days BTW - on your advice i got this board but before I could get round to a listening test seem to have blown a chip...can you advise if the chips must be isolated from the heatsink?

[@MichaelBeeny]

So sorry to hear that. Yes, the ICs MUST be isolated from the heatsink. Did you not get mica or plastic washers with the amplifiers? You should have. The ICs not only need isolation from the heatsink but also from each other. In fact 99% of all amplifiers need to be isolated.

[@ronschauer839]

Yes, they *must* be electrically isolated from the heatsinks if the heatsinks are grounded or connected to anything else (including another channel).

[@Douglas_Blake_579]

@@ronschauer839 IIRC ... the substrate plate on the back is connected to V- supply, is it not? That would mandate ... heat sink compound, mica insulator, more compound and a plastic washer on the mounting bolt.

[@ronschauer839]

@@Douglas_Blake_579 Per the specification sheet for the TDA7293, the tab is connected to pin 8 which is the inverting signal input pin, however its name on the data sheet is a bit confusing. The V- (power supply input) is actually on pin 15. Quite oddly however, they labeled BOTH pin 8 and pin 15 as "-Vs", followed by "(POWER)" or "(SIGNAL)" which is all that differentiates them..! Why they did not call pin 8 "-Vsig" and pin 9 "+Vsig" to reduce or eliminate any potential for confusion I don't know. Regardless, the device tabs *must* be isolated from the heatsink for them to function properly.

[@brucewarren692]

Hi Michael thanks for the great content. I see boards on ebay with 3 tda 7293 chips on them. Do you think this would be an advantage or is it just for power? I want to build a stereo amp for music and to integrate with my home theater and you videos have me thinking tda7293 is the way to go.

[@MichaelBeeny]

I have not tested the 3-chip version. It does not contain the servo circuit, however. It would allow higher power, if you need more, simply because you can up the voltage and the increased current is shared between the 3 chips.

[@billkalina3089]

Lol ...I don't know about the gold plated case, but maybe glue a high end logo on there? I'm certain many "golden ear" audiophiles would be shocked and maybe embarrassed in a blind test!

[@ronschauer839]

Gold plating (if thin enough) can be relatively inexpensive, and it could (in some potential reviewers' eyes anyway) greatly enhance the sonic clarity, soundstage, depth, ambience, detail, impact, presence, *blah, blah, blah*- ad nauseum. As for a logo pasted on, I think that is an excellent idea..! Pick anything north of $10,000 (per channel *without* a preamp of course...). Do you suppose a golden-eared audiophile would give it an true blind A-B test *and* admit the results? 😆

[@billkalina3089]

@@ronschauer839 lol ....I don't think you would get a bunch of those in the same room anyway. Besides, they hate to be proven wrong: that their 2,000 dollar, kryptonite infused, cryogenic, spun pure oxygen free rhodium is better than the pair you picked up at a thrift store.

[@ronschauer839]

@@billkalina3089 Oh, but you forgot the separately pressurized dry nitrogen isolation jackets for the interconnect cables, the pristine glacier ice cable standoffs (with individual refrigeration units naturally) and $1000 a meter power cables with rhodium-plated wall socket contacts... and several dozen magic quantum wall dots thrown in for good measure. 🤣😂 LOL..! But seriously, these little boards can be really good, expecially for the price. And there are *triple*-chip versions that should be even better at driving low impedance loads (but they lack the DC offset servo circuit).

[@billkalina3089]

Doh! I almost forgot!

[@FrancescoBalena]

Dear Mr. Beeny, I have a special request for you. I have seen your videos on the TDA7293 integrated amp, but I would be curious to know what you think about the LM3886. I used this integrated amp a while ago and it sounded much better than many high level amps. Can you help me? I repeat this message on some of your videos, hoping to be intercepted by you. Sorry for the inconvenience. FB

[@thegrowl8874]

your 9th viewer sir from Philippines

[@MichaelBeeny]

Thank you for viewing, but did you like the video??

[@Steven-pv1xr]

You actually dont want 5hz. Record warp will drive you crazy as hell if your amp gets down there

[@MichaelBeeny]

I personally don't use records at all. So only an issue for these that do. Such filtering should be provided on the RIAA circuit. Other inputs don't need restrictions.