These kits are wonderful. I know they save you a ton of time and hair pulling from doing an amp from scratch. The time savings must 10x.
It is a cap coupled output though and single rail - which is not necessarily a bad thing as you have built in folk proof DC offset speaker protection. But detail and resolution would probably increase a bit with a DC coupled amp with a solid state relay DC protect.
I have built a TDA7293 chip amp (the Xmas amp by Jhofland and Dibya) and I really like it’s sound. I suspect it might have to do with the quasi complimentary DMOS output topology as it sounds and measures with a dominant second/third harmonic much like SE Class A amps but with low overall THD. If you are feeling more ambitious, an open source design is available here for anyone to order their own PCBs and make their own amp.
Very simple build, here is PCB design:
I gave away 50 of these PCBs last year around the holidays, hence they were called “Xmas amps”.
I still listen to mine, as it is still in the main system rack, albeit, no chassis. I use it when I want to listen to a “chip amp” reference. Good for up to 100w if you have the right lower transformer. I run mine at 50wpc max and it drives everything I need.
It is a cap coupled output though and single rail - which is not necessarily a bad thing as you have built in folk proof DC offset speaker protection. But detail and resolution would probably increase a bit with a DC coupled amp with a solid state relay DC protect.
I have built a TDA7293 chip amp (the Xmas amp by Jhofland and Dibya) and I really like it’s sound. I suspect it might have to do with the quasi complimentary DMOS output topology as it sounds and measures with a dominant second/third harmonic much like SE Class A amps but with low overall THD. If you are feeling more ambitious, an open source design is available here for anyone to order their own PCBs and make their own amp.
Very simple build, here is PCB design:
I gave away 50 of these PCBs last year around the holidays, hence they were called “Xmas amps”.
I still listen to mine, as it is still in the main system rack, albeit, no chassis. I use it when I want to listen to a “chip amp” reference. Good for up to 100w if you have the right lower transformer. I run mine at 50wpc max and it drives everything I need.
Last edited:
Again; the Update my Dynaco / Akitka Amp is good .. unexpectedly good.
Largish Capacitance reservoirs give good bass 🙂
And Greedy Pig increases of which give audible bass depth increase 😉 .
V affordable as happy bonus.
Largish Capacitance reservoirs give good bass 🙂
And Greedy Pig increases of which give audible bass depth increase 😉 .
V affordable as happy bonus.
Having done at least 3 chip amps before the GT-102, in my opinion, it is much closer to the sound of discrete amps I've built (M2 & ESP P3A) than any of the prior chip amps.
I haven't heard the TDA7293 amp shown above for comparison though and it looks more complex than the other chip amps I built so it may escape the 'chip amp sound' better too.
I haven't heard the TDA7293 amp shown above for comparison though and it looks more complex than the other chip amps I built so it may escape the 'chip amp sound' better too.
It is a very big "chip".
At idle we "bias" the two BIG power transistors to about 1% of their maximum current. Idle current is very twichy, also temperature sensitive. In discrete we pick a small diode or transistor and try to make it track the big parts. This has to be try-and-try-again. But when NatSemi (now TI) makes ALL the parts on the chip, they can know they will be similar, all the chips in a lot, and pre-set the appropriate trim. Each lot is spot-checked and if the process seems to be drifting out of design goal, they can put a spot of layout tape on the mask to change the value of a resistor and change the preset. (OK, no tape, it's all computers now.)
Ah, that makes sense. I have to assume that having the bias set at fabrication, has the result of having it be more stable over time?
I am curious about how the Z4 version is revised - with different parts, certainly - but specifically? Is there more robust parts capable of higher current, or is the circuit design tweaked?
I am curious about how the Z4 version is revised - with different parts, certainly - but specifically? Is there more robust parts capable of higher current, or is the circuit design tweaked?
I am not sure what "Greedy Pig" is in this context - is it a mod for the GT-102?
Similar question - what is the "right lower transformer" mean? Is this a way to give the power supply more overhead reserve?
I bought the AkitikA GT-102 Z4 kit and assembled it in about 22 hours of work. It tested and runs perfectly, the first time!
This is my favorite amp, that I have ever heard. It is way beyond what I imagined possible.
After building mine, I found a 9 part series on a YouTube channel called StereoPolice, and he is similarly impressed. He did a number of test bench measurements to confirm the specs. And right at the end of his review, he mentions a possible tweak that might address the least wonderful aspect - and that is the higher treble extension / detail reproduction. This is something that is better than the Schiit Vidar, but not quite as good as the B&K ST-140.
The thing that Stereo Police gentleman suggests is that if the power supply impedance is increasing as the higher frequencies, then possibly adding a lower value capacitor in parallel with the output cap on the power supply board (which I believe is the 10,000uF on the left, close to the RF shield?), that this might flatten out the treble a bit more, and improve this?
This is my favorite amp, that I have ever heard. It is way beyond what I imagined possible.
After building mine, I found a 9 part series on a YouTube channel called StereoPolice, and he is similarly impressed. He did a number of test bench measurements to confirm the specs. And right at the end of his review, he mentions a possible tweak that might address the least wonderful aspect - and that is the higher treble extension / detail reproduction. This is something that is better than the Schiit Vidar, but not quite as good as the B&K ST-140.
The thing that Stereo Police gentleman suggests is that if the power supply impedance is increasing as the higher frequencies, then possibly adding a lower value capacitor in parallel with the output cap on the power supply board (which I believe is the 10,000uF on the left, close to the RF shield?), that this might flatten out the treble a bit more, and improve this?
I think adding a 4.7uF MKP film cap in parallel with the big electrolytic output cap will improve the treble as well. This is an excellent amp from what I can see. Nice job!
Now sit back and enjoy.
Now sit back and enjoy.
Right - enjoying this amp is easy! I am thinking of asking Dan Joffe about adding that cap, but if it is that easy to open up the higher treble, and adding a touch more detail, then this would make it essentially perfect.
I have a newbie question - what is an "MKP" film cap? What are the aspects of a given cap that would make it better in this situation?
The cap that comes up on a Internet search, are WIMA MKP 4 Series. There is one cap of a similar type already on the power supply board - it is in the middle of the four big diodes.
Last edited:
MKP (polypropylene metal film) is one of the better dielectrics for film caps. Generally considered the best sounding of the common caps. I say common meaning circa $2 to $10 caps that you can get as a standard item.
There are boutique caps made from the skin of virgins plated with gold and dipped in whale fat with ambergris thrown in for good measure. They may or may not sound better. But they will cost a lot more.
There are boutique caps made from the skin of virgins plated with gold and dipped in whale fat with ambergris thrown in for good measure. They may or may not sound better. But they will cost a lot more.
I have been using Audyn Q4 MKP caps in my two crossovers, and they seem to be very good. Doing a bit more searching, I have other possibilities, too:
Audyn Q4 Series MKP 4.7Uf for $3.54 each: https://www.parts-express.com/Audyn-Cap-Q4-4.7uF-400V-MKP-Foil-Capacitor-027-114
Mundorf 4.7uF MCap Classic (MKP) for $6.55 each: https://www.partsconnexion.com/MUNDORF-87060.hmtl
ClarityCap 4.7uF CSA CopperConnect for $10.60 each: https://www.madisoundspeakerstore.c...ycap-4.7-mfd-csa-copperconnect-poly-cap-250v/
Audyn Q4 Series MKP 4.7Uf for $3.54 each: https://www.parts-express.com/Audyn-Cap-Q4-4.7uF-400V-MKP-Foil-Capacitor-027-114
Mundorf 4.7uF MCap Classic (MKP) for $6.55 each: https://www.partsconnexion.com/MUNDORF-87060.hmtl
ClarityCap 4.7uF CSA CopperConnect for $10.60 each: https://www.madisoundspeakerstore.c...ycap-4.7-mfd-csa-copperconnect-poly-cap-250v/
All of those are fine brands. I use Audyn Q4 also.
Metalized polycarbonate caps are also very good sounding.
Metalized polycarbonate caps are also very good sounding.
Based on a comment by the Stereo Police gentleman, and the discussion about this here, I have ordered a 4.7uF ClarityCap CSA CopperConnect cap (~$11) from Madisound, and I will try things with that installed in parallel to the 10,000uF output cap on the power supply board. My other option was an 4.7uF Audyn Q4 MKP ($4) from Parts Express.
The idea is that it is likely that the impedance of the power supply is going up as the frequency rises, and by adding a low ESR/ESL cap it will gain slightly better treble extension, and gain a bit more of the subtle details and quiet clarity that the B&K ST-140 had in abundance. If this is the case, then I think the AkitikA will be virtually perfect, in every sonic aspect that I can think of.
I will certainly post about it, however it goes.
The idea is that it is likely that the impedance of the power supply is going up as the frequency rises, and by adding a low ESR/ESL cap it will gain slightly better treble extension, and gain a bit more of the subtle details and quiet clarity that the B&K ST-140 had in abundance. If this is the case, then I think the AkitikA will be virtually perfect, in every sonic aspect that I can think of.
I will certainly post about it, however it goes.
I think it’s more important on the output cap to the speakers. Where the coil of wire is wrapped around it. The signal to the speaker has to ultimately go through that cap. The one back at the power supply is farther upstream and I think less important to treble.
Okay, I will keep that in mind. I did wonder if the context of this mod was being missed. I will try the power supply mod, because I only ordered the one cap, and let folks know if it helps or not.
I am happy to report, that as I have continued to listen to the AkitikA GT-102 Z4 amplifier - that it is now fully "opened up" - and as far as I can tell, is perfectly reproducing all the lowest level details and dynamics, on each recording. I will not be attempting any mods of this amp - I don't want to, because I think it is better than any amp I have ever heard. In every way - it exceeds what I imagined possible.
With my speakers at least - which are fairly sensitive (at least 90dB and possibly as high as 92dB?) and fairly low impedance - they are ~3.1 ohm minimum.
I have heard/seen the term "transparent" used to describe the sound quality, for decades. And now, I am pretty sure I know what it means - that each recording you listen to is reproduced with its sonic qualities are clearly audible. The system - source to speakers - has virtually no discernible coloration of its own. Almost every recording sounds like you are hearing some version of the living and breathing humans making music - somewhere in your room. The space they were in is there on the better recordings, or in your room if it is closely miked.
With my speakers at least - which are fairly sensitive (at least 90dB and possibly as high as 92dB?) and fairly low impedance - they are ~3.1 ohm minimum.
I have heard/seen the term "transparent" used to describe the sound quality, for decades. And now, I am pretty sure I know what it means - that each recording you listen to is reproduced with its sonic qualities are clearly audible. The system - source to speakers - has virtually no discernible coloration of its own. Almost every recording sounds like you are hearing some version of the living and breathing humans making music - somewhere in your room. The space they were in is there on the better recordings, or in your room if it is closely miked.
You guys are not helping me not spend my children's inheritance! During Covid lock down I was looking for something to do. 50 or so years ago I built a bunch of Dynaco, Heathkit, Elco, Knight.....So I decided to build an amplifier...I bought and built a Tubes4HiFi 60w/ch Dynaco clone "ST120".....Came out great and sounds magnificent!.....A few weeks ago I found and bought a First Watt F5 v2 as I always wanted a Pass Designed amp. Thought about the ACA but driving Magnapan MMGs and/or KEF LS50 Metas in a 24 X 12 X 8 room, I just couldn't live with the low power, even a pair in mono. (Maybe I was wrong?)......Anyways, I'm totally blown away by the F5 v2.....I've never heard these Magnapans sound like this. And it's not a small difference! It is easily discernable(Other amps in the arsenal for comparison (Cambridge Audio 840 Azur, Sound Artist SA200ia, Yamaha M45, NAD C320BEE....Yeah I know, I got a bad problem!)....I can't tell you how many times I go to the Akitika web page and nearly order a GT-102 Z4 !......And then as a sign from God almighty....My wife goes out last week and buys herself a midlife sportscar ..... a BMW Z4....It's an omen I tell ya!
So, you got one of their last ones? They are about to release the GT-104 which is a discrete version of the same circuit.