I have not built the original design but instead a variation using Jimsaudio PCB but with different output MOSFET (and consequently adapted parts and values accordingly) and I do not notice the "subdued" sound... It may be my hearing that is subdued....
Fab
First of all, the difference is subtle but certainly there. Without a good A/B setup, it may not be as noticeable. I have checked it many times against many different amps. Of course, if you built it differently than the designer all bets are off.
Yes that has been observed with this amp. For example, a DCB1 preamp +SKA (modified) gave better dynamics than an LDR attenuator + SKA (modified).
Fab
Yes, I do not doubt what you say, I just want to say that a modified SKA with different MOSFET (2SK1530Y/SJ201Y) may sound good.... Of course, it is a matter of taste.
Fab
Member
Joined 2009
Paid Member
My TGM7 is based on the same topology and I believe it would sound very much like the SKA. I don't find it subdued - just neutral beyond belief - amazingly clean treble and super tight bass - perhaps to the point that it would hold no interest without some 'flavour' from other components in the system. It might be the ideal power amp to pair with a tube buffered volume control. My TGM8 doesn't have the same limitation - even Terry liked it
The SKA has very high unity gain and gobs of feedback - I don't see component variations being too relevant so long as they don't perform poorly enough to pull the open loop gain down. I take my hat off to Greg for the design, it's a really superb amp when you want something very clean and neutral.
The SKA has very high unity gain and gobs of feedback - I don't see component variations being too relevant so long as they don't perform poorly enough to pull the open loop gain down. I take my hat off to Greg for the design, it's a really superb amp when you want something very clean and neutral.
Last edited:
I built two SKA GB150s. One using mica caps and one using multi layer ceramics. They sound identical. When you say your amp is "based on" that doesn't meant it sounds identical. I would actually be surprised if Lab's choice of output transistors has much of an affect on the sound. IRFP outputs have been used for years on some very popular amps. I even have a Slewmaster OPS built using them and there is no sound change that I can hear. The SKA sounds very good but compared to almost every amp I have built it sounds subdued. That may however, be very pleasing to some folks.
Member
Joined 2009
Paid Member
My TGM7 amp is based on 99% the SKA topology - the difference is that the SKA uses a diode-wired transistor in the CCS as the thermal feedback sensor from the output stage to stabilize the amp. I saw that this was a redundant component and removed it. Instead, I use the CCS transistor itself as the thermal sensor and it works very well. Where there is a larger difference is that my pcb is much more advanced design than the SKA. My design is far more compact, has better thermal feedback, requires no fly-wires and achieves lower inductance traces with much better signal routing. It took me 3 goes to get the pcb right and when I posted it on Greg's forum for some help I didn't get very much. Unfortunately, Greg was so upset when I posted a photo of my working pcb on his forum that he deleted my posts and closed the thread.
There could be some sound differences because of the layout but I've no way of knowing.
There could be some sound differences because of the layout but I've no way of knowing.
Last edited:
If "Lab" refers to me, I wonder how you know how my amp sounds ...
If you believe that the SKA sound can not be improved and the IRFP transistors are the best for audio then it is your own prerogative.
Fab
Last edited:
Yes, I have no idea what Greg was expecting when he decided to release the design to public domain but it is clear he was not really interested in seeing what others would do with it. I think it is a cool little amp and worth building. Do you think your TGM7 sounds identical to the TGM8?
Last edited:
If "Lab" refers to me, I wonder how you know how my amp sounds ...
If you believe that the SKA sound can not be improved and the IRFP transistors are the best for audio then it is your own prerogative.
Fab
Please forgive my typo on your name.
I'm not saying that the IRFP's are the best. I'm just not convinced that different output devices have much influence on sound. Power handling, yes, SOA, yes, but sound? I have my doubts. YMMV.
Yes I have 2 of your tiny TGM7-II PCB but I have not found the time to mount them yet because too busy with other projects.. I intend to try the CFA implementation because I tend to like CFA Amps these days.... There are also SMD parts on the PCB and I am not that much confident with my soldering skills for those small parts..:
maybe that is what is preventing me have begun earlier....Fab
Please forgive my typo on your name.
I'm not saying that the IRFP's are the best. I'm just not convinced that different output devices have much influence on sound. Power handling, yes, SOA, yes, but sound? I have my doubts. YMMV.
You are still forgiven...
I am surprised that you have not experienced difference in sound using different output devices for a given amp. For example, even with "equivalent" parameters for Lateral MOSFET, peoples have reported different sound when using different manufacturers (Exicon, Semelab, etc). The 2SK1530/2Sj201 have been specifically designed by Toshiba for audio applications.
Fab
I have built over 50 different amps in the last 18 months. All have been tested on my A/B setup. I haven't tried all makes of output devices but quite a large variety including lateral and vertical MOSFETs and of course a variety of BJTs. When the output volume is equalized there is virtually no audible difference between 90% of the amps tested. The other day a fellow forum member posted this link to a Clark study. I must say I agree with it.
Richard Clark Amplifier Challenge FAQ
That said, the SKA boards I have built do display a more subdued sound in the mids and highs. I'm not opposed to trying different things to see if it can be improved.
Richard Clark Amplifier Challenge FAQ
That said, the SKA boards I have built do display a more subdued sound in the mids and highs. I'm not opposed to trying different things to see if it can be improved.
You are still forgiven...
I am surprised that you have not experienced difference in sound using different output devices for a given amp. For example, even with "equivalent" parameters for Lateral MOSFET, peoples have reported different sound when using different manufacturers (Exicon, Semelab, etc). The 2SK1530/2Sj201 have been specifically designed by Toshiba for audio applications.
Fab
People report all sorts of things. Exicon is a brand name of Semelab, so these people are hearing differences between the same parts.
Member
Joined 2009
Paid Member
I'd have to go back and listen to them side by side - a bit of wiring needed to do that but it's possible. My recollection is that I preferred the '8 over the '7 but as you say, there's less difference between good amps, especially SS with gnf.
Renesas and Alfet (from Semelab) if you prefer.
My point is that the 2SK1530/2Sj201 have different characteristics than IRFP9240/IRFP240 thus a different sound can be obtained for the same amp. Depending on other factors the difference could not be heard if you have limiting elements in your whole audio chain....
If someone believes that all power MOSFET and all power BJT sounds the same them it is pointless to persue this discussion for me....
Fab
Last edited:
I agree that different characteristics will often provide a different sound. I was just a bit concerned that the printing on the can could affect the sound (other than in the mind)
BTW, I have no idea why Semelab need so many brand names and have their own (not very proactive) disti called Magnatec. Perhaps it's a pricing strategy thing?
I've noted most of the people building on Jim's boards prefer a higher bias even though Greg designed these to be "green amps" operating mostly in Class B (less than 15mA of bias). Does raising the bias skew the distortion numbers higher when the signal is less attenuated (approaches full power)? Member entenschreck seems to have successfully built this amp as intended by Greg and seems to be happy with it. I would like to follow his example.
Last edited: