Salas,
I ran several bias simulations and it would appear as if the harmonic distribution remains unchanged between 120mA and 1,4 Amp with second harmonic predominant and smooth decay of all higher orders.
Dropping the bias below 110 mA third, fifth and ninth harmonic becomes dominant and THD increases by an order of magnitude.
I will make measurements on the week-end as I do THD measurements manually and it is a tedious process.
Regards
Nico
I ran several bias simulations and it would appear as if the harmonic distribution remains unchanged between 120mA and 1,4 Amp with second harmonic predominant and smooth decay of all higher orders.
Dropping the bias below 110 mA third, fifth and ninth harmonic becomes dominant and THD increases by an order of magnitude.
I will make measurements on the week-end as I do THD measurements manually and it is a tedious process.
Regards
Nico
Salas,
I tried to optimize bias for best harmonic distribution but as I said it remains almost constant. I conclude that there is no benefit to run a bias current higher that 140 mA. I am referring to my circuit only, Lazy Cat may have different results using the isolated gate set-up.
Regards
Nico
I tried to optimize bias for best harmonic distribution but as I said it remains almost constant. I conclude that there is no benefit to run a bias current higher that 140 mA. I am referring to my circuit only, Lazy Cat may have different results using the isolated gate set-up.
Regards
Nico
Member
Joined 2009
Paid Member
Hi Bigun, welcome here. Although Lazy Cat has mentioned an all BJT design I do not think anyone else has built anything yet. Lazy Cat did suggest that all three types could be available in the near future and I think much is going to depend on Alex's willingness and time to assist in laying three different PCBs.
Nico
Nico
Member
Joined 2009
Paid Member
Hi Nico, I don't want to ask for time from Alex when priority is the 'popular' version of this amp - besides I may need to make my own pcb yet, in order to suit the physical constraints. If I go ahead with this project I'll be fitting 3 channels into an old Onkyo stereo chasis. But I was hoping there may be some practical experience of building the all-BJT version I could learn from and so help me decide which amp to make.
Yes, indeed excellent, Nico.
I was surprised too, so I double checked and 33ns it is, as you said almost the same as 29ns from simulation, very close to the theoretical limit.
Closed loop gain and phase shift are perfect at 100kHz and far beyond, which is clearly audible as a "no amp present" effect, music only.
So what I have here is super fast, linear and stable amp, but still easy to built even to a DIY newbie.
I run an amp with bias current set at 150 mA all the time, lower is non-sense, higher will check.
Bigun please reward us with your active presence to this thread, an all BJT version of SSA could be yours participation to DIY-ers, since I will prepare mosfet driver/BJT output version
Member
Joined 2009
Paid Member
I'm very tempted, not sure I can keep up with you guys though as I have other projects on the go but I will investigate this option more seriously now...
I'm curious why this amp produces predominantly 2nd harmonic when it is symmetrical (from Nico's simulations) ? is this a consequence of generally poor complementary matching between n-type and p-type FETs. One reason I built only asymmetrical to date is because I like the harmonic signature.
I'm curious why this amp produces predominantly 2nd harmonic when it is symmetrical (from Nico's simulations) ? is this a consequence of generally poor complementary matching between n-type and p-type FETs. One reason I built only asymmetrical to date is because I like the harmonic signature.
Last edited:
This triple darlington configuration is of course possible for the output stage to achieve a super high current gain factor, but the question is thermal, electrical stability and the benefits of this arrangement.
Certainly you could check behavior of this compound transistors in any amp that enables bias setting/thermal regulation for all Vbe voltages of the BJT's mounted on the same(main) heatsink.
I suggest quiescent curents through R1: 5mA, R2: 25mA, R3: 150mA
Certainly you could check behavior of this compound transistors in any amp that enables bias setting/thermal regulation for all Vbe voltages of the BJT's mounted on the same(main) heatsink.
I suggest quiescent curents through R1: 5mA, R2: 25mA, R3: 150mA
Your Ib suggestions are exactly/precisely the same as I was thinking.I suggest quiescent currents through R1: 5mA, R2: 25mA, R3: 150mA
That is an interesting question Bigun, since the driver stage almost contain higher odd order products.
When changing the bias I tend to think that the addition of the harmonic content of the output stage adds even order products until biased to the point that the second harmonic becomes predominant.
This amp by no means sounds "class A", warm and snugly, the harmonic distribution decays almost linearly making it sound crisp, clean, vibrant and refined.
Maybe not in every ones taste but certainly I remain very impressed. Switching to my normal amp on my PC makes mine (also about 40 watt) sound a little bland and lifeless and it takes getting used to again.
Member
Joined 2009
Paid Member
Hi Nico,
Thanks for the info. For my HT project I believe a clean and vibrant sound would fit the bill nicely. This seems consistent with low distortion and not a lot of H2. Your distortion plots show all distortions as being low. Of course with an all-BJT output it's a different amplifier but the front end and OLG should all be the same.
I'm not familiar with your 'normal amp' ? - I find it better to have some idea of how you feel the sound compares with other amplifiers rather than as an absolute as this may help me relate better to my own tastes and experiences.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.