The soft start controls the turn on surge. Not the bias.
The .15 resistors help stabilize the bias. They could be larger than .15, maybe as large as .47 but I haven't tried it.
Greg used 1 .15 per rail for simplicity. He likes to keep everything super simple if possible. I used a .15 per mosfet. More parts, yes. PITA to install, yes. But it works better.
Don't over think all this and look for problems that aren't there.
Mike
The .15 resistors help stabilize the bias. They could be larger than .15, maybe as large as .47 but I haven't tried it.
Greg used 1 .15 per rail for simplicity. He likes to keep everything super simple if possible. I used a .15 per mosfet. More parts, yes. PITA to install, yes. But it works better.
Don't over think all this and look for problems that aren't there.
Mike
Member
Joined 2009
Paid Member
The source resistors play multiple roles, including reducing turn-on surge current and helping with thermal stability of the bias. Increasing the value will have other effects (such as open loop gain of the FETs) so I'd just stick with 0R15. If you happen to have 0R22 in your junk box, for example, then I'd say you can use them without any worries.
I was talking about the Mosfet types there, since the resistors were added when these were changed and the PCBs amended. There was a brief explanation on the SKA open forum at the time when I assumed the replacement type had lower Rds but no, I don't have an original to compare, though gallery pics on the forum support this since only two 10R/1W resistors are of any size in there. So, that's why the suggestion to compare the Mosfet types fitted in the modified and original designs, as is AndrewT's.
Greg's comment at the time of release when Toshiba 2SK1530/J201 were suggested as options, was no - the design requires a relatively high Vgs of 3.5V to work properly. AFAIK, nothing's changed since then.
A single shared Source resistor has an effect equivalent to two separate double value Source resistors.
i.e. two 0r3 gives the same bias stabilising effect as a single 0r15
0r3 for a 2pair FET output stage is quite high for Source Resistors.
That is a very big change in philosophy for a stage that was supposedly bias stable by just selecting Vgs to within 10mVgs.
i.e. two 0r3 gives the same bias stabilising effect as a single 0r15
0r3 for a 2pair FET output stage is quite high for Source Resistors.
That is a very big change in philosophy for a stage that was supposedly bias stable by just selecting Vgs to within 10mVgs.
Member
Joined 2009
Paid Member
I got some spare time for basic measurements of my v1 board (I will have to improve it a bit). Now is time for listening some tunes. The bias is way to much overcompensated so i will try to add another diode to diode connected transistor and increase resistor for CCS. In my case I am using bd139/140-16 for bias sensors (maybe different transistors will be beter but I will try diode anyway). There is no hum noises at all, amp is rock dead with input shorted.
Now first impresions, the bass is really ok, the hights I like but they are different to the other amps I had. Really nice amp.
The measures show falling down harmonics spectrum - maybe this is the key to the good sound --> I do not know. I had to set bias at minimum 150mA to get really good sound out of it.
Big Thanks for Greg --> good job.
thx for bigun for support and answering my stupid questions
Now first impresions, the bass is really ok, the hights I like but they are different to the other amps I had. Really nice amp.
The measures show falling down harmonics spectrum - maybe this is the key to the good sound --> I do not know. I had to set bias at minimum 150mA to get really good sound out of it.
Big Thanks for Greg --> good job.
thx for bigun for support and answering my stupid questions
Attachments
Last edited:
Member
Joined 2009
Paid Member
Very nice to see measurements - you must have some nice equipment. What is happening as the frequency goes up - looks like D3 is decreasing between 1k and 10k ??? - I assume "D3" is the third harmonic. Anyhow, even though I don't understand all these things I do like the distortion profile, it looks very nice being dominated by D2 (2nd harmonic).
You are right about the listening impressions, I also found the highs are different to other amps. I didn't get a chance to listen enough yet though.
Last edited:
Hi G; interestingly, Greg provides the choice in the build guide for the builder to leave out the bootstrap caps (C4,5) for a 'fuller, more tube like sound' of increased lower order harmonic distortion, predominantly 2nd. I haven't heard of anyone trying it, but it's intriguing...
Member
Joined 2009
Paid Member
One channel just finished..
Just finished one channel..I was very impatient . So, bias was set it up, and I connected my very old Tesla speaker. I never ever heard so clear sound, deep bass... Then I have to connect my another speaker, hand made using Seas speakers.. Simply the best, what i ever heard.. I have to finished second chanel, power supply and listen and listen
Just finished one channel..I was very impatient . So, bias was set it up, and I connected my very old Tesla speaker. I never ever heard so clear sound, deep bass... Then I have to connect my another speaker, hand made using Seas speakers.. Simply the best, what i ever heard.. I have to finished second chanel, power supply and listen and listen
Attachments
