VSSA Through-Hole Version by Jason

Status
Not open for further replies.
Why are the VAS devices NOT on the main heat sink? My VSSA layouts are intended to place the VAS devices onto the main heat sink under the board. EXACTLY which circuit / version did you build?

I haven't heard too many people complain they are unhappy with the sound. I generally consider it very crisp, clear and detailed. It might sound a little thin in the bass region at first, but it is actually very accurate once you realize the detail that is being revealed.
 
I think guitar.mod might be right. I'm going to check the feedback resistors.

I absolutely agree with you Jason about the VAS transistors. They should be placed onto the main heatsink and that's one of the reasons I'm waiting for your boards 😉
I used an old pair of boards from PMI I had with me. It's Rev. B without CCS. There is no way of placing the VAS transistors onto the main heatsink with these boards.
 
Ah, OK. We are not actually talking about my one of my layouts at this point. Whew! I'm actually a bit relieved. I had a set of PMI's revision C boards but I only built, tested and then sold them. The version you built is actually more like Shaan's PeeCeeBee than LazyCat's VSSA since there is no CCS.

The feedback loop resistors should be within 1% of each other, or closer if you can select from a small batch of 1% devices. Match the two 2K2 and the 100R to ensure the positive half and negative half are operating with the same amount of closed loop gain.
 
Last edited:
Why are the VAS devices NOT on the main heat sink? My VSSA layouts are intended to place the VAS devices onto the main heat sink under the board. EXACTLY which circuit / version did you build?

I haven't heard too many people complain they are unhappy with the sound. I generally consider it very crisp, clear and detailed. It might sound a little thin in the bass region at first, but it is actually very accurate once you realize the detail that is being revealed.

I generally consider it very crisp, clear and detailed. It might sound a little thin in the bass region at first, but it is actually very accurate once you realize the detail that is being revealed
Agree!
I have built PMI version on my hand made P.C.B
I have built the Shaan version also.http://www.diyaudio.com/forums/solid-state/270418-vssa-pmi-version-vs-peeceebee-shaan-version.html
 
Last edited:
I had a 2k4 resistor paired with a 2k2:headbash::headbash::headbash:
Now I need to fix the wall😀

Replaced those evil 2k4 with two 2k2 0.1% in both channels. What a difference!
As you guys said, sound is crispy, detailed, beautifull highs, lovely natural mid range and the bass is just right - deep and tight! I'm not a basshead, the bass seems right natural more of it and it wouldn't be as natural. Also the soundstage is pretty big!

Thank you guys for your help 🙂
 
You are welcome! 🙂

p.s. You measured 2k4 value from 2k2 lot resistors?

No, I have my resistors organized in small plastic bags and that bag was labeled 2k2. Turns out it had 2k4 mixed with 2k2 resistors. As they were Vishay Dale CMF55 the printings are vey small and I must have missread them when I stored them in that pastic bag. Now I measured the 2k2 with the DMM just to be sure. I have to go through all those plastic bags and check all the resistors so I don't make the same mistake again.
 
Last edited:
Gain can be adjusted some, as with anything pushing too far one way or the other can lead to reduced performance. I think someone did experiment a little with gain but ended up going with the recommended values after all since what was being corrected for was a wrong value resistor in one of the builds.
 
I don't set it by voltage reading. Put an ammeter inline with one of the power leads and actually measure the bias. There are detailed set-up instructions posted a few times, at least one linked in the first post. That should help with getting things up and running.

Like most amplifiers, start at minimum bias setting and adjust the Vbe multiplier until the target is achieved.
 
yes please, 🙂

Thank you!

From what I remember, from LC's VSSA thread he got 1.5165V across TLVH431 and it stayed there once it warmed up. I cannot source the TLVH431 here, so for easy bias the Vbe multiplier is the best option for me which you have used on your build.
 
I'm sure the MJE's will work. Others have used them in various designs and they seem to work out, though I don't know for sure if anyone has used them in a PeeCeeBee or a VSSA. If your supply is <+/-40V you can use BD139/BD140. It is a tolerant design.

Use NP0 / C0G ceramic or a film capacitors if the silver mica is too exotic in your locale.Ceramics like X7R are not good for compensation, only bypassing.
 
My Rev'B' build is biased a little lower than some, but I have a spread of 1.358V that gives me about 100mA through the output MOSFETs.

Thank you Jason,

Are you still researching on using jfet for ccs?
I have build a prototype using your scheme of Vbe multiplier and used jfet for ccs (2SK170 and 2SK246) and they are very good. 😀

I will try to post pics
 
Last edited:
Status
Not open for further replies.