Replace R27 with 620/680 ohm. Expect the 2.4 volt to rise a little bit, when you do this, so 680 might be right.
YES...... D1 is glowing constantly. With R27=680 ohm it will glow a little less, but still be constantly lit up. This is indeed normal 😉
YES...... D1 is glowing constantly. With R27=680 ohm it will glow a little less, but still be constantly lit up. This is indeed normal 😉
Hello meanman1964,
if you have the 'original' resistorvalue of 220 Ohm at R27 and you measure 2.4 V over it, that equals 0,0109 A or 10.9 mA.
So, your bias is too high. The bias should be anywhere near 4 to 5 mA. I would try a 470 Ohm or 560 Ohm resistor at R27.
As long as your PEARL 3 is switched on, as long D1 should lit up. D1 is a part of the biasing-circuit (Q6 / Q7 and U1B).
If you have big caps in your PSU, it can take some time, till D1 shuts off - after switching off the PSU.
Cheers
Dirk
if you have the 'original' resistorvalue of 220 Ohm at R27 and you measure 2.4 V over it, that equals 0,0109 A or 10.9 mA.
So, your bias is too high. The bias should be anywhere near 4 to 5 mA. I would try a 470 Ohm or 560 Ohm resistor at R27.
As long as your PEARL 3 is switched on, as long D1 should lit up. D1 is a part of the biasing-circuit (Q6 / Q7 and U1B).
If you have big caps in your PSU, it can take some time, till D1 shuts off - after switching off the PSU.
Cheers
Dirk
Dirk , thanks .
With 560 Ohm I get around 0,0048...
With 560 Ohm I get around 0,0048...
Last edited:
Why were surface mount parts used in this project when there is through hole parts available?
Assuming you are referring to Q1-Q4 (or maybe even U1), perhaps a slightly altered set of questions may get you to the answer you seek.
Why did a thoughtful designer allow multiple options for both surface mount parts and through hole parts? Why did they take the time to carefully note the options in the schematic, when some people won't look at the schematic? Why were the much less expensive and commonly available surface mount parts included in the kits vs. the vastly more expensive, obsolete through-hole parts? Why did the designer make the effort of leaving a through-hole option for the people that already have or want to source the comparatively rare, comparatively expensive, obsolete through-hole parts that for best performance need to be tightly matched adding additional time and expense?
That's just a few example questions... <sarc> intended, but not for ill-intent, just fun.
Short answer... You may not know that you can use through-hole parts if you want. 🙂 There is an exceptional set of build documents. Read them.
Last edited:
Big grins and up-with-the-thumbs on this one
No, there is a danger of possibly learning something .... and I usually avoid that .........
Actually I was wondering if the best part for the job was only available in surface mount and the thru hole parts were available as an option for those that cant solder surface mount parts. Can you please show me where it addresses this? I must be too stupid to have noticed it. But thanks for reminding me why I avoid these forums. Seriously, what a dick thing to say. You didn't even understand the question.
Last edited:
Best part for the job is probably a subjective human judgement; different builders probably have different opinions. In my own opinion the 2SK209BL / 2SK209GR is the best part for the job but others may disagree. Even better if you can have your lab technician measure 50 parts and hand pick two tightly matched quads, one per channel. Personally I'd also be favorably disposed to pay the higher cost, to get 0.1% tolerance parts for the JFET source resistors. Mouser.com sells them for $0.34 each in quantity=10. P/N 279-YR1B10RCC . However that might not be "best" in the opinion of others.
Last edited:
Thanks.. I thought there might have been a difference in part performance as most people wouldn't choose to use surface mount parts on a board with so many thru hole parts which negate the space savings angle.
which parts shipped with the kits for the input fets? The chips have an X G on them if I recall correctly.
Properly laid question will get proper answer
Simple as that
You didn't explained core of your dilemma, and you presume ill will where just good intended fun is present
But, one need some effort to recognize anything in any group of people, online or not
If you want quick and straight service, pay for professional one
Here you'll always get everything and more, for free ... but with some nice mockery
^ First, I'd recommend lightening up. Your call. You seem to want people to read your mind vs. what you actually write. That's never a great idea, IMO.
If your overall questions is similar to: "Will the SMD parts provided in the kits, with no other modifications to the recommended circuit, outperform whatever it is that I have in my brain that I have not yet expressed in terms of how I might use some as-yet-to-be-mentioned through-hole parts?"; then I'm sorry I didn't pick up on that immediately. The answer is - I haven't the foggiest of clues.
What I'd recommend is -
If you are not comfortable with surface mount parts; either get someone to do it for you, practice, or use through-hole parts. For through-hole; I'd recommend the "170s" from either Toshiba or LS. As you've seen you'll get lots of opinions. Mine is just one.
If you are comfortable with surface mount parts, use those provided. Mark also provided some fun ideas to really "take it to the next level". They're the 209s, but I'm basing that on a memory that should not be trusted.
I'd bet $100 USD that you'll never-ever in a controlled study be able to hear the difference between properly built versions of the versions with through-hole or SMD with or without tweaks, but it's DIY. Take it where the path leads. We do all sorts of fun things. Or, as Kasey Musgrave put it so eloquently... "Follow Your Arrow".
Seriously, as I wrote, there was no ill-intent in my sarcasm. It's a fun place to hang. I hope you'll stick around.
Good luck with your build.
If your overall questions is similar to: "Will the SMD parts provided in the kits, with no other modifications to the recommended circuit, outperform whatever it is that I have in my brain that I have not yet expressed in terms of how I might use some as-yet-to-be-mentioned through-hole parts?"; then I'm sorry I didn't pick up on that immediately. The answer is - I haven't the foggiest of clues.
What I'd recommend is -
If you are not comfortable with surface mount parts; either get someone to do it for you, practice, or use through-hole parts. For through-hole; I'd recommend the "170s" from either Toshiba or LS. As you've seen you'll get lots of opinions. Mine is just one.
If you are comfortable with surface mount parts, use those provided. Mark also provided some fun ideas to really "take it to the next level". They're the 209s, but I'm basing that on a memory that should not be trusted.
I'd bet $100 USD that you'll never-ever in a controlled study be able to hear the difference between properly built versions of the versions with through-hole or SMD with or without tweaks, but it's DIY. Take it where the path leads. We do all sorts of fun things. Or, as Kasey Musgrave put it so eloquently... "Follow Your Arrow".
Seriously, as I wrote, there was no ill-intent in my sarcasm. It's a fun place to hang. I hope you'll stick around.
Good luck with your build.
If a question is unclear asking for clarification seems the better course than putting somebody down or getting sarcastic. Its a good thing people don't act this way in person. Only when there is no accountability.
So I take it all the alternate parts listed are pretty much equal. I have no problem with smd parts. I was just asking to see if there was a noticable difference as I could discern no other reason to make smd an option if the thru hole parts were readily available. (unless the majority of the board was smd in which case space savings would be a good reason).
So I take it all the alternate parts listed are pretty much equal. I have no problem with smd parts. I was just asking to see if there was a noticable difference as I could discern no other reason to make smd an option if the thru hole parts were readily available. (unless the majority of the board was smd in which case space savings would be a good reason).
Don't forget to investigate the money difference. Maybe it will matter; maybe not. (It's less than a bottle of French champagne at a white tablecloth restaurant).
@jwags81818 - as others have indicated, the option to use 2SK209s (or several others as shown on the schematic in Post #1 of this thread) cost about $0.50 each, or about half that if you buy 100. The through-hole parts are obsolete but some can be bought already matched in the diyaudio store for about $10 each. Big delta. Those parts were accommodated in large part, I believe, because there is now a long history of Pass diy projects using those parts because Nelson likes them so much - they've become de facto identifiers of many diy Pass projects. Fortunately for those of us who don't want to spend that much on a few parts, Wayne designed the board to accommodate either.
I built my TH version with the 209s, then as an experiment, removed them and installed matched 2SK170s. I really couldn't tell much difference - swapping op amps seemed to me to have a greater effect.
Subsequently I bought 100 2SK209 parts and spent time matching them closely (I don't have a lab assistant, just my very accurate DC current source and power supply instruments, and my own test jig). My view now is that using tightly matched SMD JFETs along with 0.1% tolerance source resistors is about as good as it gets, and is totally consistent with the parameters of this circuit.
You will find a lot of potential questions answered in the excellent build documentation for this project - both Randy's very detailed build document, and Jim's as-always excellently photo-illustrated build guide. There is also a whole bunch of additional builder-generated data available within this now long thread.
One thing not apparent to many people, it seems, is just how much effort a relatively small handful of people put into making these projects happen, and they often have to answer the same questions many times, or questions that are answered in the original documentation. Your initial question was direct and simple, but was also easy to read into or misinterpret. Hopefully you have had your immediate question answered, and will be interested in reviewing the great documentation.
Cheers
I built my TH version with the 209s, then as an experiment, removed them and installed matched 2SK170s. I really couldn't tell much difference - swapping op amps seemed to me to have a greater effect.
Subsequently I bought 100 2SK209 parts and spent time matching them closely (I don't have a lab assistant, just my very accurate DC current source and power supply instruments, and my own test jig). My view now is that using tightly matched SMD JFETs along with 0.1% tolerance source resistors is about as good as it gets, and is totally consistent with the parameters of this circuit.
You will find a lot of potential questions answered in the excellent build documentation for this project - both Randy's very detailed build document, and Jim's as-always excellently photo-illustrated build guide. There is also a whole bunch of additional builder-generated data available within this now long thread.
One thing not apparent to many people, it seems, is just how much effort a relatively small handful of people put into making these projects happen, and they often have to answer the same questions many times, or questions that are answered in the original documentation. Your initial question was direct and simple, but was also easy to read into or misinterpret. Hopefully you have had your immediate question answered, and will be interested in reviewing the great documentation.
Cheers
As they say online: "Porque no los dos?" (English: "Why not both?")
Why not purchase an excessive number of 0.1% tolerance resistors and then hand-match those for even tighter tolerance? Buy 50 of them (total cost: $19.00) and pick out two of the best-matching quads.
An absolutely worst-possible-scenario analysis with pencil and paper, tells you that if you start with a population of 50 parts, you're guaranteed to get two quads, each quad matched within ±0.012%. ([± 0.1% / 49] * 3). In real world scenarios, with random resistances that follow a uniformly probability density (or, better yet, a Gaussian bell-curve), the matches will be significantly better. Monte Carlo simulations can tell you exactly how much better, but the time it would take you to create and debug and run the simulation, is undoubtedly worth more than $19. So go ahead and buy fifty parts, then spend an enjoyable hobby hour, testing resistors.
A few years back I did such a Monte Carlo analysis, looking for matched pairs (not quads) of resistors to make the diyAudio IPS6 and Cedarburg amplifier cards really sing. Shockingly few candidates were needed to find shockingly tight-matched pairs. It's somewhere here on the Forums but I lack the motivation to go searching. I have a feeling that a Monte Carlo simulation hunting for quads, would require a lot more "housekeeping" to get everything right. But my pessimism may be unwarranted.
Last edited: