Jean Paul is right , few doesnt mean anything ,remenber there are some newbies around .....
a good advice is one I can use
with love anyway
.
Last edited:
Long before you joined the forum, this is what Nelson said about the Erse inductors:
http://www.diyaudio.com/forums/pass-labs/1898-soz-choke-input-filter-2.html#post21827
I just breadboarded one channel of the B1 R2. Seems to work fine, but I have a couple questions:
1. The offset seems to move around, at values <1mv. What should be the target value?
2. I tried various supply voltages, but didn't notice any real difference in distortion (typically well below 0.01%, usually about half that). Any downside to running the circuit at +/-15V vs +/-12V?
Thanks,
Mike
1. The offset seems to move around, at values <1mv. What should be the target value?
2. I tried various supply voltages, but didn't notice any real difference in distortion (typically well below 0.01%, usually about half that). Any downside to running the circuit at +/-15V vs +/-12V?
Thanks,
Mike
I think because rev1 is well estsblished, it has boards from passdiy.com store, can get boards with matched jfets.
The rev2 is very simple, even for a beginner to try point to point soldering.
I heard both and rev2 is excellent, without cap coloration of rev1. I have the rev1 myself and wouldn't hesitate to build the rev2.
The rev2 is very simple, even for a beginner to try point to point soldering.
I heard both and rev2 is excellent, without cap coloration of rev1. I have the rev1 myself and wouldn't hesitate to build the rev2.
Yes, it looks like it. It seems that copper perfboard or stripboard would be the best way to do this (combination of efficiency, elegance, and performance)? ...I've not built anything from scratch before, but thought I might give the B1 or B1 Rev2 a try, in part because it has the chance of equaling or bettering my existing preamp which has seen off most competitors under $1500. I've done a bit of replacing and upgrading parts on existing components.
Biggest thing for you to overcome is schematic and component placement. Examine the parts for polarity and pin assignments, like the jfets gate, source and drain connections and how they apply to the schematic. Envision what you will do before doing it and go over it, over and over, until it makes sense. Try to install active components last, if possible. Reduces the chance of static discharge damage.
Some schematic questions
Thanks for that. A few questions:
Thanks for that. A few questions:
- What is the P102, a 50R trim pot? If so, how is it used?
- Can P101 be a 25k or 50k as with the the orig B1?
- My understanding was that NP recommended or used a 24v SMPS, so input to the schematic at V+ and V- is 24VDC, right? And D+ and D- is the output to the channel signal circuit? (I've not seen the D abbrev before.)
- So The right bit is applied to smooth the DC input that may have already come off of a smoothed linear PSU? (I'm thinking schottky bridge with LM317, smoothing caps, and any approp support.)
- The PS smoothing on the right is one circuit that is then split to both channels (L + R), or is it repeated for each channel? I see no channel numbering convention on it (101 vs. 201).
- Can it be a single 24+0VDC input or dual 12+12-VDC? Any possible benefit?
- If running two boards for balanced operation, might there be any benefit achieved by separate windings and PSU circuit for each board?
- By 'active' you mean the JFETS I imagine?
Last edited:
50 ohm pot is for zeroing out dc offset.
Stick with what is recommended, if it's 25k, use 20 or 25k.
The little ps circuit allows the use of a +24v supply in to a 12V +/- supply. You don't have to use smps, but they aren't bad either. I recently built a boz-J with a 24v smps that feed a pair of LM317 regulatorS and it sounded great. Kind of miss it...
You can use a dual linear -12v/+12v supply if you like.
I don't know about running it balanced.
Active devices are semiconductors and diodes, opamps, etc
Stick with what is recommended, if it's 25k, use 20 or 25k.
The little ps circuit allows the use of a +24v supply in to a 12V +/- supply. You don't have to use smps, but they aren't bad either. I recently built a boz-J with a 24v smps that feed a pair of LM317 regulatorS and it sounded great. Kind of miss it...
You can use a dual linear -12v/+12v supply if you like.
I don't know about running it balanced.
Active devices are semiconductors and diodes, opamps, etc
I suspected that. So I can re-learn about measuring DC offset separately, offline.
So if I am running a dual linear -12v/+12v LM317-based supply, I could omit the little PS circuit in the schematic altogether, not gaining any benefit from it?
Then only the JFETs for this schematic, though of course I'll have a few active devices on my linear PSU.
If you place the positive lead of a multi-meter (set to DC voltage reading) to the output connection after the 50 ohm pot and the negative lead to ground of the circuit, you might get a VDC reading. Adjust the pot until you can get it as close as possible to 0V (zero volts).
Because the circuit can use a dual output power supply of +/-12v, the cap on the output can be omitted in this design, but we need to be sure the DC component at the output is very low or zero DC. If a cap is used, such as in the Rev1, there is no DC, because caps don't pass DC. They pass a DC version of an AC signal. I hope I have that part right. Print out both circuits and have a look.
Yes, you can use a LM317 or LT1083/85. They are also a good example of an active device.
Vince
Because the circuit can use a dual output power supply of +/-12v, the cap on the output can be omitted in this design, but we need to be sure the DC component at the output is very low or zero DC. If a cap is used, such as in the Rev1, there is no DC, because caps don't pass DC. They pass a DC version of an AC signal. I hope I have that part right.
Print out both circuits and have a look. Yes, you can use a LM317 or LT1083/85. They are also a good example of an active device.
Vince
The other thing I wanted to add is that when building, insert the resistors first, then caps, diodes and lastly transistors and active devices. Don't get too hung up on this when doing point to point wiring, because it's not always practical, but do your best to follow the rule. Makes life easier.
Lastly, try to imaging what you will do before you do it, especially difficult tasks. Check and recheck your work, then check it again before applying power. Technically, the circuit still works before you turn it on. It's not until a catastrophic failure that it doesn't.
Good Luck! We are all pulling for you!
Vince
Lastly, try to imaging what you will do before you do it, especially difficult tasks. Check and recheck your work, then check it again before applying power. Technically, the circuit still works before you turn it on. It's not until a catastrophic failure that it doesn't.
Good Luck! We are all pulling for you!
Vince
Last edited:
I see the rec'd input voltage is 24v. What is the practical range of voltage input to achieve optimal sound performance? Do we have enough builds to know?
In regards to regs, do I see correctly that LM317 is available both in 3v drop and LDO?
Those LT1083 are pricey! Are they worth all that?
In regards to regs, do I see correctly that LM317 is available both in 3v drop and LDO?
Those LT1083 are pricey! Are they worth all that?
The linear technology regs are ldo, I don't know about 317/337s.
There are some good to-92 case ldo regs, but I've never used them.
This circuit is so low current draw that they may be an option.
I've never done a head to head comparison, but I think the LT1083 provides good regulation and low noise. Someone else my disagree.
There are some good to-92 case ldo regs, but I've never used them.
This circuit is so low current draw that they may be an option.
I've never done a head to head comparison, but I think the LT1083 provides good regulation and low noise. Someone else my disagree.
The 24v in is from the smps. It splits into +12/-12 via the small divider /floating ps. I don't know that this circuit benefits from higher supplies or tweaking, but I could be wrong.
For a beginner, you might want to stick with the circuit as shown. Also, the nature of the buffer keeps noise down.
The max voltage on the jfets is 28v, but is know to work at higher voltages like 32v and bit higher. Only with a cascode.
Here's the thing though, as shown there is a 24v differential, so that should be plenty for an incoming voltage.
For a beginner, you might want to stick with the circuit as shown. Also, the nature of the buffer keeps noise down.
The max voltage on the jfets is 28v, but is know to work at higher voltages like 32v and bit higher. Only with a cascode.
Here's the thing though, as shown there is a 24v differential, so that should be plenty for an incoming voltage.
I recall reading on the original B1 that the voltage could be 18v, 21v, even as high as 25v if I remember. Though someone said they thought it was better at 21 than at 18.
I only have a 19v & 20v laptop supplies around. If really needing 24v, I might as well build a linear supply straight off rather than buy a 24v SMPS as a temp measure.
Sent from my GT-I9505 using Tapatalk
I only have a 19v & 20v laptop supplies around. If really needing 24v, I might as well build a linear supply straight off rather than buy a 24v SMPS as a temp measure.
Sent from my GT-I9505 using Tapatalk