JC-80 eBay PCBs

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Assuming 22 ohm Caddocks, I can (by ordering for qty=2 PCBs) get the parts and PCBs for $80/channel aside from the J74 / K109. Since I already have a matching pool of those it is hard to guess what the cost might be to j-random fanatic.

A quick run down of the BOM (without revealing it) is Holco 1% non-magnetic resistors, Nichicon Muse KZ & ERO MKP caps, and a sufficient assortment of SMD resisors for the servo to allow matching. The one advantage a 'lone Wolfsin' has is the time to carefully match parts.

The only 'original Toshiba' are the singletons so if other components need special consideration, please raise a red flag. I am satisfied with that level of expense but feel certain that a GB could result in $75 JC-80s -- a price point at which I would expect considerable demand.
 
Dale CMF60 or similar brown colored Dale resistors are an inexpensive high quality resistor. Both Digikey and Mouser carry them.

At 18 volts or so above 20,000 ohms 50 PPM tempco should be fine. 25 PPM for lower values if you are fussy.

Above 100K you might consider the standard Xicons.
 
Dale CMF60 or similar brown colored Dale resistors are an inexpensive high quality resistor. Both Digikey and Mouser carry them.

The Holco are equivalent I think. Dale custom assortment could be had at additional shipping of $4 or ordering from DK or Mouser for about $6 more shipping. So it is $10, $14, or $16 for the thru-hole resists for two channels.

Not sure how to interpret non-quoted part of post, @simon7000
 
Thanks for the QC, John

Let me take a stab at what that earlier post was meant to communicate. Please correct if I stray. Matching is good but temperature coefficient must be considered as well, especially for low resistance values (most of the thru holes) or operating at voltages less than 18 volts (i.e. the entire servo section SMDs). Thus it might be necessary to make finer distinctions than merely SMD & thru hole as in the BOM.
 
Has "Sponge Bob" succeeded again?


sponge bob has been brought under control ...


bob_esponja_britney.gif
 
Housekeeping

I am down to three items that I cannot fully spec on Mouser or DK. 1)dual JFETs discussed earlier will be matched/built with singletons, 2)2SK2013 & its complementary are modestly priced on eBay, 3)tubular caps C2 & C3.

The pair of axial tubular caps marked C2 and C3 on silk screen parallel the electrolytics to ground on the 24v rails. The BOM description is insufficient to identify the part on DK. Would this ideally be polypropylene? Is the value critical?
 
Member
Joined 2005
Paid Member
The pc board has space for eletrolytiics in parrallel with a film cap. The on board regulator is a cap multiplier. The on-board regulator is used as the final stage of a multi-stage regulation scheme and is more of a noise filter and local energy storage.

Using an R core transformer (split bobbins are OK too) do not use a torroid as they too wide band , seperate positive and negative soft recovery bridges are an optional detail which would be good,, CRC or CLC filter then LM 317/337 preregulators, then shunt regulators if you want to go all out as John did in his Blowtorch. If you research the first Blowtorch thread (not the current one) you'll find a lot of power supply design concepts in first few hundred pages.

The film caps on the preamp pc board should be polypropelene at a minimum and polystyrene or teflon if you want to go all out (MIT are good choice). If you do the extensive power supplies you don't need or want the electrolytics on the output of the on board regulator at all - just the film cap is required - the value or tolerance isn't critical it's construction (film and foil prefered) and non magnetic construction are important details. You do want the electrolytic bypassing the resistor on the regulator or you don't get the noise filtering.

The power supply for these boards must be more complex than just a diode bridge, transformer, and some filter caps as it actually won't work properly without some form of off board pre-regulation.

The sonic charcteristics of the final preamp is the result of the circuit design, passive parts section and quality of the power supply and the implimentation of them in the chassis (it swould be nice if the tranformer(s) and the first part of the DC filters were in a seperate chassis. A dual mono power supply would be part of an all out implimentation

If you want a cost effective version a single transfomer, single soft recovery bridge CRC or CLC filter and LM 337/317 regulators would be the minimum to consider

This is John's original design and I would defer to his opinion on any of this.
He points out that a carefull selection of passive components (typically non-magnetic and chosen by listening tests over decades should be adhered to. This and the power supply is likely more complex ( and more expensive) than you might have had in mind and there are some cost effective choices we can discuss here - such as John's Dale resistor suggestion. It would waste your Caddocks if the power supply isn't up to the circuit design and parts selection.
 
Last edited:
TNP, I agree with most of your suggestions. The power supply does need a 337/317 DC regulator after the power supply caps, and the power transformer should be EI, or R core.
The shunt regulator should not be necessary. High speed, soft recovery diodes are important, as well.
See, we really have progressed over 30 years! '-)
 
Last edited:
Your exchange provided an increased appreciation for the distributed regulation on the PCB. So that realestate is anything but wasted! Thanks to both of you.

If you want a cost effective version a single transfomer, single soft recovery bridge CRC or CLC filter and LM 337/317 regulators would be the minimum to consider . . . the power supply is likely more complex ( and more expensive) than you might have had in mind /QUOTE]

I had started the power supply for a center tapped toroid feeding a bank of TO-220 center tapped Schottky MBR41H100CT rectifiers. Sounds as if dual windings in an EI or R-core feeding split rectifier banks would be better. Should two LM317/337s in series follow the caps since "[JC] found that 2 regulators in series is better than one 'super' regulator." Torch-PartI .

Are those rectifiers OK? How important is a separate enclosure if EI is used?
 
It is actually better to NOT use the center tap of the transformer, so you should go to at least 4 diodes for the power supply. An E-I or R core transformer is better than any toroid because of the problem with high frequency transmission through the toroid. Schottky diodes are usually OK, unless they are 'super fast'. Soft recovery is what we are trying for.
The Blowtorch was more complex, and perhaps (over the top), but we followed the rectifier bridges with a single load cap, fully floating, for both supplies, that was split in voltage by a common mode choke and then 2 somewhat larger caps to provide a 'ground return'. This is practical with both the JC-80 and the Blowtorch circuits, because they are inherently symmetrical in current drain to both supplies, and so, do not pull the voltages to one side.
The 317/337 regulators follow the twin (supply) capacitors, and near the line stage, the cap multiplier removes any extra noise and transient problems that the 317/337 regulators have.
 
Last edited:
It is actually better to NOT use the center tap of the transformer, so you should go to at least 4 diodes for the power supply. An E-I or R core transformer is better than any toroid because of the problem with high frequency transmission through the toroid. Schottky diodes are usually OK, unless they are 'super fast'. Soft recovery is what we are trying for.
The Blowtorch was more complex, and perhaps (over the top), but we followed the rectifier bridges with a single load cap, fully floating, for both supplies, that was split in voltage by a common mode choke and then 2 somewhat larger caps to provide a 'ground return'. This is practical with both the JC-80 and the Blowtorch circuits, because they are inherently symmetrical in current drain to both supplies, and so, do not pull the voltages to one side.
The 317/337 regulators follow the twin (supply) capacitors, and near the line stage, the cap multiplier removes any extra noise and transient problems that the 317/337 regulators have.
Mr. Curl, what is the purpose of resistors across the CM chokes coil?
 
Images of JC-80 (servo in middle) & candidate PS

@kamis -- "resistors across the CM chokes coil?" should make sense to me because I am reading very carefully. Is that a reference to material outside this thread?

I will reread TorchPartI for details of the caps 'n choke combo used because I do not yet understand it well enough to spec parts. The PS PCB I had started is sufficiently flexible to allow for all of the requirements I understand except, possibly, the choke. I am attaching a photo of that board in the hope that a reader may know a source for additional copies. It is gold plated on a nickel substrate and measures 92x95mm.
 

Attachments

  • PCB_images.jpg
    PCB_images.jpg
    255.8 KB · Views: 1,083
Progress delayed, sorry.

From ticknpop, "just the film cap is required - the value or tolerance isn't critical it's construction (film and foil prefered) [is]" There are LOTs of available Soviet teflon caps ranging from 0.01 to 0.1uf. The smaller values physically fit better and are silver. They are NOS "well aged" about 20 years. A .022 fits perfectly and the shipped cost is below new tin-polyprop. Is this a good choice?

Have ordered enough used 22R Caddocks and NOS Toshiba 2SK2013/2SJ313 for two PCBs. The caps are the last non-PS part to not be ordered thru DK or Mouser. Anyone wishing to build can use the BOM supplied with the PCBs to identify components not specifically called out in this thread. The PS will need some special parts and I plan to eventually upload a BOM for it. I shaved about $50 off the cost of a pair of these boards by "going cheap" -- that's about 25% cost savings -- but hope to have not compromised the build quality.
 
I have attached a copy of a schematic from TorchPartI as a basis of reference for discussion. It does not show resistors for the CM choke referenced earlier by @kamis. I would like to cut back to dual secondaries on a single transformer, same diode count, CM choke, single pair of LM317/337 (i.e. shunt OR . . . left out) and no linear on PS. Is there a different, more representative, schematic I should have used? Am I wanting to leave out too much? Where should the anti-ringing resistors be added if required?
 

Attachments

  • TorchPS.pdf
    21.5 KB · Views: 547
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.