Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 8, 2013 10:05:47 GMT
Thanks Alex,
DOH!
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 8, 2013 15:36:34 GMT
A sideways thought going back to scuppered ideas of an SMT version.
Would it be worth concocting a simplified i.e. closer to the original design (negating the need for Tosh equivalents for ex.), version using SMT to create a "super reg" ? Possibly a cheap alt. to PH regs etc.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 8, 2013 21:18:49 GMT
Chris Is the size of Greg's dual such a problem that we couldn't use another single mounted on top of the other ? Part of the success of the present JLH will be due to relatively close matching of the BC550C/BC560C. How are we going to readily check a tiny SMD transistor for HFE ? Alex
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 9, 2013 3:36:13 GMT
One thing that we haven't mentioned here is the role that the Current Limiter, and in particular the value of it's emitter resistor, plays in the time that the JLH reaches full output voltage. The JLH can provide a turn on delay which may be useful in some applications. The delay can range from a couple of hundred milliseconds with a 1 ohm emitter resistor ,to several seconds with a much larger value emitter resistor. Similar can be achieved by using a PW5 series resistor at the input of the non current limited version, and using a suitable relay connected at the output to short circuit the series input resistor when output voltage approaches that of the power supply itself.
Alex
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 9, 2013 12:10:02 GMT
Chris Is the size of Greg's dual such a problem that we couldn't use another single mounted on top of the other ? I was thinking along the lines of a three pin, direct replacement for standard regs. However, if the raison d'être is lost in the process then obviously not much point.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 9, 2013 12:25:12 GMT
Chris Is the size of Greg's dual such a problem that we couldn't use another single mounted on top of the other ? I was thinking along the lines of a three pin, direct replacement for standard regs. However, if the raison d'être is lost in the process then obviously not much point. Hi Chris For around $50 plus P and P and several months on a waiting list,you can get a Paul Hynes regulator like this, but they are fairly low current devices (150mA ? ) . Most shunt regs as seen in DIY Audio are usually much bigger with quite a few components. The PH regs that Syd gave me are fairly tall, although they have a 3 pin pinout, and in my case I had to bend it over to fit in the SC DAC 's 1U rack case. What the JLH has going for it, is that it is not voltage specific. In other words it can operate over a wide voltage range, usually without modifications., and with a fairly low standby current, unlike Salas and some others. Regards Alex
|
|
Will
Been here a while!
Ribena abuser!
Member since 2008
Posts: 2,164
|
Post by Will on Jul 10, 2013 20:30:04 GMT
Unless any more performance can be wrung from the existing JLH circuit, it seems that going SMT is the next step. Making into 3 pin reg would be nice. The main problems, as I see them, is that you'll need enough copper area to act as a heat sink if going above 500mA or so when using smt transistors. If the the new reg was limited to lower currents then it could be easier to implement. Also, as Alex mentions, the lack of decent smd caps. Perhaps going mixed technology and using thru-hole caps? Still this brings up size issues again.
A few years back Uriah over at diyAudio mentioned to me that he was doing a smt jlh for use in his LDR based amp that he was doing. Anybody still in contact with him, he may share his thoughts?
A different thought, but is there a better/different way to approach the CL at the start of the JLH - easier to set up for higher currents or perhaps a all in one PCB for extra high currents, similar to the power resisitor/relay setup that Alec uses to supply his Blu ray player? Just throwing a few ideas about.
Regarding Paul Hynes, the guy is currently several months behind schedule (6months+) as the poor guy has had a run of bad luck with PC's crashing and also health problems. More info on this at Audiocircle.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 11, 2013 22:46:26 GMT
Perhaps if the layout could be rejigged to having the through hole caps Will mentions on one side of the board, say laid length ways, end to end with the lead ends facing each other. For the 16v,or less, 2200uf caps that's not tooooo big. That might leave enough room the other side to include the heat sinking copper area?
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 11, 2013 23:13:14 GMT
Perhaps if the layout could be rejigged to having the through hole caps Will mentions on one side of the board, say laid length ways, end to end with the lead ends facing each other. For the 16v,or less, 2200uf caps that's not tooooo big. That might leave enough room the other side to include the heat sinking copper area? Hi Chris I am not saying that we can't make further running improvements, but personally, I like Greg's PCB for normal use, and Will's PCB for tight places ,or where only a single rail is needed. Presently we are working on ways to get the 100uF electro closer to the output terminals, as Will has done in his single, , and the addition of something like a 100nF near the input terminals. Greg has some other interesting ideas that he may discuss as he gets his initial thoughts more focussed in that area. Myself, I am curious to see how replacing R11 with a J511 CRD (4.7mA) would go with lower current versions using the CL. I could be very wrong, but I am wondering if the bias voltage is kept constant with varying load requirements, if this will give additional ripple rejection. Kind Regards Alex
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 11, 2013 23:32:26 GMT
I hear you Alex. It was only a sideways thought with the intention of getting a single JLH right in there, no leads.
I too am very curious as to Greg's thoughts on possible improvements to the current dual version.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 12, 2013 0:42:31 GMT
I hear you Alex. It was only a sideways thought with the intention of getting a single JLH right in there, no leads. I too am very curious as to Greg's thoughts on possible improvements to the current dual version. On my wish list, I would like to see a version where the CL is replaced by a bypass able, adjustable LM317/LM337 (or similar pin out) voltage regulator with a small heatsink, that would not only provide current limiting set by the type of LM317/LM337 used, but also be able to be used with a basic rectified and filtered input, or an existing higher voltage regulated supply.Regards Alex
|
|