Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Feb 19, 2013 21:12:45 GMT
The only real way forward is to be able to compare what you have against very good gear from others. We never seem to stop learning or improving things when theory suggests they should already be damn near a straight wire with gain. already. Some even learn not to use 3mm thick double sided PCBs !  There may even be further benefits from going a little higher than + and - 20V supply rails , just as Robert Dwyer said about going from + and - 15V rails.. Seems that the JLHs when using the CL section aren't as good in practice as they are in theory., and there is some minor interaction between channels due to that.The original JLH Current Limiter circuit has always been a weak point of his design, and could undoubtedly be further improved. The problem is that the HA/preamp doesn't sound right when the CL section is bypassed. Regards Alex |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Feb 20, 2013 0:51:48 GMT
Chong and Will
There is nothing stopping you from using the PCB designed for the 15W Class A , as it is the same basic circuitry with a few different components and values. You would need to bend the legs of the output devices to suit the different pitch, and the heatsinks would overhang. The Zobel filter would be omitted and bypassed. There is nothing too involved there.
Alex
|
|
|
|
Post by gommer on Feb 20, 2013 9:13:35 GMT
Once I get the nod form Marc, I'll send over the revised pcb (and anything else wanted) and the pcb's can be on the way It's very heartening that another couple of AK Class A's are going to be made  I ain't nodding yet. Getting my head round everything and trying to decide how far i will take this. I'm considering the following things: HA PCB: sure LS devices: sure Other active parts for HA: maybe Caps and heatsink for HA: maybe resistors: no way PS PCB: maybe (dual LM317 version i guess, although the beefier one also makes sense) Transistor matching: maybe Basically, as I'm starting from scratch with this, I might as well order everything in bulk, there are not that many participants that it will become a logistic nightmare. And matching makes sense, as it would justify to buy 100 or even 250 transistors in order to easily pick matched pairs or quads. I'll decide one of these days what i can commit to, as i have to keep family life in mind too. Cheers, Marc |
|
XTRProf
Fully Modded
Pssst ! Got any spare capacitors ?
Posts: 5,689
|
Post by XTRProf on Feb 20, 2013 11:25:42 GMT
Arghhh, seems I wouldn't be on the list as no dual PSs. What's the best way to modify Will's PCBs to dual PS? I mean where to cut trace is best for least defects like instability and others.
|
|
pingu
Been here a while! 
Reallising what has been taken on with the GB
Posts: 332
|
Post by pingu on Feb 20, 2013 14:22:31 GMT
Once I get the nod form Marc, I'll send over the revised pcb (and anything else wanted) and the pcb's can be on the way It's very heartening that another couple of AK Class A's are going to be made I ain't nodding yet. Getting my head round everything and trying to decide how far i will take this. I'm considering the following things: HA PCB: sure LS devices: sure Other active parts for HA: maybe Caps and heatsink for HA: maybe resistors: no way PS PCB: maybe (dual LM317 version i guess, although the beefier one also makes sense) Transistor matching: maybe Basically, as I'm starting from scratch with this, I might as well order everything in bulk, there are not that many participants that it will become a logistic nightmare. And matching makes sense, as it would justify to buy 100 or even 250 transistors in order to easily pick matched pairs or quads. I'll decide one of these days what i can commit to, as i have to keep family life in mind too. Cheers, Marc Hi, Pls count me in for 5 PSUs boardsto go with the anp boards. I'll also join in the GB for the other bits as well seems to make sense to bulk buy. All the best Martin |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Feb 20, 2013 18:19:43 GMT
Martin reallising what he has taken on with the GB(s)  PS If there is a GB for Jaycar stuffs going on I might like some speaker protection kits! |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Feb 20, 2013 20:54:52 GMT
|
|
Will
Been here a while!
Ribena abuser!
Member since 2008
Posts: 2,164
|
Post by Will on Feb 20, 2013 20:58:58 GMT
Arghhh, seems I wouldn't be on the list as no dual PSs. What's the best way to modify Will's PCBs to dual PS? I mean where to cut trace is best for least defects like instability and others. Hi Chong, I've sent you an email with the details, save muddying the waters in this thread  To be quite honest, I'm in shock that I've had the same idea as you!   Does this mean I get to be as handsome as you are in that photo?  |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Feb 20, 2013 21:29:00 GMT
Perhaps you have both been eating those contaminated Lasagna s ? Lotsa luck taming 2 lots of JLHs working from the same regulated PSU. Of course, there is always the possibilty that there will be no problems whatsoever and with suitable earthing you may even notice a tiny improvement. ( try proving it with a DBT !  Just remember though that you are unlikely to measure an improvement in S/'N or other parameters without suitable gear.Even then..... Even using a CRO with a low noise 10 X preamp, I could only see <1/3 of a division of SMPS residual apparently coming from my big LCD TV next to it. Regards Alex |
|
XTRProf
Fully Modded
Pssst ! Got any spare capacitors ?
Posts: 5,689
|
Post by XTRProf on Feb 21, 2013 2:33:52 GMT
I've sent you an email with the details, save muddying the waters in this thread Thanks.  To be quite honest, I'm in shock that I've had the same idea as you! Er, you having lessons with a top Audio Zen master? Does this mean I get to be as handsome as you are in that photo? Oh dear, CJ is the culprit. Now disease has spread internationally like SAR. Danger, danger, danger ........... All personnels please stay in your home. It's plague curfew time. The audio disease has spread. I must report this to WHO. Arghhhh, not rocking with the British group. Help! Oh dear, now Beatles .............. Heh, heh, heh, maybe having an overdose of OMD new songs, GI Joe and Bond in combination as the actions are strongly infected from these in combination.  Send in Team 6 to be sure as they will leave nobody (except Women and Children) alive to spread the disease even when you have not fired any single shot! No POWs ............ Only body bags ............ |
|
pingu
Been here a while!
Reallising what has been taken on with the GB
Posts: 332
|
Post by pingu on Feb 21, 2013 9:36:01 GMT
LOL
Alex and CJ
I love it how can I get this as my avatar picture thingy, it will not copy.
2xHA's on the cards one for me one fro son in law. (when all mods buying is settled) Several ripple eaters now finished and awaiting installing. Projects crack on.
Anyone going to Bristol BTW.( Bit off subject ....sorry )
Martin
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Feb 21, 2013 9:41:00 GMT
LOL Alex and CJ I love it how can I get this as my avatar picture thingy, it will not copy. 2xHA's on the cards one for me one fro son in law. (when all mods buying is settled) Several ripple eaters now finished and awaiting installing. Projects crack on. Anyone going to Bristol BTW.( Bit off subject ....sorry ) Martin Martin Have the REs had some time on them, on the bench ? There's nothing worse than completing a project. then having to wait for a PSU to come up to speed before you can hear it properly. Regards Alex |
|
pingu
Been here a while!
Reallising what has been taken on with the GB
Posts: 332
|
Post by pingu on Feb 21, 2013 10:03:12 GMT
Hi Alex They are going on the bench tomorrow or Sun. i have bypassed the c/l circuit as you suggested with the 4,700 uF caps but upped the voltage to 35v so will cover 12 or 15 volt psu's. Hope that is OK and what was needed quote; I recommend using a 4,700uF 10V low ESR electro and a normal 4,700uF 16V in parallel on both halves of the PCB. Thanks Martin ......... still laughing over the Pingu |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Feb 21, 2013 10:24:26 GMT
Hi Alex They are going on the bench tomorrow or Sun. i have bypassed the c/l circuit as you suggested with the 4,700 uF caps but upped the voltage to 35v so will cover 12 or 15 volt psu's. Hope that is OK and what was needed quote; I recommend using a 4,700uF 10V low ESR electro and a normal 4,700uF 16V in parallel on both halves of the PCB. Thanks Martin ......... still laughing over the Pingu Martin Those 4,700uF caps should be lower voltage types. as the voltage across them when working is only around 600 Millivolts. We could have even used 6.3 volt types, but these have a much higher ripple rating which is better for longevity. The 10uF and the 100uF electros can be 35V types if desired. Regards Alex |
|
pingu
Been here a while!
Reallising what has been taken on with the GB
Posts: 332
|
Post by pingu on Feb 21, 2013 10:33:17 GMT
Hi Alex They are going on the bench tomorrow or Sun. i have bypassed the c/l circuit as you suggested with the 4,700 uF caps but upped the voltage to 35v so will cover 12 or 15 volt psu's. Hope that is OK and what was needed quote; I recommend using a 4,700uF 10V low ESR electro and a normal 4,700uF 16V in parallel on both halves of the PCB. Thanks Martin ......... still laughing over the Pingu Martin Those 4,700uF caps should be lower voltage types. as the voltage across them when working is only around 600 Millivolts. We could have even used 6.3 volt types, but these have a much higher ripple rating which is better for longevity. The 10uF and the 100uF electros can be 35V types if desired. Regards Alex[/quote Hi Alex All noted Alex and will be done before Sunday and powering up on bench. As you can see sorted Avatar out now Thanks again Martin |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Feb 21, 2013 11:08:49 GMT
''i have bypassed the c/l circuit as you suggested with the 4,700 uF caps but upped the voltage to 35v so will cover 12 or 15 volt psu's''
Hi Martin
Just to be clear. Does the above mean that you've used 4700uf caps to ''bypass'' the current limiter?
If so I have a feeling that is not quite what Alex intended (but I may be wrong but i've not seen that done before) most of us just use a plain piece of wire to bypass with.
Maybe a picture would help.
Take care
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Feb 21, 2013 12:37:30 GMT
Hi Shaun
I am about to pull the plug for the night,so feel free to take over.
Kind regards
Alex
|
|
pingu
Been here a while!
Reallising what has been taken on with the GB
Posts: 332
|
Post by pingu on Feb 21, 2013 12:42:59 GMT
''i have bypassed the c/l circuit as you suggested with the 4,700 uF caps but upped the voltage to 35v so will cover 12 or 15 volt psu's'' Hi Martin Just to be clear. Does the above mean that you've used 4700uf caps to ''bypass'' the current limiter? If so I have a feeling that is not quite what Alex intended (but I may be wrong but i've not seen that done before) most of us just use a plain piece of wire to bypass with. Maybe a picture would help. Take care Hi Shaun, This was from the thread on JLH RE group buy on DIY a week or so ago. Yes I did not use the c/l circuit in the JLH RE at all and inputted the power via the two caps in parallel straight to R1 and R2. So input from psu to JLH RE then the 4.700uF (+/- pins) caps then to R1 R2. Have I miss-interpreted something? Alex was talking about not using the c/l circuit if I was running a regulated psu and the RE not getting working correctly if I was running more than 100ma. I wanted up to 500ma so opted for the non c/l choice. Maybe I have it wrong? Martin |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Feb 21, 2013 14:04:51 GMT
''i have bypassed the c/l circuit as you suggested with the 4,700 uF caps but upped the voltage to 35v so will cover 12 or 15 volt psu's'' Hi Martin Just to be clear. Does the above mean that you've used 4700uf caps to ''bypass'' the current limiter? If so I have a feeling that is not quite what Alex intended (but I may be wrong but i've not seen that done before) most of us just use a plain piece of wire to bypass with. Maybe a picture would help. Take care Hi Shaun, This was from the thread on JLH RE group buy on DIY a week or so ago. Yes I did not use the c/l circuit in the JLH RE at all and inputted the power via the two caps in parallel straight to R1 and R2. So input from psu to JLH RE then the 4.700uF (+/- pins) caps then to R1 R2. Have I miss-interpreted something? Alex was talking about not using the c/l circuit if I was running a regulated psu and the RE not getting working correctly if I was running more than 100ma. I wanted up to 500ma so opted for the non c/l choice. Maybe I have it wrong? Martin Hi Martin mmmmmmmmmmm something does not sound quite right with the above description. You should not have 4700uf caps on the input of the JLH as all that is needed to bypass the CL is a wire link. So I think we may have crossed wires here. TBH from your description it’s hard to work out exactly what you’ve done but you should only need to use caps where they have been marked on the board. When we say bypass the CL it’s not the same as bypassing to ground with a cap it just means disabling that part of the circuit. So what you should have is If you are using Greggs lovely boards is R12 removed (should make life easier) Then link from the R12’s hole on the outside of the board to that little hole just in front of it (unmarked and between R12 and R1) That bypasses the CL part of the circuit. No need for any extra caps in that spot as it will degrade the performance. A picture would help quite a bit Hope that helps. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Feb 21, 2013 14:27:11 GMT
Hi Martin here is a picture it's from Gregs site ( hope that's ok with you Greg) the link location is marked in red  take care |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Feb 21, 2013 14:59:38 GMT
Well I think Shaun has covered it but here's an actual board marked for no CL....  |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Feb 21, 2013 15:18:34 GMT
Hi Chris Thanks You always seem to make things much more obvious than I manage to. Good work Take care |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Feb 22, 2013 3:59:42 GMT
Just a note for clarity. If you decide not to use the current limiter, you can use output terminals for the input wires as well. In theory, this should work slightly better. I use my JLHs like this and it means you don't need the links installed. I'm not sure if this mod has been given the full nod of approval yet, so feedback will be appreciated.
BTW: I started doing this on the earlier versions of the boards that had thinner copper and when used in high power applications. From memory, Alex has measured a voltage drop from the input to the output. On the newer PCB I doubled the copper thickness as well as making the tracks wider so any benefit will be less due to the reduced resistance.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Feb 22, 2013 11:01:02 GMT
Hi Greg
I’m almost finished building up my first pair so I’ll post a few pictures of what I have over the next few days.
Yes
You are right
the + and - voltages can be tapped in after the link and that makes sense to me but using terminal blocks does make it easier for the in out in out board checking.
If I may just add
These boards really are superb quality and a joy to use.
Great work from all involved with the design, construction and distribution.
Take care
|
|
pingu
Been here a while!
Reallising what has been taken on with the GB
Posts: 332
|
Post by pingu on Feb 22, 2013 12:42:33 GMT
Thanks all, Very grateful Advice much appreciated. Couple of boards going bench Sunday (I hope). Then just waiting for the HA bits and bobs with the boards. All the best to everyone Martin |
|
Send in Team 6 to be sure as they will leave nobody (except Women and Children) alive to spread the disease even when you have not fired any single shot! No POWs ............