Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Mar 19, 2013 11:54:35 GMT
Hi Alex, just realised I missed your last question out - setting the LTP offset was pretty easy, I got it to within 0.3mV or so, with the lid on after 30 minutes of being powered up to make sure it's thermally stable. I'll re-test it when I come to finishing the amp off for good. Re. tantalums in the PSU, i'm going to try out some regular electrolytics in this position as well and see how they affect the sound; i'll report back on here when I get round to it. Hi Sean I am pleased that you had no problems with getting the front end balance O.K. A few had problems, possibly due to diode forward voltage variations. You will need to leave the PSU Adjust Bypass cap until you have both modules working into your good speakers, as the difference is likely to be quite subtle. Regards Alex |
|
|
|
Post by sean on Mar 19, 2013 15:14:54 GMT
Yes it was all good thanks, I used the updated resistor values for colder climes which probably helped and it worked no problem at all. I'm going try swapping around the adjust caps when everything else is finished, i've got a few different ones that I can try out. That'll be in early April now, as i'm going away soon for a bit.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Mar 21, 2013 8:19:38 GMT
Hi y'all, Sean has just updated me with the news that both amps are complete and working fine with his test speakers. He now is taking a break somewhere until 1st April when he plans to drop them into his main system and 'fine tune' balance and things technical  . Dave. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Mar 21, 2013 8:41:45 GMT
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Mar 21, 2013 9:53:55 GMT
Hi y'all, Sean has just updated me with the news that both amps are complete and working fine with his test speakers. He now is taking a break somewhere until 1st April when he plans to drop them into his main system and 'fine tune' balance and things technical . Dave. Dave All that should need doing, is a final fine adjustment of the front end balance and bias adjustments, as well as replacement of the tantalums in the PSU with 10uF electrolytic capacitors. Perhaps you should consider Shaun's suggestion, as the 2nd GB for the preamp/HA will soon close, and is highly unlikely to be repeated again. Regards Alex. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Mar 21, 2013 10:26:41 GMT
Hi Dave,
I'm delighted, you will now have a great amp, and Sean's build is very impressive as is Jon's makes mine look positively "Rustic"
X3 echoing Shaun and Alex
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Mar 21, 2013 12:00:28 GMT
Hi y'all, Sean has just updated me with the news that both amps are complete and working fine with his test speakers. He now is taking a break somewhere until 1st April when he plans to drop them into his main system and 'fine tune' balance and things technical . Dave. Dave All that should need doing, is a final fine adjustment of the front end balance and bias adjustments, as well as replacement of the tantalums in the PSU with 10uF electrolytic capacitors. Perhaps you should consider Shaun's suggestion, as the 2nd GB for the preamp/HA will soon close, and is highly unlikely to be repeated again. Regards Alex. Hi Dave Alex's matching PRE/HA is a superb match for his PA and also makes a really good HA. It matches my K701 superbly and brings them to life (love the bass) in a way that no others I’ve tried get anywhere near including my TP Ventus which is no slouch. If you check out the thread that I’ve linked you'll probable notice that quite a few builders are on their second build of the HA/PRE. Enough said. The pre really does do the business and out performs any of my others with the class A PA by quite some distance. It’s not such an expensive build I had 2x18V 80VA transformers from farnell for the crazy price of £10 each which are a little over speck but work very nicely. The main expense apart from that is case work which is much cheaper than the PA. I have my build in a matching Modushop HIFI2000 2U pesante which looks cool (with the lid on   ). As Alex said The opportunity to have this wonderful HA/PRE won't be coming up too often so well worth thinking on. Just out o9f interest What speakers have you? Take care |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Mar 21, 2013 13:20:07 GMT
Hi Shaun, My speakers are Mission 774s to which I have added some very decent ribbon super-tweeters (with crossovers etc) with much help from forum friends. I cannot hear anything coming from them but I can say that I prefer listening to music with them connected rather than with them out of circuit. Having said that, my ancient old ears have lost both ends of the hearing spectrum due to 'Industry related hearing loss'. At the moment my amps are two Mini-Ts (one driving each channel) with a third Mini-T driving both super-tweeters - might be unorthodox but it works for me . Cheers, Dave |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Mar 21, 2013 14:38:31 GMT
Hi Dave ''might be unorthodox but it works for me '' all that matters Dave  I remember the 774’s quite well as a friend of mine owned a pair and loved them. From what I remember they sounded pretty good if a little on the dry side bass wise but @ 89DB 1W and nominal 8ohm should give the Class A no problems what so ever. I think that your ears are in for a tread Super T’s Which ones are you using? I keep meaning to have a play with some of those. Take care |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Mar 21, 2013 16:31:59 GMT
Shaun,
When Sean delivers my Class A I've got 3 Mini-T's that you can play with indefinitely.
Dave.
|
|
|
|
Post by sean on Apr 9, 2013 21:20:01 GMT
Hi all, as Dave's already reported I have the amps both working perfectly well into the test speakers (which are 8 Ohm), but when I tried out one of the amps on my Dynaudio 52s (which are 4 Ohm) it was fine for the first few seconds of music - then it started to oscillate at around 5Hz, you could see the bass driver moving back and forth by about a centimetre (so a large amplitude oscillation, looked quite smooth i'd guess it was roughly sinusoidal), there was also a low thudding which I assume was something like a periodic 50/100Hz oscillation as well. I turned the amp off right away, tried it back on the 8 Ohm test speaker and everything was perfectly fine again, had to playing for the last hour on the test speaker no problem.
Anyone have any idea what the problem might be, and if it's fixable? Dave has 8 Ohm speakers so i'm guessing that they should be fine too, but of course I want to get the amps fully working on 4 and 8 Ohm speakers.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 9, 2013 22:20:55 GMT
Hi all, as Dave's already reported I have the amps both working perfectly well into the test speakers (which are 8 Ohm), but when I tried out one of the amps on my Dynaudio 52s (which are 4 Ohm) it was fine for the first few seconds of music - then it started to oscillate at around 5Hz, you could see the bass driver moving back and forth by about a centimetre (so a large amplitude oscillation, looked quite smooth i'd guess it was roughly sinusoidal), there was also a low thudding which I assume was something like a periodic 50/100Hz oscillation as well. I turned the amp off right away, tried it back on the 8 Ohm test speaker and everything was perfectly fine again, had to playing for the last hour on the test speaker no problem. Anyone have any idea what the problem might be, and if it's fixable? Dave has 8 Ohm speakers so i'm guessing that they should be fine too, but of course I want to get the amps fully working on 4 and 8 Ohm speakers. Hi Sean That sounds like a PSU or JLH issue. I have never seen that kind of problem before over many builds of this P.A. over the years. Is this happening on both channels ? If it's only one channel, I would start by doing some measurements around the JLH s for that channel. You could also try bypassing the JLHs for that channel to see if the problem goes away. The main things to check are the voltage drops of the current limiter emitter resistors, the collector voltages of the BC550C and BC560C which should be fairly close, and the approx. 1.8V at the BC6xx end of the 120K resistor. Compare those voltages against that of the good channel. You will also need to check the output voltages of the voltage regulators. What type of regulators are you using ? Don't forget that you are no longer in Class A into 4 ohm speakers after 7.5W , so the PSU becomes more critical. With suitable power supplies the original SC design was capable of up to 30W into 4 ohms. I have also had a previous version working well into a friend's big 4 ohm OhmWalshe speakers, although it wasn't fully DC coupled back then.. Regards Alex |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 9, 2013 22:56:16 GMT
Hi Sean
I’ve had my amp plugged into a lowish six ohm speaker and also 4 ohm stunt jobs with no problemo. I suspect that Alex is right and it could be that the current limiter/JLH/PSU is the first place to look.
The Dynaudio speakers are a tough drive and require muscle to get them singing from what I remember but that should not be a problem with this amp.
Which regs have you used in the main PSU?
What CL do you have set on the JLH? Are those babies getting super-hot at switch on? My ones get temporarily warm but no more.
And errrrr just remind me again about the VA of those transformers that you've used.
As you say they are fine into 8 but maybe there is a problem somewhere that would be easier to sort out now.
Let us know how it goes.
Take care
|
|
|
|
Post by sean on Apr 9, 2013 23:53:49 GMT
Thanks for the quick reply. The fault is on both channels, so it is probably down to an implementation fault rather than individual component failures.
The PSU boards were built as per the BOM, and as per your advice Alex are now using 10uF electrolytics on the adjust pin and not tantalums. The regulators are the LM1086 (317 equivalent, up to 1.5A output) and the 337. Main caps are Nichicon KG 10,000uF 35V. The output voltages on each amp are +/-22.1V into the amp board, and the JLH boards each feed +/-22.0V back to the front end of the amp. It stays rock steady with either the 8 ohm speaker or with no speaker connected. Bias is set to have 260mV across one of the 1R resistors, and the offset on the front end is less than 0.5mV
Transformers are 120VA each (remember, i've built the amps as monoblocks), and the CL on the JLH is set as per Alex's recommendation, 1R and 1K8 resistors and the CL transistors each have a small heatsink fitted.
At switch-on with test speakers fitted, the JLH transistors don't get particularly warm at all - hardly above ambient. I'd rather avoid connecting the Dynaudios back up to try measuring voltages and temperatures as it takes a short while to do, all that cone movement can't be good for them and I fear for the tweeters too :S
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 10, 2013 0:09:11 GMT
LM1086 (317 equivalent, up to 1.5A output) and the 337 A quick ask round anyone else using these regs? i have the 3A units fitted in my build because 1.5A seemed a little marginal in that spot. I think that Alan also used the 3A regs without issue and that's the only stand out difference as far as i can see (which is not very far) between builds. I may be wrong so some opinions will help thinking on this. I’m wondering if the extra load is maybe somehow causing the reg's thermal shutdown to kick in and out. just a wild guess but it could be useful to measure the raw voltage with a 4ohm dummy hooked up just to rule that out. ''I'd rather avoid connecting the Dynaudios back up to try measuring voltages and temperatures as it takes a short while to do, all that cone movement can't be good for them and I fear for the tweeters too :S'' a 4ohm dummy load would be a good bet to save tears at switch on.  take care |
|
|
|
Post by sean on Apr 10, 2013 1:23:40 GMT
Doh, yes of course a 4 Ohm dummy load would do just fine. I've got a 4R7 50 Watt power resistor that'd do the job, i'll try it out tomorrow. Re. the regs, the BOM did indicate that 317/337 were OK to use, and they're certainly well heatsinked. I'll get an osciloscope onto the power rails to see if anything is oscillating with the dummy load whilst i'm at it.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 10, 2013 1:56:42 GMT
Sean My existing build uses a pair of the original S.C. PSUs with a pass transistor. My new PCBs are using 3A regulators, which I have recommended for this new build, especially as many are using the higher VA transformers and decent heatsinks. I am using a 21-0-21 toroidal , and we couldn't use the original SC PSUs with the more commonly available 18-0-18 toroidals due to lack of voltage overhead. The B.O.M. wasn't prepared by me, but I should have checked it more closely, especially as a few are using 4 ohm speakers .  This also gives the opportunity to increase the bias current a little, as well as the possibility of increasing the supply rails another couple of volts if desired. Kind Regards Alex P.S. It is possible that if the LM317 and LM337 are starting to lose control near their maximum ratings, that this may cause the JLHs to "hunt" a bit. Is the bias current no more than 1A when fully cased up ? Do the CLs in the JLH have 1.8K base resistors ? |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 10, 2013 9:09:27 GMT
Top Man Looking at the high quality work that you've done I’m pretty sure it's going to be something small. My 3A regs are well heatsinked but still get warm especially at switch on so that could be the cause of your builds instability. I have a dual mono build BTW so two transformers/PSU and still went with 3A regs just for long term reliability. I used the 3A TO-220 LT1033/LT1085 (Farnell have them still I think) in my build which seemed to give a more reliable overhead IMHO. Those 4.7R resistors that you’ve mentioned may l need to be heatsinked. Those Ali cased resistors are rated at whatever watts but then in the very small print it usually gives a lower rating without extra heat sinking. Yup I know it’s teaching my grandmother to suck eggs but others may need to do the same later so just for general info. Take care |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 10, 2013 14:24:03 GMT
To save one lazy sod trawling the thread, again, could someone remind me of the best arrangement for the large caps on the JLH boards (C4A & B on the blue singles) specific to this application  e.g. one x 4700 10v & one x 2200 10v etc. Cheers me dears.... |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 10, 2013 16:10:05 GMT
To save one lazy sod trawling the thread, again, could someone remind me of the best arrangement for the large caps on the JLH boards (C4A & B on the blue singles) specific to this application e.g. one x 4700 10v & one x 2200 10v etc. Cheers me dears.... Hi Chris I’ve used 2x2200 10V for my build with CL fitted. but lets see what the boss thinks  Take care |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 10, 2013 18:05:47 GMT
Hi Shaun,
Now you say that, my memory seems to be agreeing. The other options were down to non CL versions and lower voltages e.g. 5v.
I can't get low ESR 2200uf 10v locally so will to order again.
Cheers
ps
I also tried to get some 1N4148 locally and was given some old ITT 600. The 4148 cross refs OK to replace the 600 but is the opposite way around also OK?
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 10, 2013 21:24:24 GMT
Hi Sean,
Shaun said.................
Shaun's Traffo/ PSU configuration is same as mine Two 160VA into two PSU/JLH running LT1033/LT1085 I have two working builds of the 15Watt with this config for the power supply.
Regards,
Alan
PS I only used the LT1086 regs with my HA/PREAMP build.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 10, 2013 21:31:42 GMT
To save one lazy sod trawling the thread, again, could someone remind me of the best arrangement for the large caps on the JLH boards (C4A & B on the blue singles) specific to this application e.g. one x 4700 10v & one x 2200 10v etc. Cheers me dears.... Hi Chris I’ve used 2x2200 10V for my build with CL fitted. but lets see what the boss thinks Take care Shaun and Chris Where you are using several JLHs in several different pieces of equipment, it is best to use a 2,200uF 10V low ESR and a 2,200uF 16V normal electro in parallel where the CL section is used. This is due to the simplicity of the CL itself as the 4,700uFs tend to cause it to hunt, or even on occasion overheat as the electros form. With my original 3 uploaded .wav files for M.C. he also thought as I did, that the use of 2 low ESR types in parallel resulted in a smidgin too much HF detail. I had left it that way as the guys here this end thought it sounded O.K. when I asked them. The very recent ones I sent M.C. were apparently the best sounding so far, and used one of each type in parallel. If your source is slightly HF lacking you may prefer 2 low ESR in parallel. However, there isn't much in it. Regards Alex |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 11, 2013 6:43:37 GMT
Oops, thanks Alex.
I've 4700s in the aclass pre/ha.
Better wip 'em out!
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 11, 2013 6:52:56 GMT
Oops, thanks Alex. I've 4700s in the aclass pre/ha. Better wip 'em out! Hi Chris What you could try with a bench PSU if you have one with + and - outputs, is to burn them in at quite a bit lower than + and - 20 V, and check that there is no overheating of the Toshiba devices. After say 48 hours with them, try a dummy load and check with your DMM that the output voltage doesn't cycle up and down by several mV. You may be able to use them this way, but I haven't tried this myself yet. There is a possibility to perform a little better again, if they remain stable. If they don't, just fit the 2,200uFs. Regards Alex |
|
.
).
