Luckiestmanalive Posted June 26, 2022 Author Posted June 26, 2022 (edited) Thanks, X - that makes sense. I don't think I'll mess with the power connector. The hum isn't loud enough to bother me even when I'm not playing music. What I am trying to do now is to find an alternative speaker connector to the stock screws. I found cheap connectors at Jaycar where the thread is the same as the original screw: https://www.jaycar.com.au/red-deluxe-binding-post/p/PT0453?pos=11&queryId=0df678907a8486d0fa84c3f90ce4c7fe&sort=relevance I bought four and they are working fine. However, they stick out too far so aren't a forever solution. Also, the post you link to above suggests you found a good substitute for the 6240G? It looks like I have original input tubes - Mitsubishi Electric 6240Gs and NEC 6AQ8. Should I look to change these and what would you recommend? Edited June 26, 2022 by Luckiestmanalive
xlr8or Posted June 27, 2022 Posted June 27, 2022 (edited) Try changing the 6AQ8 to a vintage ECC85. They are quite cheap to acquire. For the 6240G the 6CG7 can be used, but I prefer to use a tube socket adapter instead to retrofit 12BH7's and E80CC's. Edited June 28, 2022 by xlr8or 2
Luckiestmanalive Posted June 28, 2022 Author Posted June 28, 2022 Thanks, X. Now the fun begins - what to choose, what to choose!? I see some of the sellers test and post output figures for the tubes they are selling, for instance, the GM, la1 and la2 for the ECC85. I'm not sure what I'm looking for? Being a dual triode, am I looking for la1 and la2 figures to be the same and higher than 10ma? And a high GM is best?
xlr8or Posted June 28, 2022 Posted June 28, 2022 I would recommend getting this one: https://www.ebay.com.au/itm/274986913446 1
Luckiestmanalive Posted June 28, 2022 Author Posted June 28, 2022 What about a pair of 6N1Ps or a 6GU7s in place of the 6240Gs? The 6GU7 is down on gain (17 v 35x) but seems otherwise to be a good match and the 6N1P is very close except on transconductance (7500 v 3500umhos).
xlr8or Posted June 28, 2022 Posted June 28, 2022 (edited) 21 minutes ago, Luckiestmanalive said: What about a pair of 6N1Ps or a 6GU7s in place of the 6240Gs? The 6GU7 is down on gain (17 v 35x) but seems otherwise to be a good match and the 6N1P is very close except on transconductance (7500 v 3500umhos). I think you'll find their grid voltages, transconductance and plate loads are very different to the 6240G. Edit: Added E80CC datasheet info below. I'm using it with a tube adapter to allow conversion from a 6CG7 to 12AU7. Edited June 28, 2022 by xlr8or 1
Luckiestmanalive Posted June 28, 2022 Author Posted June 28, 2022 I've picked up alternatives for the 6240G from various threads on forums talking about substitutes the owners have supposedly tried. I've constructed a table of the alternatives and their characteristics from datasheets I've found. None of the alternatives are perfect... Description Units 6240G 6N1P-EV 6GU7 6FQ7 6CG7 Vf Heater voltage V 6.3 6.3 6.3 6.3 6.3 … Heater current mA 600 600 600 600 600 μ Amplification factor mu 35 35 17 20 20 Ia Plate current mA 7 7.5 11.5 9 9 S Mutual conductance mA/V 4.5 Pa Plate dissipation W 3 2.2 3 4 2.75 Va max Max plate voltage V 800 300 330 330 330 Vp Plate voltage V 250 250 250 250 250 Vg Grid voltage V -5 -1.5 -10.5 -8 -8 Gm Transconductance Umhos 3500 7500 3100 2600 2600 … Plate resistance ohms 4400 5500 7700 7700
Luckiestmanalive Posted June 29, 2022 Author Posted June 29, 2022 (edited) Apologies, but I may have some terms wrong and some are missing so feel free to correct me and I will edit the table. One figure that stands out in the 6240G data is the Max plate voltage - 800V is so much higher than the others. I guess, being a max it doesn't matter much. Edited June 29, 2022 by Luckiestmanalive
Luckiestmanalive Posted June 29, 2022 Author Posted June 29, 2022 I don't mind experimenting with a tube (part of the fun, isn't it?) as long as it isn't dangerous or likely to harm the amp.
muon* Posted June 29, 2022 Posted June 29, 2022 Using the wrong tube (even if it does work and doesn't damage the amp) can change the operating points in that part of the circuit, so only use ones that are compatible. 1
Luckiestmanalive Posted June 29, 2022 Author Posted June 29, 2022 (edited) Cheers, M. I am worried about that - it is why I am researching the best 9-pin alternatives to the original 6240G tube, which hasn't been available for decades. It appears even Luxman's recommended alternatives, the 6FQ7 or 6CG7, aren't a perfect match and I've found the other ones in online posts by Luxman owners who have used them to replace the original 6240G tube. Edited June 29, 2022 by Luckiestmanalive
muon* Posted June 29, 2022 Posted June 29, 2022 (edited) Follow Kirks recommendations, he knows tubes better than anyone here. Edit: the internet is full of folk suggesting tubes as being equivalents when they are in some instances not. Edited June 29, 2022 by muon* 3 1
xlr8or Posted June 29, 2022 Posted June 29, 2022 This one just got listed. It's a tu0 version as opposed to the tu1 version I posted earlier: https://www.ebay.com.au/itm/155055569105
Luckiestmanalive Posted June 29, 2022 Author Posted June 29, 2022 Thanks, X and M. I appreciate your advice. I'll let you know if I pull the trigger - I'm pretty happy with the music it makes right now and the thrill of the hunt for alternative tubes is proving to be fun in itself! What makes the Philips ECC85 tube catch your eye, X? Neither are NOS and appear to be well-used if the test results are anything to go by?
xlr8or Posted June 30, 2022 Posted June 30, 2022 (edited) Firstly, you don't need new and matched sections for the ECC85 input tube. UOS works just fine and will probably outlive you, me and everyone else reading the thread here. Secondly, vinatge Philips Heerlen production is in demand for giving some of the best tonal characteristics one can find in tubes. The tu0 version also represents the first of its kind. It was over-engineered for the highest standard in telephony applications, hand-assembled by lovely ladies and internal construction features would be representative of using the best techniques and materials that wouldn't even be considered today for manufacturing tubes due to high costs and unavailability. Just to give an example, Heerlen were using gold plated grid posts in 9-pin mini construction during the time the 2 tubes above were manufactured. Edited June 30, 2022 by xlr8or 3
Luckiestmanalive Posted June 30, 2022 Author Posted June 30, 2022 (edited) You've convinced me! I pushed the button on the tu1. And now my quad socket bias measurement doo-dad has arrived! So tomorrow I will take the KT88s out, insert the four adapters, reinsert the KT88s on top, switch the amp on and take measurements after 15 min, 30 min, one hour, and two hours. Edited July 12, 2022 by Luckiestmanalive Slip of keystroke 2 - I bought the tu1 not the tu0! 1
Luckiestmanalive Posted June 30, 2022 Author Posted June 30, 2022 Oh, btw, Kirk, I checked each of the changes for ultralinear and triode rewiring you provided and the adjustments made to my A3600 seem to be consistent with the list for ultralinear connection. The giveaway is the extra wires between pins 2 and 5 of the primary winding side of the output transformers and pin 4 of each of the output tube sockets. 1
xlr8or Posted June 30, 2022 Posted June 30, 2022 This is what my one is wired for: Also, I forgot to mention, if you want to use KT120's in there the cage needs to come off, due to the internal height clearance restriction meeting a KT88 or anything less in height. 1
Luckiestmanalive Posted June 30, 2022 Author Posted June 30, 2022 Thanks, mate! Yes, I compared your wiring with mine in the pics we both put up on the first page of my thread. I've read there are differences in the performance and character of the amp between Triode and Ultralinear wiring. In general, Triode mode is more intimate, relaxed and provides greater instrument separation while Ultralinear mode is more powerful, dynamic and presents a larger soundstage. And I have been pondering options for the cage if I were to spring for a set of KT120s... Measuring the output of the KT88s is postponed until I replace the fuse in my multi meter (I forgot it was blown). 1
muon* Posted July 1, 2022 Posted July 1, 2022 On 26/06/2022 at 6:45 PM, Luckiestmanalive said: Thanks, X - that makes sense. I don't think I'll mess with the power connector. The hum isn't loud enough to bother me even when I'm not playing music. What I am trying to do now is to find an alternative speaker connector to the stock screws. I found cheap connectors at Jaycar where the thread is the same as the original screw: https://www.jaycar.com.au/red-deluxe-binding-post/p/PT0453?pos=11&queryId=0df678907a8486d0fa84c3f90ce4c7fe&sort=relevance I bought four and they are working fine. However, they stick out too far so aren't a forever solution. Also, the post you link to above suggests you found a good substitute for the 6240G? It looks like I have original input tubes - Mitsubishi Electric 6240Gs and NEC 6AQ8. Should I look to change these and what would you recommend? https://www.audiophonics.fr/en/binding-posts/eizz-ez-301-gold-plated-ofc-copper-binding-posts-o16mm-x-61mm-pair-p-9899.html
Luckiestmanalive Posted July 1, 2022 Author Posted July 1, 2022 1 hour ago, muon* said: https://www.audiophonics.fr/en/binding-posts/eizz-ez-301-gold-plated-ofc-copper-binding-posts-o16mm-x-61mm-pair-p-9899.html Thanks, M - I love the look of these! However, the rectangular window in the steel on the back of the amp has a height of 12mm at a pinch. Also, the width of the window is only 50mm, which has to fit four connections, unless I want to miss the 16ohm connection out and mess with the original look. I'm mulling over my options...
Luckiestmanalive Posted July 1, 2022 Author Posted July 1, 2022 (edited) I have replaced the blown fuse in my multimeter and taken some readings from the output tubes. This is obviously my first time taking readings at all and first time with this doodad. The readings took a while to stop rising (maybe about 20-30 min). Also, switching the device from reading one tube to another could be having an effect on the readings because after I switched from one to another (something I had to do slowly because the speakers would make an audible but reasonably soft thump) the reading would be roughly along these lines (mA): A to D (V4 to V7) = 75, 74, 74, 79 These numbers are rough because the readings would fluctuate quite a bit when the switch first happens and trend upwards right away. After I'd remained switched to a tube for about five minutes the readings would stop rising and stabilise (varying by about +/-0.2mA): A to D (V4 to V7) = 76.9, 76.9, 78.1, 80 These readings are a bit surprising as I thought they'd be around 50mA, as per the conversion instructions. Edited July 1, 2022 by Luckiestmanalive
xlr8or Posted July 1, 2022 Posted July 1, 2022 9 hours ago, Luckiestmanalive said: I have replaced the blown fuse in my multimeter and taken some readings from the output tubes. This is obviously my first time taking readings at all and first time with this doodad. The readings took a while to stop rising (maybe about 20-30 min). Also, switching the device from reading one tube to another could be having an effect on the readings because after I switched from one to another (something I had to do slowly because the speakers would make an audible but reasonably soft thump) the reading would be roughly along these lines (mA): A to D (V4 to V7) = 75, 74, 74, 79 These numbers are rough because the readings would fluctuate quite a bit when the switch first happens and trend upwards right away. After I'd remained switched to a tube for about five minutes the readings would stop rising and stabilise (varying by about +/-0.2mA): A to D (V4 to V7) = 76.9, 76.9, 78.1, 80 These readings are a bit surprising as I thought they'd be around 50mA, as per the conversion instructions. Someone has been naughty and biased up those tubes. They need to be brought back down. They are set too high.
Luckiestmanalive Posted July 2, 2022 Author Posted July 2, 2022 Yeah, that's what I thought. So this morning I repeated the procedure with the quad bias tester but with the bottom off and the amp angled backward using a couple of laptop ramps so I could get to the pots. It's a real beggar that the two pots to adjust the bias are set at the back (making them hard to get to) and the front two only balance the two pairs of output tubes. I managed to reduce the output of V6 and V7 down to the 50s fine but I couldn't turn the other pot for V4 and V5 any lower than 70mA. So I turned the other one up to match, made sure the output tube pairs were balanced and turned it off. It looks like I will need to do more troubleshooting or find someone who knows what they are doing to look at it. The fact that one pot is fine and the other isn't suggests the two sides of the V1 (6AQ5) and/or V2 and V3 (6240Gs) are unbalanced? And the bias circuit has an incorrect or failing component?
Luckiestmanalive Posted July 2, 2022 Author Posted July 2, 2022 (edited) I've put it all back together and switched it on and there is no problem I can detect in its behaviour or music-making (hooray!). However, I think I've balanced the two sides a little better as the imaging is no longer slightly lop-sided. One thing I should have done before I put the bottom plate back is check the problem pot wasn't just sticky. I might do that now... Edited July 2, 2022 by Luckiestmanalive
Recommended Posts