Luxman L-85v Integrated Amplifier

[sabristol]

I tried a brief simulation and the simulated 1kHz THD+N vs. power is somewhere between 0.001% and 0.01% (-100dB and -80dB), so probably the real world would not be better than -80dB. I have overlooked the high gain (82k/680R), so it is as it should be. Thanks for bringing this circuit into ASR!

View attachment 117151

View attachment 117152

Excellent. I've got it full modelled in LTSpice too - although I'm a newbie and crusty so it takes me ages to workout what to do, and I don't know all the tricks.

Many thanks.

 

[restorer-john]

Great! And I might have built the circuit. Wouldn't it be a great project to have 3 similar units to compare?
I posted a smile but I am serious.

Absolutely. It won't happen overnight, but I'm all for it. I really should get on to the restoration of this L-85V. I have all the parts waiting and ready and as you can see, it's partially done.

It will be essentially original except all the electros are Nichicon UFGs, some key noisy semis replaced, and everything (right down to the switch contacts on every selector rebuilt and cleaned). OPTs and the basic power stage will be stock NOS or orignal semis. The "blob' diodes are already replaced with paired Si diodes and shrink covered.

 

[restorer-john]

This is what I was querying- the floating almost touching T0-220 tag. :

Quite different to your CAD drawing above.

 

[sabristol]

Your bottom end rolloff is typical of the tested and reviewed FR plots of the L-85V. The top end however rolls off much more aggressively in your amplifier than both my experience with Lux L-xx amplifiers, and the UK technical review I have at hand. The L-85V review I have shows perhaps a maximum of -1dB at 40kHz and ruler flat to at least 30kHz. You are -1.25dB at 20kHz and only testing at 1W. Let's see what your PBW is.

The measured half power (40W/-3dB) bandwidth of the amplifier when new was 43kHz.

PS. What are the values of these components BTW? I presume your signal ground lift resistor is 68R.
View attachment 117148

Well spotted ;-)

I had hum problem even after multiple iterations of board design. What you spotted are the bodges I had to make on the as fitted; removed R2 and C5 and replaced R3 with the 68nF cap.

 

[restorer-john]

I see you've used Nichicon FG. I decided against that as there are some better suited for PSU decoupling and for the input and feedback coupling the books recommend bipolarised UES series.

I like UFGs. They may well be rated 85deg, but I have come across commercial products where the UFGs have been sitting on seriously hot sections, up against charred PCBs and they not only survive, but stay in spec. Much of it has to do with the generous physical dimensions and the ability to get of heat, unlike some of the sub miniature 105deg caps which seem to fail or go high ESR prematurely.

Sadly Nichicon is/has discontinued a lot of the small value med/high voltage UFGs. Muse is also another option, but their range has shrunk massivley too.

 

[sabristol]

This is what I was querying- the floating almost touching T0-220 tag. :

View attachment 117158

Quite different to your CAD drawing above.

Good man - you have an eye for detail

The photos were during development, but should the V2.1. I drilled and tapped the heatsink for the three transistors Q13, Q14 and Q15 and mounted them. The picture must have been taken just before I'd finished that.

As there seems to be interest I'll take some current photos when I get a chance.

In the meantime here is the layout:

 

[sabristol]

I like how you've used the original TO-3 transistor sockets and the placement of the two boards look lovely. What's the deal with that TO-220 that is almost touching the rear panel?

I've got a few L-81s at Dad's and an L-85V here partially restored for him.

The 2SC-1345s are the noisy Hitachi "tombstone" types IIRC.
View attachment 117140

If the leads are black (oxide) get them out fast.

Here's some pics of the original powerstage boards and the Lux L-85V I have here:

@restorer-john what do you direct replace the 2SC1345 with btw?

 

[pma]

The HF roll off of the amplifier seems to be about 100 kHz in simulation. So there should not be a problem. More probably input RC parts and generator output impedance.

 

[restorer-john]

I had hum problem even after multiple iterations of board design. What you spotted are the bodges I had to make on the as fitted; removed R2 and C5 and replaced R3 with the 68nF cap.

You are in the UK, right? So you have a proper earthed unit, not a double insulated one? Like us here.

Lux seriously messed around with chassis, single point, multiple point and never really committed to proper star earthing or separated the cap/0v CT/chassis tie points.

I'll have to pull out the L-85V that's here (right next to my bench in two boxes) and check the earthing/chassis arrangement, but IIRC, the units we received were multivoltage (set for 240V) and had some issues with suppression caps to chassis earth depending on mains wiring polarity too. You would have likely had the same models in the UK. If you get a cap to chassis, you get eddy currents in the chassis itself which can/may capacitively couple through the T0-3's large collectors to the heatsink/chassis. I had that problem with an amp a while back. Drove me nuts until I floated the entire amp module from the chassis. Never forgot that little exercise in frustration.

 

[restorer-john]

@restorer-john what do you direct replace the 2SC1345 with btw?

I have NOS 2SC-1843s but you can source 2SC-1844s that will be fine.

 

[restorer-john]

More probably input RC parts and generator output impedance.

That's why I was querying his input filter component values. Read too much Self and Cordell and you'll have a pile of stacked filters up front. (Not that it'a bad thing, but...)

 

[AnalogSteph]

I have NOS 2SC-1843s but you can source 2SC-1844s that will be fine.

The go-to NPN small signal TO-92 transistor these days is KSC1845... probably a whole lot easier to find than any of those. (I no longer find a product page for these. Don't tell me they're EOL, too. ) KSC945 should work as well, a little tighter on Vceo than the originals. (In terms of PNPs, KSA992 is good as well but KSA733 is no great shakes if memory serves.)

 

[restorer-john]

The go-to NPN small signal TO-92 transistor these days is KSC1845... probably a whole lot easier to find than any of those. (I no longer find a product page for these. Don't tell me they're EOL, too. ) KSC945 should work as well, a little tighter on Vceo than the originals. (In terms of PNPs, KSA992 is good as well but KSA733 is no great shakes if memory serves.)

It was. The KSC-1845 is non-stocked by Mouser. Wanna put together a group buy for 2000+. I could be in it. Just order 2000 and split it two or three ways.

Even then, you needed a bag full to match the suckers. And forget so-called NOS on the secondary market- they could be anything.

The original 2SA-733 (G/GR) was amazing. I still have plenty, but they are as rare as rocking horse you know what.

No, not the 2SC-945. That's the BC-548/9 of epitax world. I'm showing my age, I still have bags of original Philips 547/8/9, 557/8 and I think some 639/640s- tin whiskers on the legs and all.

Actually, we should do a thread with ancient (and rare) semi close up pics.

 

[pma]

Lux seriously messed around with chassis, single point, multiple point and never really committed to proper star earthing or separated the cap/0v CT/chassis tie points.

That's quite usual issue of many amps. I was about to ask about measured hum/buzz spectral components. They kill the resulting S/N often. I am seriously thinking about several mods to improve noise and usable swing - the amp has high supply rails 2 x 55V but is very ineffective regarding usable output voltage swing and power. I will also add a DC servo, the simulation shows high uncontrolled output DC offset.

 

[pma]

I'm showing my age,

Hehe, don't be funny, I am 65

 

[sabristol]

You are in the UK, right? So you have a proper earthed unit, not a double insulated one? Like us here.

Lux seriously messed around with chassis, single point, multiple point and never really committed to proper star earthing or separated the cap/0v CT/chassis tie points.

I'll have to pull out the L-85V that's here (right next to my bench in two boxes) and check the earthing/chassis arrangement, but IIRC, the units we received were multivoltage (set for 240V) and had some issues with suppression caps to chassis earth depending on mains wiring polarity too. You would have likely had the same models in the UK. If you get a cap to chassis, you get eddy currents in the chassis itself which can/may capacitively couple through the T0-3's large collectors to the heatsink/chassis. I had that problem with an amp a while back. Drove me nuts until I floated the entire amp module from the chassis. Never forgot that little exercise in frustration.

Yes in the UK the metal chassis / any exposed metalwork has to be connected to mains earth. The 0V is tied to the chassis too.

I can't say I fully understood what Luxman had done with grounding, especially power and grounding of the Tone Control Amplifier. My confusion is partially because some of it isn't in accordance with the service manual schematic.

I was having a breather from doing the Luxman L-85v as the hum problem was frustrating me and repairing a Trio receiver instead. Looking at the Trio circuit made me realise what was wrong with mine and voila - sorted!

 

[Plcamp]

You are in the UK, right? So you have a proper earthed unit, not a double insulated one? Like us here.

Lux seriously messed around with chassis, single point, multiple point and never really committed to proper star earthing or separated the cap/0v CT/chassis tie points.

I'll have to pull out the L-85V that's here (right next to my bench in two boxes) and check the earthing/chassis arrangement, but IIRC, the units we received were multivoltage (set for 240V) and had some issues with suppression caps to chassis earth depending on mains wiring polarity too. You would have likely had the same models in the UK. If you get a cap to chassis, you get eddy currents in the chassis itself which can/may capacitively couple through the T0-3's large collectors to the heatsink/chassis. I had that problem with an amp a while back. Drove me nuts until I floated the entire amp module from the chassis. Never forgot that little exercise in frustration.

I think I need to discuss this with you before I finish grounding scheme for the DH220C I am about to bring up. Some questions include...

a) Should I change to 3 wire ac in to ground the chassis?
b) The old unit directly chassis grounded both inputs, should they be 10 ohm lifted with a bypass rf cap instead?
c) The old unit routed the driver card ground directly to speaker return and then to star ground. I am wondering if the driver card should instead directly attach to star ground?

 

[restorer-john]

I was having a breather from doing the Luxman L-85v as the hum problem was frustrating me and repairing a Trio receiver instead.

The Lux gear from the 70s was still very much steeped in the tube designs they had done previously. It was almost like they had dropped in silicon where vacuum state was previously. Their chassis designs were set up for tube/valves. I even have some early pieces where they stuck transistors in rubber grommets placed in holes in the chassis and went point to point. (a few brands did that- Sansui was another).

All their amplifiers of that era were variations on the same topological theme, but they sounded really good within their limits. The L-81 was my first real amplifier after building several EA/ETI designs I didn't like. Then came a Sansui AU-D5 and the Lux was pensioned off. Then came an HK PM-660, then big pre/powers etc.

But I still have a soft spot for the Lux amplifiers and have several in the restoration queue.

 

[restorer-john]

I think I need to discuss this with you before I finish grounding scheme for the DH220C I am about to bring up. Some questions include...

a) Should I change to 3 wire ac in to ground the chassis?
b) The old unit directly chassis grounded both inputs, should they be 10 ohm lifted with a bypass rf cap instead?
c) The old unit routed the driver card ground directly to speaker return and then to star ground. I am wondering if the driver card should instead directly attach to star ground?

That's a whole new thread, start one perhaps? It's not really DIY, but we have no restoration forums either.

Floating input returns/grounds and various stages is an art. More often than not, it results in vaporised low value resistors. Ask me how I know that.

I have some commercial examples that are just brilliant, and plenty of others where there are so many return points you have no idea where the problem starts or finishes.

@pma, can you weigh in?

 

[Plcamp]

You can see in post 40 of the Hafler rebuild thread where I described and put a picture of some of what I have done. I have many questions about what I see as compromise decisions on grounding.