HK Audio Actor / Lucas power amps , generic problem?

[N_Cook]

They have an otherwise original assembly method of the final contact to the
pins of TO3 power transistors by 2 zero ohm "resistor" links, so 2 current
paths to the pcb traces. Anyone else observed bad solder joints to these
links on the pcb? and cause? believed conventional solder not PbF

 

[N_Cook]

I repair many of these, and can honestly say that I have *never* observed
any bad joints at the place you mention. By far the commonest problem is
failed output transistors. This occurs because the heatsinks are only just
about adequate with proper cooling. As soon as the fan air intake vents
start to clog up with fluff, the outputs start to run too hot. Over a period
of time, this dries out the heatsink paste to a powder, leading ultimately
to transistor failure.

Whenever I get one, I always remove the other pair of transistors as well,
clean down their heatsinks, and re-paste them, not forgetting the flatpak
transistor that's in contact with the underside of one heatsink on each
channel.

The manufacturers recommend that when the outputs are replaced, two of the
BC546Bs nearby are replaced as well (T7 / 8 on one channel, 10 / 11 on the
other). Check also C3 and C21 to make sure that they are not bulging.

Other than this, these amps are very well behaved, and new outputs and fuses
will, in 99.9% of cases, effect a complete cure. Note, however, that they
have proper differential inputs, so are not that easy to drive correctly,
unless you have a proper balanced XLR source, and that they don't like
earthed test equipment connected to their outputs / inputs simultaneously. I
usually hook a completely isolated speaker to them for final check, as the
music shop which sends these to me for repair, often remove the amp chassis
from the cab, to ease the transport, and save me having to strip it all out.
He now tells customers when they collect the repaired unit, that they should
brush out the air vents at three monthly intervals.

Arfa

Looks as though repaired 2 years ago,in UK, with (see logo thread ) with
MJ11015/16 power Darlingtons that I suspected could be pirates as could not
find logo.
2 of the TO3 devices shorted C-E on one side of the "H"
1 of those 2 are original colour but other three have a dirty brown/grey
colouration of cap. I cannot figure out the failure mechanism. Will have to
remove the other good pair to check under and will heatsink the 0R links
while doing so. I doubt I disrupted the pcb solder when desoldering the B &
E pins with the failed ones.

 

[N_Cook]

I agree with Arfa, I`ve repaired dozens of these and always found the
solderwork to be excellent. Perhaps someone else has been in there
before you. There was a mod for early units, different output
transistors and the addition of a couple of 1N7007 on the print side of
the board.

Ron(UK)

Amp seems 2004 but replacement devices of matching dates on each pair of
2006 and 2007, looks wll enough done repair, he would have seen any
associated duff solder points in the process, surely. I don't like the
closeness of the +/- live vanes of the h/s on a board that can easily flex
with heat.

 

[N_Cook]

I repair many of these, and can honestly say that I have *never* observed
any bad joints at the place you mention. By far the commonest problem is
failed output transistors. This occurs because the heatsinks are only just
about adequate with proper cooling. As soon as the fan air intake vents
start to clog up with fluff, the outputs start to run too hot. Over a period
of time, this dries out the heatsink paste to a powder, leading ultimately
to transistor failure.

Whenever I get one, I always remove the other pair of transistors as well,
clean down their heatsinks, and re-paste them, not forgetting the flatpak
transistor that's in contact with the underside of one heatsink on each
channel.

The manufacturers recommend that when the outputs are replaced, two of the
BC546Bs nearby are replaced as well (T7 / 8 on one channel, 10 / 11 on the
other). Check also C3 and C21 to make sure that they are not bulging.

Other than this, these amps are very well behaved, and new outputs and fuses
will, in 99.9% of cases, effect a complete cure. Note, however, that they
have proper differential inputs, so are not that easy to drive correctly,
unless you have a proper balanced XLR source, and that they don't like
earthed test equipment connected to their outputs / inputs simultaneously. I
usually hook a completely isolated speaker to them for final check, as the
music shop which sends these to me for repair, often remove the amp chassis
from the cab, to ease the transport, and save me having to strip it all out.
He now tells customers when they collect the repaired unit, that they should
brush out the air vents at three monthly intervals.

Arfa

Thanks for that - both of those caps are bulging, I had not noticed up to
now. I will replace those and the TO92s. A tip in reply for these and other
apparatus with that grey interconnect ribbon. Before removing any such
boards I run some hot-melt glue down the join of the cables at and to the
board. So any bending, you have to do, is then only in the cable run which
consists of multistrand but soldered wire, so will easily break where not
supported by the grey plastic sheathing, ie at the ends, where it otherwise
would choose to bend.
I would have thought the ducted air design was about ideal for a given fan
size, but I will bend inwards the outer fingers between the +/- dc carrying
adjascent heatsinks. Perhaps the airstream takes the path of least
resistance and does not pass through the 44 vane version of this type
http://www.elfa.lv/images/tn/d9b8dde0-803b-11dd-9ab0-005056b94d7a.jpg
staggered fingers of those square format heatsinks and goes around outside
instead. Perhaps bending inwards all 20 internal run fingers of each h/s
would be a sensible mod, less vortexing/blocking, more open, would make less
of a dust trap as well

 

[N_Cook]

The sort of wide pitched grey ribbon that has end stripped and conductors
soldered directly to pcb, rather than headers

 

[N_Cook]

Having a great deal of experience of these, I feel they are a fundamentally
well-designed and constructed amp - both electronically and mechanically,
and I don't think that I would contemplate altering any aspect away from the
way it was designed, including the arrangement of the heatsink fins, and the
airflow paths. The heatsinking and cooling of the heatsinks *is* adequate as
designed, but becomes a bit marginal if the units are thrashed hard on a
regular basis, whilst the air intake vents are allowed to clog. If the
transistors / heatsinks of both channels are cleaned down and repasted
during the repair process, and the owner is alerted to the importance of
just scrubbing a soft paintbrush around the vents once in a while, they are
unlikely to give further trouble. I've never had any repeat failure issues
with them, anyway.

I take your point about the inner 'row' of fingers, but I don't think I
would feel inclined to bend them in more than a few degrees, otherwise, they
might actually block convection off the transistor cases ??

Arfa

The good pair had good solder points under. All white goo was in oily state,
One of those electros was o/c the other high ESR and both had more obvious
bulge at the lead end

 

[N_Cook]

Having a great deal of experience of these, I feel they are a fundamentally
well-designed and constructed amp - both electronically and mechanically,
and I don't think that I would contemplate altering any aspect away from the
way it was designed, including the arrangement of the heatsink fins, and the
airflow paths. The heatsinking and cooling of the heatsinks *is* adequate as
designed, but becomes a bit marginal if the units are thrashed hard on a
regular basis, whilst the air intake vents are allowed to clog. If the
transistors / heatsinks of both channels are cleaned down and repasted
during the repair process, and the owner is alerted to the importance of
just scrubbing a soft paintbrush around the vents once in a while, they are
unlikely to give further trouble. I've never had any repeat failure issues
with them, anyway.

I take your point about the inner 'row' of fingers, but I don't think I
would feel inclined to bend them in more than a few degrees, otherwise, they
might actually block convection off the transistor cases ??

Arfa

I have to differ on the aerodynamics. This sort of 4mm wide double wall,
staggered finger design, is for convection not passage of air through. Walls
of 4mm gap blocked by 4mm finger repeated four times each side of the fan (2
h/s in line each side) is pretty effective wind break. Then 2 "gaps down the
sides of 3mm wide x 35mm high and 70 mm long is similarly "resistive". Most
cooling would be the 2 outer edges of 12 only of the 44 vanes each h/s where
there is unobstructed passage outside the heatsinks and out to the
surrounding duct

 

[Phil Allison]

"Nutcase _Kook"



** LEARN to *** TRIM *** you FUCKING IDIOT !!

2.5 lines tacked on the end of a 6K mess is

NOT FUCKING ON !!



..... Phil

 

[N_Cook]

20 degrees is about the most or you block off access to the TO3s. Maybe only
tinkering at the heat build up problem. Most air must go through the 25mm or
so wide gap in the ps direction (over the tips of the vanes), rather than
the 2 off 5mm gaps to the duct let alone through all those vanes. I would
have thought there should be some shuttering to block off most of that large
gap along the ps side , to force most air round and through the heatsinks.

They were assembled with enough white goo to keep a goth happy for a week.
So much that she filled the B & E holes so pins pushed through it, not
cleaned off, and some white goo actually under the solder joints (not
actually failures there). Those failed caps 47uF, 63V - in similar HK amp
they were rated 40V 47uF , what is the problem in that area of the circuit?

 

[N_Cook]

If make is relevant these blown caps, 2 types so I don't know which is
original, both dark and light blue
47uF , 63V
SG and logo of an eye, type?
R4D 85deg C (M)

probably original one as amp is 2004 and previous repair (if only 1) 2007
47uF, 63V
YAGEO
08/03 (date Aug 2003? )
SK85 deg C

 

[N_Cook]

Oh, BTW, can you confirm which way the fans run? Sucking air out the
cabinet or blowing it in?



Gareth.

This one is outwards
4,000 sq mm of purpose built inlet holes and 3,300 sq mm through fan vent I
would have thought was quite adequate if the air was directed over the
heatsinks and not wind-breaked. This one permanent mounted high up away from
floor dust.
Compared with Laney CD850 , totally inadequate constriction on outlet over
fan .
3 of those I've cut away the "vent" holes over the fan, fared off and
covered with a wire grille and they've not bounced back (each was cutting
out thermally after extended but normal use)
As the Laney original design just 14 small slots with a combined area of
only 1.5 sq
inches, let alone vortexing, so overloading the 3 inch diameter fan
with through ducted area of about 6 sq ins , ridiculous aerodynamics

 

[N_Cook]

Well, I agree with Arfa and Ron. I've repaired lots too, since the cooling
is also IMHO inadequate, which is quickly made worse by fluff around the fan
vents.
Never had any dry joints, but seen plenty of the bulging caps Ron mentioned.
I replace them with 105 degree types regardless of apparent condition.

One warning - it is easy to accidentally short the solder tags the big
diodes are soldered to, to the zero ohm links as you tighten the nuts. If
you power up fully in this condition with 4 amp fuses you will blow the
Darlingtons again.
You have probably noticed both speaker output tags have amplifiers on them,
there is no grounded speaker connection, though each output is referenced to
ground.



Gareth.

I see no circuitwise reason for those repeated cap failures. The Darlington
heatsinks are elevated off the pcb (trackside) with spacers and although
the minor component side of the pa pcb is un-fanned I would not have thought
pcb and working caps would get to 85 deg C
Will make airflow mods and RTV-stick non-resettable thermochromic dots
around various items , for if it bounces back.

 

[N_Cook]

Thought so.
I did once think about reversing the fan, blowing air into the amplifier,
but there is too much risk of the impeller being fouled by a blow to the
casing around the vents. Surely this would mean more internal turbulence,
thus better cooling?
It must have been designed this way round for a reason, though, (?) so I
gave up on that idea.



Gareth.

The main disadvantage that way round is all internal stuff gets heated up to
some intermediary temperature, between ambient and heatsink temp

 

[N_Cook]

Maybe a batch of dodgy caps, along with the dodgy Darlingtons?!

100v 105C caps are not exactly expensive or physically large, so a bit of a
no brainer for bullet proofing this application.




Gareth.

no room for that size, unless put on track side of the pcb . As it happens
space there and as the temp sensing TO220 s and fan air flow/temp sensor
that side already I will solder replacements on that side I think