Warwick bass combo overheating and blowing output transistors.

[Sam Spoons]

I have this lovely little Warwick bass combo in the studio* which has an excessively noisy cooling fan (a little 40mm computer jobbie) so I disconnected it for general recording duties. But I forgot I had done so and we used it for a couple of rehearsals, the most recent was a bit louder than usual and the amp died (thankfully during the last tune). This has happened once before (without the fan disconnected) on a gig when I had loaned it to a keyboard/piano player who was sitting in). In that case it was the output transistors which I duly replaced and have had several years of service since.

I will try to diagnose and fix it next week but, in the interests of preventing such a disaster reoccurring I'm wondering if replacing the very crude bent aluminium heat sinks with something more sophisticated like this https://www.ebay.co.uk/itm/297193488995 ... R6qVkYjPZQ might allow me to throttle the fan down to reduce noise or remove the need for a fan entirely (or at least improve reliability under load)?

Plan B is to connect the fan via a switch on the back panel but that relies on me remembering to switch it back on for loud rehearsals (I won't forget to turn it off when recording, that's for sure )

I realise having the fan thermally controlled would be ideal but that may be a bit more complicated than I would like to attempt, at least in the short term.

* Technically I guess it belongs to #2 son but he doesn't use or need it these days so it's on permanent loan to me.

[Folderol]

A large properly fitted heatsink will make a lot of difference, especially if the fan is positioned so the flow is along the fins rather than straight at or across them - less turbulence and also quieter. That's assuming there is a clear route for the air to get in and out of the case!

[Wonks]

Though to some extent whilst it makes sure the heat is getting away from the transistors, they will also ensure the heat is warming up the insides on the amp chassis. You may find that the fan stops/starts a lot more.

Given the wide range of computer fans a available, could you find a slower, quieter fan? I doubt you need the full airflow rate the current fan provides.

[resistorman]

You might look into a Noctua fan, less than 20 bucks US
https://noctua.at/en/nf-a4x10-flx

it might also be possible to feed it less voltage to slow it down.

[James Perrett]

I realise having the fan thermally controlled would be ideal but that may be a bit more complicated than I would like to attempt, at least in the short term.

Adding a simple thermal switch is probably simpler than adding a heatsink. Something like this

https://www.switchelectronics.co.uk/pro ... rminals-no

perhaps?

[ajay_m]

A bigger heatsink has more thermal inertia and therefore will take longer to reach a given temperature but unless there is some way for this heat to escape, ultimately it will not really help. What is needed is airflow and as has been said a low noise fan- Noctua are as suggested a very good choice - will be important.

[Sam Spoons]

Thanks everybody, It really is a nice little bass amp with decent eq, 150 real watts and a 10" driver in a ported cab. It works nicely with electric bass and with my EUB in the studio and is adequate for low volume gigs. It just doesn't seem to like being driven flat out for long.

Since it seems likely I will have to replace the power transistors anyway swopping the heat sink for a better one is not too much of an extra step so I'll probably do that anyway (assuming I can find one that will fit in the available space) and I will look into a quieter, more efficient fan. Even if it only improves reliability that will be worth doing (though two failures in 20 years* is only borderline unreliable TBH).

I have been looking at those heat activated switches recently in my two!!! dodgy microwave ovens. That seems like a good solution and looks very simple to fix.

* Especially with one being self inflicted

[ajay_m]

Of course unfortunately a fair few amp designs don't always fully consider the safe operating area in the semiconductor datasheet. In fairness to the designers this can be tricky as the load is inductive and challenging to model. But those data sheets mean what they say and even a brief excursion outside the SOA can lead to device destruction, even though junction thermal limits were not exceeded.
Found this out the hard way with the original Quad "current dumping" amp design and a couple of failures at high but not clipping output levels. Of course, monitoring the device case temperature and shutting down via a thermal switch won't hurt but it might not prevent further failures.

[James Perrett]

Since it seems likely I will have to replace the power transistors anyway...

If it is a typical direct coupled class AB amplifier then you may find yourself going round in circles as the failed power transistors may just be a symptom of some other problem. If you replace the power transistors without finding the real problem then the replacements will be destroyed fairly rapidly.

[Sam Spoons]

It worked as a fix several years ago and the amp has continued to work until Thursday evening but I'll try to find a schematic and post a link for further fault finding advice later.

[Sam Spoons]

A bit of serious googling this arvo reveals that small Warwick bass amps have a bit of a rep for overheating and the next model after mine has a larger, more open, chassis with two, larger, fans and vents in the sides of the cabinet so it sounds like forced cooling is a bit essential I'll add some vents in the sides to improve airflow but it looks like I'm stuck with the fan.

I have also found a schematic for this later model (CL ND4) but the circuit has also been redesigned so is not particularly helpful...

I've emailed the UK distributer/service centre and asked if they are able/willing to share the circuit diagrams and PCB layout.

[Wonks]

I'll add some vents in the sides to improve airflow but it looks like I'm stuck with the fan.

Don’t knock it. It may be the only fan you ever get!

Be careful when making extra vents. You need to make sure the fan draws air across the transistors and their heatsinks. If you provide an extra low resistance path for the air, make sure the air path to the fan from it/them is over the heatsinks. Otherwise you could end up with a bigger problem than before.

[Sam Spoons]

I'll have you know I've got three huge fans (they're all in my new Mac Pro and hardly ever make any noise )

Yes, good advice Wonks I'll bear that in mind. I don't yet know for sure if it's the power trannies that have popped, I just know that was what happened in similar circumstances all those years ago. With the annual gig coming up I'll have to put the repair on the back burner for a couple of weeks but hopefully by then the importer will have sent me a schematic to work from.

[S2]

A person more cruel than me would be tempted to say that it's the best thing to happen to a Warwick amp!

[Sam Spoons]

You ain't the only one going off their user forum... But I really like this little bass combo, it sounds nice with electric bass or EUB, is loud enough (when some numpty doesn't disconnect the cooling fan ) for moderately loud rock band rehearsals with a live drummer and is small and compact so sits nicely tucked away a corner of my tiny project studio.

[Sam Spoons]

Gig out of the way I have fired up the little Warwick combo and lo and behold it has returned from the dead... I have no clue what happened to it but I am saved the effort of repairing it. I will look into that Noctua fan resistorman linked to before adding additional heat sinks or thermostatic control.

[Hugh Robjohns]

Sounds like it went into an over-current or thermal shutdown mode.

Does it have any self-resetting fuses or the like?

[Sam Spoons]

I'm guessing that it must. I'll investigate further when I get (hopefully, I emailed the UK distributor 2 weeks ago but haven't had a reply yet) a circuit diagram from Warwick.

[Sam Spoons]

The reply arrived from Warwick's UK distributor yesterday saying they have them for internal use only but they attached the .pdf to the email anyway. I have said I won't share them on a public forum but I guess I could pass a copy on privately if anybody can use it.

I will peruse later an see if I can work out what happened.