Samson XP308i Expedition. Lost in the upper reaches of the Unpopo. I wish.

This a nice picture of a Samson XP308i.


Shortly after I’d dropped it off a cliff. Or maybe just before. In this case it wouldn’t have made much difference because it didn’t work.



I know this might sound Neanderthal, but I don’t like switchmode power supplies. Uggy-wuggy-grunt. There you go, I am Neanderthal. I’m Stone Age enough not to run my business on bullshit and subterfuge, which currently shrinks my playing field down to the size of a postage stamp.

So what is wrong with switch mode power supplies? More to the point, what is a switchmode power supply? And also why is it? And what is the alternative?

All these riveting issues you’ve fallen asleep over, will be rivetingly detailed in our next week’s riveting issue, called ‘How to sell something you haven’t got.’ No, not really…… onwards and inwards.


This is a picture of the main bit of a switchmode power supply. The finned thing is the heatsink, upon which is screwed a switching regulator (in this case a Fairchild  KA1M0880B power switch which is the nearest device you see screwed to it. The next one along is a high current diode and the one furthest away is a mosfet transistor which beefs up the output from the switching regulator to supply the amp circuitry. There are few different ways to design these things but the idea is that the mains input is directly rectified to produce something like 330 volts dc. This an impractical voltage as it stands. It is too high to do anything with for a semiconductor amp, and also, d.c. is not readily manipulated to produce the various different voltages that the amp needs; in this case probably +15 volts, -15 volts, 5 volt digital supply, and then maybe +30 volts or so, and -30 volts ish.

So we’re starting with a voltage from the supply that is way too high, and we can’t easily modify it.

An orthodox power supply has an isolating mains transformer. The primary side is connected to the mains supply, and there is no direct connection to the electronic gubbins in the amplifier. The secondary side of this transformer has several windings (or coils of wire, if you like) wound on the same lump of metal as the primary winding. By having various sized windings, many different voltages can be available, and the whole idea is relatively simple.

The switchmode idea has no isolation from the mains, it goes straight into a bridge rectifier and comes out of that as high voltage d.c. as we’ve already said. So where do the voltages come from? Well, the switching regulator we mentioned ‘chops’ the d.c. and produces a high frequency (in the Samson case 75 kHz) and this can be transformed to give the various voltages by much smaller transformers, because of the high frequencies involved.

So what was wrong with the traditional power supply, that we need to do away with it?

One of these things you might have already guessed. Wait for it……..a switchmode……

IS CHEAP!!!!! Yayyyy! Even including the little transformers a switchmode like this is unlikely to cost the manufacturer more than a few quid.  On the other hand the ‘old’ transformer arrangement is probably ten times the price to build.

Another big plus for the switching supply, and especially in the case of toy p.a. stuff like this Samson is, is that it’s lightweight. A traditional power supply is a lot of things but it is definitely not that.

So why don’t I like them? It’s fair question. Might it be that I don’t like anything less than fifty years old? Well………….alright, you got me there. But I do like ideas that save the good old planet’s resources, and it does save a lot of copper and steel by not building a big transformer. But……………………

Switchmode has many problems that I don’t like. First (whatever they say) I see a lot that are blown to hell. They have many protection devices in there (overvoltage, overcurrent, overtemperature, overblown claims of longevity…… well maybe not that) but they still seem to hit the rocks very regularly. Second, there is no isolation between you and the mains supply. I Don’t like that. Third, it doesn’t have the same inherent filtering of mains noise of an isolating transformer. Fourth, as with many unnecessarily ‘complicated’ things there is a lot more to go wrong.  Fifth, (again, whatever they say) the switching noise generated inside the unit will leak into the mains supply. We don’t need more crap in the mains that we already have. Just one of these things, well, ok; but a hundred thousand, say? It just seems to be a step in the wrong direction, to me.

I’ve actually led you astray, here. (So what’s new?) Because this Samson Expedition thing I haven’t fixed. Yet.


This is where I get into another ‘Pet hates’ tirade. The pic shown is of the sheaf of paper I have so far assembled in the interests repairing this amplifier. Why did I need to do this? It’s because of the increasingly secretive nature of the so-called ‘free information society’. After spending several hours on the web searching for schematics for the XP308i, I was forced to draw the conclusion that there isn’t any.

I’m sure there are a bagful of bullshit reasons for this, but the obvious one, you rarely find mentioned. Samson don’t want these things repairing. Just a guess, you understand.

Being a very bloody-minded person, that, I find is a great encouragement to figuring out what it’s all about, how it works, and fixing the damned thing. I’ll bring news of this job in due course, but just as a hopefully useful aside, to anybody unfortunate enough to get stuck inside one of these:- Beware of applying too much heat to the tracks, they’re very flimsy. Check that you have 330 volts dc out of the mains bridge rectifier. Check, that from pin one on the KA1M0880B(that’s the nearest pin to the edge of the heat sink) to the negative pin on the main rectifier, you have a significant ac signal. Don’t use a grounded scope (or anything else grounded) on this switching part of the circuit, as it is all referred to virtual ground. If none of this makes any sense, probably best not to go there.

We will be back on this amp, we will fix it, and I will let you know how it was done. And hopefully Samson, and all those of a similarly secretive fraternity will get it as a sound kick up the arse.

Time for tea.


Fender Deluxe reverb

Marc is a nice bloke, and an intelligent one, until it comes to the issue of GAS. When this rears its bankrupt head, it’s true that most of us of a musical persuasion lose any grip we might ever have had on the excuse for reality that we have forced on us.

For the uninitiated, GAS is the acronym for ‘Gear Acquisition Syndrome’, although you would have lived forever at the bottom of a well, not to have come across it.

I have a Stetson hat and welding goggles for my GAS attacks. “How does that work?” you might wonder in an idle moment under the duvet. It goes like this. The bloke who walks into the music shop with welding goggles and a Stetson with the brim pulled down to his feet, is me. I stare at the floor, ask for my bag of strings and get the hell out before anything hanging on the wall commands me to buy it.

What has all this to do with Marc and his Fender Deluxe reverb? Well, he does seem to buy a lot of things that end up in my hands being fixed. The latest being his nice Deluxe that didn’t sound right when he cranked it up. Bearing in mind that the amp is only (but worthily) rated at 22 watts rms, might he just be asking for it to do more than it reasonably is capable of?

It was a strange fault, and this is why I’m jotting about it. Strange, but in essence, simple; when you know what it is. The same could be said about Blackpool rock, Dark Matter, quantum physics, and the Dyson vacuum cleaner. All simple after it’s been done; by somebody else.  

If somebody complains about something not sounding right, crossover distortion and intermodulation distortion come high on my list of things to look at, because they both have symptoms not easy to describe. If your amp sounds like a bear’s arse being filed down with a hatchet it’s not a subtle thing. That depends on your degree of sensitivity and whether or not you happen to be the bear. But the subtle faults are notoriously hard to describe.

Crossover distortion often makes an amp sound ‘weak’. If biasing is overdone on the output valve control grids, the positive and negative -going signals cross over badly and you get a weak sound and early clipping, so lower power output.

Intermodulation distortion is when your nice guitar sound gets mixed in with some other noise like, say 100hz buzz (poor smoothing, check the reservoir/smoothing caps) or 50 hz hum (check the hum balance if you’ve got one).

This was neither of those things, and in the first place, didn’t exist at all. One of the things you figure out early on in your engineering training is the effect a bench has on all things broken. It often makes them work. Briefly. So the MesaBoogie just brought in by a suicidal musician who watched it disappear under the wheels of the 8.24 from Paddington, is shovelled onto the bench where it rests for a second or two, before producing the most glorious music. So he says “It never did that for me, after the train ran over it”.  

Extreme case, I grant you, but it happens all the time. Back to Marc’s Fender Deluxe. It was perfect. These things just look right on a ‘scope if you’ve looked at scopes for a long, long time. So he had it back. No fix, no charge. I’d be very wealthy if I charged for the things I didn’t fix. You know, like the microwave in the Aspidistra Hotel in Llandudno. I’ve never been there but it’s the same principle. Just send a bill somewhere if I get a bit short of the readies. I’d probably do well in the tax office. Anyway, he got it back home and it was just the same. I was getting the ‘Black Hole ’ feeling about it.

To cut an already tedious story to the slightly less boring. 

It turned out that many(!) if the power valvebase contacts had been poorly manufactured (it had to be a manufacturing fault, because the thing looked nearly new even though it wasn’t) a bit of a jiggle around in the car, and the screen grid contact on one of the 6V6′s open circuited, and resulted in that particular valve putting out a fraction of its share of the power. I slotted a watchmaker’s screwdiver down each side of all the valve base contacts to tighten them up, and everything was, and still is, hunky dory.

Often the repair is far easier than the figuring out of the what it is you should be repairing.

Tea time.