DIY Valve Guitar Amp Head

[Marcel]

Soundcraft were state of the art back in the '80's. Very top end gear along with SSL and Yamaha. Interestingly back then on the wholesale and the servicing side both Yamaha and Marshall in Australia were handled by the same people. Not sure if it is still the case today.

For a cap replacement request I'd be tempted to do audio frequency response tests first across the whole desk, find the worst slots, repair them, and then repeat the repair on all the other channels. That way you end up only replacing what's needed to restore 100% performance. Unless the client specifically asks for it and is prepared to pay then doing a wholesale replacement can get substantially more expensive in both time and money, and may not necessarily cure all problems.

I have a 20/8/2 Soundtraks Megas Studio desk from the late '80's that has an interesting Aux send fault. None of the 6 sends work and the supply resistor for the sends in the Master bus module gives up its smoke the moment you turn the desk on. Have swapped the chips and most of the caps but without an accurate circuit this MIDI controlled desk is proving difficult (a bastard in every possible sense of the word) to repair which is a shame as it is a fantastic home studio desk if I could get the sends working... 4 band dual parametric EQ, 6 sends, 16 mono and 4 stereo modules, 8 bus, dual monitor volumes, LED metering, MIDI mute control/recall,..... Thankfully nil mice have ever been in there before me...lol

Those dirty contact faults can be annoying and pesky....lol ... and I'm surprised you didn't get a replacement ribbon.

[DrNomis_44]

Soundcraft were state of the art back in the '80's. Very top end gear along with SSL and Yamaha. Interestingly back then on the wholesale and the servicing side both Yamaha and Marshall in Australia were handled by the same people. Not sure if it is still the case today.

For a cap replacement request I'd be tempted to do audio frequency response tests first across the whole desk, find the worst slots, repair them, and then repeat the repair on all the other channels. That way you end up only replacing what's needed to restore 100% performance. Unless the client specifically asks for it and is prepared to pay then doing a wholesale replacement can get substantially more expensive in both time and money, and may not necessarily cure all problems.

I have a 20/8/2 Soundtraks Megas Studio desk from the late '80's that has an interesting Aux send fault. None of the 6 sends work and the supply resistor for the sends in the Master bus module gives up its smoke the moment you turn the desk on. Have swapped the chips and most of the caps but without an accurate circuit this MIDI controlled desk is proving difficult (a bastard in every possible sense of the word) to repair which is a shame as it is a fantastic home studio desk if I could get the sends working... 4 band dual parametric EQ, 6 sends, 16 mono and 4 stereo modules, 8 bus, dual monitor volumes, LED metering, MIDI mute control/recall,..... Thankfully nil mice have ever been in there before me...lol

Those dirty contact faults can be annoying and pesky....lol ... and I'm surprised you didn't get a replacement ribbon.

Reason why I didn't get a replacement ribbon was that there weren't any available so I literally had to go through and cut out all the worst chewed bits and splice all the good bits together, it took a while but I managed to get it all done, it was great for exercising my patience....lol.

It sounds to me that there might be a short-circuit somewhere after that supply resistor for the sends in the master bus module, that's the first thing I'd be suspecting, and it's probably the reason why the sends aren't working, you could try having a good look underneath the circuit board with a magnifier to see if there are any slivers of solder creating unintentional shorts, I'd also try measuring the voltage on either side of that resistor too, I agree, trying to service a piece of gear without a good circuit diagram is like trying to solve a mystery blindfolded, I wish all manufacturers would get their heads out of that "gotta protect our designs" mentality and start producing decent, ie clearly legible, circuit diagrams and easily obtainable service manuals, but a lot of electronic gear these days are designed with planned obsolescence included in it, back in the early days of electronic gear manufacture, things were designed and built like tanks.

[Marcel]

I know exactly where you are on the map with that Doc, and I'm right beside you.... Protecting IP by not publishing circuits for technicians in my mind is the quickest way to go from fault to rubbish dump and earning yourself a rubbish reputation at the same time. Sadly many management types are taught and only see the situation as 'turnover of product'....Which is fair enough if the item is sold for a few bucks, but that thinking makes no sense once the product gets past a weeks wage worth in value... Unfortunately most computerised products like smart phones and the like have morphed into their own little modular world in the 'circuits' discussion...

For the moment the Soundtraks desk is safe-n-sound sheltering in the too hard basket. Multi layer boards stacked 3 deep on a 2 module wide frame along with short ribbon cable lengths and nil circuit, and a huge lack of urgency. But one day, when I'm more in the mood and have the need...
I did find an UK enthusiast group for the desk, I got channel module circuits from them but they don't have the right circuit for the master module of my desk. Actually they were surprised to learn that my desk has MIDI which makes it a rather special rarity. Also found out that in the UK it is a very popular type of desk for analogue affectionardo small studio owners as it has nil digital audio sections at all and enjoys the highly desirable traits of a superb noise floor and very effective and acoustically pleasing EQ.

[Marcel]

Found this of FB. Thought it might be of relevance when choosing caps for an amp (or guitar) build...

http://zerocapcable.com/?page_id=224

[Simon Barden]

Not a very good test at all IMO. They measured the capacitance, then they put a recording of a guitar pickup through a digital filter using the capacitor value as either a value in the filter or using it to determine the filter's cut-off point. You then hear (if there is a sound clip) the recorded signal played back through the filter - which gives the capacitor only signal (as if the tone control has been turned town to 0).

Now you aren't going to hear any real tone difference when so much of the signal has been filtered out (and using mp3s doesn't help either), so you do need to hear it with just bit of the tone rolled off - which didn't happen. Also there must be more subtle things happening that you aren't going to hear with a muffled and data compressed sample. I don't expect to hear much difference at all between the capacitor types, but I'm sure that the different dielectrics must have some varying frequency-dependent characteristics.

Also, whilst fairly slight, there would still be some interaction between the tone control and the pickups as the tone control is adjusted. Their sound clip was taken direct from a pickup (hopefully into a hi-Z input) with no tone pot in circuit. If you want to do a test, you should do it properly. I'd have used a multi-turn pot with a lock for the tone control, so I could set some mid-position settings accurately and with no chance of accidentally moving the pot between different capacitors.

I'm not sure how you could get a consistent signal through the pickup (I get their point), But there must be a means of inducing a repeatable signal from an external device (e.g. TC Electronics send messages to their effects pedals through pickups with sound from a mobile phone).

Then I'd look at frequency response and frequency over time response charts of the signals to compare them. If there was next to no difference, then I'd say that there really no difference between capacitors. However, they do mention that ceramic capacitors are prone to microphony, so best not to use them. But other capacitor types, though less prone to microphony, may still be affected slightly by loud noise, so this should also play a part in the test, so add a guitar amp playing back at both quiet and loud volumes and see if there is any difference.

[Marcel]

I had good laugh when I saw this report, and I never bothered listening to the sound samples as I concluded after reading the entire text that by their test methods they were not going to see any differences anyway.

My own feelings about this topic are mixed. On one hand as a tech I can see that a 0.022uF cap is exactly that and given a fixed frequency and circuit to work in then the performance of that cap should not vary if it is replaced with any other cap of the same value.... BUT.... we are not dealing with a fixed frequency, and we are dealing with a complex LCR circuit which does have a semi-random mechanical aspect to the equation, and we are dealing with human hearing which has immeasurable variables from person to person so it is conceivable to me that replacing one cap with another will have an effect which is interpreted differently between different people.

To quantify that 'interpretation' is a difficult if not near impossible task. Minor changes in capacitance value, ESR, stray coupling, the hand that strikes the string are but few of the things that will 'colour' any persons interpretation on 'how it sounds' or if 'it sounds good'. And those few variables are relatively minor when compared to the biggest factor of 'what the listener believes'. Faith in all its forms has a huge impact on our choices and opinions.

Personally I think belief plays an important part of any performance. Be it a politician on a podium or a musician on a stage or a soldier in a army or a pilot in an aircraft or a cook in a kitchen,.... if you believe in yourself, your gear, those around you, your tools, those supporting you then your performance will excel.... and 'tone caps' are a (small) part of that excellence...

[Simon Barden]

I'm not one for using the tone control on my guitars except on rare occasions and only to get a more jazz-like sound, so any of the finer nuances would be generally lost on me anyway, though I do now normally fit Sprague orange drops to my guitars (If only because their size and stiff legs make them easy to fit) but have bought some PIO caps from Bareknuckle to fit on my ES-3 kit as I was also buying some other bits from them at the time and thought - why not!. They weren't ridiculously expensive, maybe twice their price for orange drops, so I thought it worth a go.

[Marcel]

I'm guilty of that too Simon. Bought a 10 pack of 0.022uF Orange drops at 3 times the price that I can buy Vishnay or Panasonic caps of the same value off my wholesaler. It was a bit of a waste as most of my current heard of guitars are fitted with humbuckers... You live and learn...
I agree that the Orange drops on a physical/mechanical level are very nice to work with. Haven't tried any PIO caps from Barenuckle.

[DrNomis_44]

Before I moved to Darwin in 2000, I bought a kit from Jaycar Electronics which was for building a device called an ESR meter, you'll probably know what ESR stands for Marcel, but for the benefit of other members, ESR is short for Equivalent Series Resistance, now the ESR of a capacitor is an important specification, in Electrolytic Caps you want the ESR to be as low as possible, usually it's about less than an Ohm in fairly large value Electrolytic Caps, the ESR of a Cap is the Effective Series Resistance that the cap presents to AC, and is a very good indication of the health of the Cap because when the Cap goes bad and dries out, the ESR skyrockets in value, or goes high, lower value caps, eg say a .022uF/600V PIO type, will have a higher ESR than say a 47uF/450 V DC Electrolytic, Low ESR is an important spec for Caps because you tend to find a lot of low ESR Caps in modern Switchmode Power Supplies.

I might see if I can dig up my ESR meter and do some restoration work on it so that I can use it again, I think it just needs a couple of new test leads made for it, and maybe a new 9V battery clip, if I can't find it I might see if I can buy a new ESR meter kit from Jaycar cause there's supposed to be an updated one.

I think at least one of my DIY guitars uses a .1uF PIO cap as the tone cap, they are pretty good sounding, to my ears the common garden-variety 100V Greencaps seem to have a tone that has a harsh high-end to it, or maybe I'm just imagining it....lol.


When you choose a Cap for use in an electronic circuit, you have a few criteria that the Cap needs to meet, for me those criteria are as follows:


1, Working Voltage- (Very Important) The Cap needs to have a Dielectric (plate insulator) that is able to withstand the expected voltages in the circuit when the circuit is energized, or powered up, otherwise the Dielectric will puncture and develop a short-circuit.

2, Tolerance- (Not so important in some cases) The tolerance of a Cap is how much it actually measures in value +/- the marked value, most modern Electrolytic caps can actually measure anywhere from +/- 10-20% of the marked value on the Cap, in some cases, like in tuned radio circuits, or audio filters you want the tolerance to be fairly tight, note that most of the time it's not that critical.

3, Physical size- This is relatively self-explanatory, you wouldn't want to try and use a huge oil-filled can-cap in a circuit designed to fit in a case that's about the size of a matchbox.

4, Age- Again self-explanatory, you wouldn't want to use a Cap that's so old it's literally crumbling away to dust.

[Simon Barden]

And on no.1, the working voltage should ideally be half that or less of the caps rated value. Above that, the capacitance value drops off (so you might need to use a bigger value capacitor than normal if that's the case). And when working with AC circuits, if the cap only has a DC voltage value on it, then remember that the RMS voltage of an AC circuit is less than the peak voltage achieved by the AC signal. For a sine wave this is 41% greater than the RMS value.