Warranty vs Tweaks

[Stormy]

Hi all! Be gentle, I'm new...

I've just bought an X-Can V3, and I've been offered a 2-year warranty (and a free isolation platform to thwart the microphonics from the railway line I live next to!) as part of my £250.
Nice.
Except that I like to tweak things too, and quite fancy the Panasonic cap mod and nicer tubes...

Would I be better off holding on for two years with a "mediocre" sounding amp, or forsaking the whole warranty for the improvements? Bear in mind that everything I buy seems to break (through no fault of my own). How reliable are these things? There doesn't seem to be much that can go wrong (bar the odd fried resistor), and anything that does go should be fairly easy to repair, right? I will be doing the mods myself, and I trust my own soldering skill far enough not to break it while modding it. Would a good compromise be to just change the tubes? As they appear to be push-fit, would MF be able to detect that I've had the case open to swap the tubes if it does need warranty repair (assuming I put the old ones back in and remove any fingerprints, lol)? Which gives the more noticeable improvement, and is the whole kaboodle greater than the sum of its parts?

Thanks!

[thegreatroberto]

The MF gear I have had a 5 year warranty. Imagine how long I need to wait before modding.
Really this question is aboy 2 things.
1) the reliability of current MF gear
2) your ability with an iron

I can't answer either. (my X-cans V1 has been fine for over 6 years-no problems).But as to current gear and in particular the V3, i doubt it should be much different.
If you are just doing tubes, then there is no problem ( you can easily swap back) , but caps are a bit different.bare in mind that caps and tubes could cost £50/60, are you not better of getting a better head amo from the start ?
However if you are going with the MF and "must" mod, start with the tubes,and wait out the warranty period !

[Stormy]

I haven't been able to find many alternatives in the UK to be honest. I was looking at the Woo Audio 3 on eBay earlier but after postage from the 'States and import duty I'd be looking at £400. And then I'd probably end up wanting to tweak that. I think it makes things more enjoyable if you've had some involvement in the sound or look of something, and it makes things seem less expensive if you can gradually spend money tweaking a £250 amp, rather than shelling out £400 to start with.
The tubes look to be the greater part of your £50/60, and who's to say that the tubes in a different amp will be the "right" ones for me? Panasonic caps are cheap really. It's not like I'm going Black Gate. Yet.

[Fergus]

Stormy,

Like you, my MF X-Can V3 had a 2 year warranty. Unfortunately, I didn't discover this until I'd already carried out the tube upgrade and capacitor change.

The one good thing I discovered (apart from the substantial sonic benefits) was a noticeable degree of scorching around the resistors, which could lead, I suppose, to a catastrophic electronic failure.

However, if you've bought your X-CAN V3 recently this shouldn't be a problem, as MF now raise those resistors off the PCB (see the "Capacitor Upgrade to X-CAN V3" discussion).

Start with the tube change: get rid of those "crappy" JAN/PHILLIPS valves and try something like the ELECTRO-HARMONIX 6922s or PINK FLOYD's "shit hot" 6H23N-EBs.

Then, if you can resist no longer, order the PANASONIC FCs and implement a more comprehensive upgrade, along the lines of that recommended by PINK FLOYD.

Afterall, what future fault cannot be fixed, and the X-CAN V3, believe me, is pretty easy to work on, even for someone as "kack-handed" as me.

Fergus

[Stormy]

Thanks Fergus. This is another reason why I went for the X-Can - the community looked good. I haven't seen another head-amp with as much wealth of information as this one, and I knew any escapades into modding would be "supported" by you lot, if not by Musical Fidelity!

I think I'm leaning more towards doing it now rather than later as, like you say, problems are easily fixed (often with better componenets by the sounds of things!). I'm friendly with a component expert at work who is a fellow hi-fi enthusiast (although primarily concerned with speakers not headphones) and he fully endorses replacing every component of suspect quality in a piece of kit on principal. If the sonic benefits are in fact substantial, then I think that settles it.
I may use a mixture of FCs and FMs (for the smaller values), as I hear the FMs are even better for the capacitances that you can actually get them in. Further to PinkFloyd's comment about RS not stocking 35V 2200uF FMs, I have checked the datasheet and can confirm that 1800uF is the biggest Panasonic actually make in the higher voltages.

Fergus, I saw a photograph of a PCB with scorched resistors! I don't know if that was your one...

I had heard that the E.H. and Sovtek brands were supposed to be good, and I've been keeping my eyes open for a good pair of Mullards on eBay as something to try. I will certainly contact Tony about the 6H23N-EBs too as PinkFloyd seems so taken with them. I just hope they haven't gotten too expensive...

[Fergus]

Stormy,

Go for it!

I've no personal experience of the PANASONIC FM capacitors, but, like you, I've heard they have an even lower ESR than the FCs. Whether you would be able to audibly discern the difference between the two is arguable, although PINK FLOYD says that the FMs have a "tad more sparkle and dynamics."

Certainly, go for a "mixture" of FCs and FMs. I know from experience that the FCs sound simply stunning in the X-CAN V3. So the FMs, which have a similar sonic character to the FCs, will not disappoint.

Replace the 6 2200uF input caps with FCs, either 35/50v. Then you can substitute the 4 1000uF, and all the other smaller values, with FM types. I myself am going to change the 4 1000uF/50v FCs with 2200uF/50v FCs, as PINK FLOYD recommends.

No, the scorched PCB wasn't mine, thank God. The "trouble" in my X-CAN is limited to the left hand side - at least for the time being. To remedy this, I'm going to install heatsinks on all the transistors, replace all the resistors and raise them above the PCB.

PINK FLOYD says that MF had implemented all the required change. The date of this correspondance was 16/11/05, so any X-CANS after that date SHOULD have the raised resistors.

As for the MULLARD ECC88/6922s, there are currently a number of ongoing auctions at E-BAY. For example, a matched pair of MULLARD gold pin valves (NOS) are being sold for £49.99; or 6 tested MULLARD ECC88s - current bid £13.50.

All the best,

Fergus

[Stormy]

I've put together a BOM on the RS site, I'm just seeking everyone's approval...

6x 2200uF 35V caps to be replaced with:
6x 3300uF 35V Panasonic FC (RS 315-0798) £11.35 inc VAT

2x 1000uF 35V > 2x 1500uF Panasonic FM (RS 526-1806) £2.19 inc VAT

4x 100uF 35V > 4x 150uF Panasonic FM (RS 526-1799) £0.85 inc VAT (Discovered to be a BAD IDEA, as the 100uF caps are part of a low-pass filter, and fitting bigger caps cuts off more of the upper frequencies - no-one copy me!)

2x 10uF Polar > 2x 10uF Non-Polar (RS 521-1772) £0.14 inc VAT

4x 220uF 6.3V Non-Polar > 4x 470uF 16V Non-Polar (RS 521-1895) £1.27 inc VAT

The reasoning behind the 4x 470uF instead of 2x 1000uF for the output caps is that in theory two caps in parallel should give lower ESR, and I can't be bothered with bypassing big caps with several smaller values because it involves fiddly underbelly soldering and there might be resonance problems if I just use one cap in the spare bay to bypass a big 1000uF instead. I can always change it later. If I eventually decide to bypass the 470uFs I'll probably end up quite close to the magic/ideal 1000uF in total anyway.

50% extra on most of the caps should do it though, lol. They should fit - I checked the diameters of the replacements carefully against the PCB.

I make that £15.80 inc VAT (and yes I know I'll actually be buying more than that because of RS' multiples of 5 policy, but I'm only counting what's going into the unit.).

A quick update: I took my demo unit back today because my new one had arrived. I took it out of the box and to my dismay saw August 2005 written on the backplate. HOWEVER, when I opened it up the resistors have been raised, despite the date being the same as the demo unit.
I bought some heatsinks from Maplin and fitted them (a bit of an assortment unfortunately - Maplin never seem to have more than two or three of each of their products in stock at any given time). I've fitted the four at the output end, but had to gently coax them backwards a little to stop them fouling the second board. I had a scary moment when the legs of the smaller of one pair of transistors buckled suddenly and I thought something had snapped.
Closed it up and plugged it in, and it's actually better than the other one already, even before mods. No microphonics from the valves and less hiss at full whack (not that I listen to it that loud).

[Stormy]

RS are (as everyone probably knows) having a few problems with their supply of caps at the moment. I did however receive my 3300uF Panasonic FCs and 150uF FMs today, which I've duly fitted. I'm going to be waiting on the remaining PSU caps and the input and output non-polars until sometime in March judging by the email they sent me. I picked up some 10uF non-polars from Maplin, but they're only rated at 85ºC, and I'd rather fit 105ºC ones later rather than have to unsolder the ones I've got.

Initial impressions are more clout in the bass end of things, but I'll let them settle in and form properly before I give a full verdict. If only I'd managed to get all of them in one go...

I have a pair of 6N23P-EB/6H23N-EB on order with the legendary Tony. £9.50 apiece now, and he's worried his current batch of 50 may be the penultimate or even last batch he'll ever have, so get them while they're available I guess. Why the hell don't Sovtek copy them? I was almost wishing that spy business in Russia the other day WOULD trigger another cold war so we get some more stocks of good military valves!

[Fergus]

Stormy,

Yes, I 've heard, through the GREAT ROBERTO, that RS are having some difficulties sourcing the FCs. I urged him to go to FARNELL, as they seem to have an endless supply of them. As for the FMs, I'm pretty sure FARNELL doesn't stock them.

Some time back, I fitted 4 NITAI 220/35v non-polars as output capacitors; I think they're rated at 85 degrees. So too are the SUNTAN non-polars available at RAPID and, as you say, the SAMWHAs from MAPLINS. In fact, I've personally never seen an non-polar capacitor rated at 105 degrees, although I'm sure there must be. The SAMWHAs are excellent capacitors, and are recommended by PINK FLOYD. I would fit them, find out what sonic benefits they bring, and then sometime down the line you can replace them.

I keep checking Tony's website for the 6H23N-EBs: but as yet they haven't arrived. Like you, I might try to pre-order them. That way, you're guaranteed to get a pair.

Damned Bolshies! JACK STRAW should raise the issue with PUTIN at the next G8 meeting. No tubes no foreign investment!

Fergus

[Stormy]

I e-mailed Tony about the valves, and he's just got some in - when I said I had some on order, I've actually paid him and they should be going out tomorrow. He says he's not putting any on eBay until he's honoured any pre-orders outstanding.

RS had the 3300uF FCs in stock, so they have got some of the capacities, and they are working fine in place of 2200s.I only managed to get 10uF non-polars from Maplin, as they don't go above 220uF and I see no point in replacing the output caps with the same capacity unless they have much lower ESR.
I'd rather not fit the input caps I've got because I don't want to have to desolder them again later. I'm fine with soldering, but I find desoldering very stressful as I can never get braid to work and I don't possess a pump.
I can't be bothered with ordering any 470uF or 1000uF outputs from Rapid or elsewhere because it'll mean more postage. The RS ones are definitely rated at 105ºC, and I'm fairly happy to stick out the wait for now.

My only real concern at the moment is that the 150uF FMs I ordered to replace the 4 100uF caps are 35V, while the Jamicons I removed are actually rated at 50V. I took PinkFloyd's word for it at the ime, but looking closely at the photographs on his "Tweaks" page, his ones were rated 50V too, not 35V as he says in the text. D'oh!

[Fergus]

Hi Stormy,

Got an e-mail from Tony today, saying that the much-vaunted 6H23N-EBs have arrived - eventually. Needless to say, I ordered them.

I replaced the stock caps on the output stage and substituted them for NITAIs of the same value for two reasons: firstly, I'm not that knowledgeable about electronics and so didn't have the courage to "up" the capacity, and secondly, the JAMICON caps are simply "crap", more at home - as someone has already said - in a tumble dryer than a headphone amplifier. I thought, anything but the JAMICONs, especially in such a crucial part of the amplifier. While certainly not being the best non-polar, the NITAIs are designed specifically for audio purposes - speaker crossovers etc - and are definitely better than the JAMICONS.

I checked RS, and you're right they do sell a 105 degree NK series non-polar - brilliant! I'm not sure what make they are, although I noticed CAPXON written on the side.

Fergus

[thegreatroberto]

I contacted RS yesterday ref. their lack of capacitors.They claimed to be unaware there was a problem! So I reeled of a list of what was not in stock and was going to take 6 weeks to get in ! I also asked what RS stood for , and were capacitors still regarded as "Radio Spares".The scarcasim was lost on the young lady ! Needless to say , I felt better.

I have found that with so much cap replacement on my MF gear in recent weeks , that I must have every combination of capacitance and voltage available !
I'd be weary about under rating a capacitor in terms of voltage.i.e replaceing a 50V version with a 35V one.However I can't see whay you can't do it the other way around.However , it will have a larger dia.so space may be an issue though.Any thoughts ?

Upping the capacity is fine if the capacitor is being used,well for capacity.However if the capacitor is being used as a filter (High pass) them altering the capacity will alter cross-over of the filter.So in a power supply I can't see why bigger is not better.Elsewhere on the board, its a like for like replacement for me !

And for me, de-soldering is with a pump.I just can't get on with braid !

[Stormy]

Yes, the upped capacity thing did occur to me after I had done it with regards to cut-off frequencies...
I have noticed a change in tone, and I had been trying to pin it down until I compared it with the headphone output on receiver and was shocked at the difference (but unfortunately in a bad way. The X-Can still thrashes it for bass though. /cry). I've still got "treble", but there is definitely a lot missing from the upper registers.
From a quick glance at capacitor network construction, bigger caps lower the cut-off frequency. I should have known better.

I've messed it up, haven't I, lol. Argh. Ah well - I can order 50V 100uFs next time instead of 35V ones. It's PinkFloyd's fault - he said 35V. :¬p Ooh and what a surprise, they're out of stock! RS 526-1597
Bum. They seem to be the sole suppliers in the UK according to Google.

Well if anyone wanted to know how to soften the sound of the amp, that's definitely one way to do it (although 150uF is probably slightly too severe. 120uF would be better!).
I'll buy a desoldering pump at the weekend - may have to put the dreaded Jamicons back in for now. If the pump makes things easier I might put my 10uF non-polars in too I suppose.

[PinkFloyd]

stormy said:

Well if anyone wanted to know how to soften the sound of the amp, that's definitely one way to do it (although 150uF is probably slightly too severe. 120uF would be better!).

The 100uF caps should be left @ 100uF. Sorry about the 35V typo it should read "50V" I'll change that ASAP.

As far as the output caps are concerned 1000uF is ideal especialy if you will be using low impedance 'phones... you really can't go "too high" on the output caps so don't be afraid! They are quite literally in series between the output and head socket and are essentially there to block any DC that may be present on the output. The 440uF per channel that is already fitted is fine but 1000uF is technically better and you certainly will notice a bit more bass extension if you up the capacity of the output caps.

The 10uF input caps can be safely removed if there's no DC on your sources output. If you're not too happy about removing them (no reason why you should be) then at least replace them with non polar types, again this makes a subtle yet noticeable improvement to the sound quality.

As to bypassing the 'lytics....... go for it but don't bypass them with just one film cap bypass them with 2 or three to ensure there is no resonance at audio frequency. So, if you want to bypass a 1000uF then parallel something like a 1uF, a 100nF and a 100pF onto the cap... a mixture of polypropylene, polyester and silvered Mica would be the most expensive way to go but you'd get equally good results with say a polyester, a ceramic and a polyprop...... just mix them and give them a go you really haven't got anything to lose..... if you don't like the sound then just desolder them. Most audio tweaks are really a case of suck it and see, if it sounds good keep it..... if it doesn't try something else.

Ears are as unique as fingerprints and they are also an acoustic device (ears) so no two people are going to hear things in the exact same way. What I class as "bright" someone else may call "sharp" this is why there is so much disagreement in audio circles...... it's not necessarily the equipment, it's usually the persons ears so remember that tweaks are generally quite a personal thing and you're "tuning" the sound to your own personal taste so "NEVER" be frightened to experiment (especially with things like caps and opamps) it's fun, it's cheap and it's effective

Mike.

[Stormy]

The 6H23N-EBs arrived from Tony today, so I decided to open up the amp to rectify the mistakes and fit the new valves.

I went out and bought a desoldering pump (best £9.99 I've ever spent in terms of saving aggravation I reckon!), and removed the 150uFs, replacing them with the original "crappy" Jamicons. I was just about to say that it's a shame in some respects because a load of the hiss has come back, but that's kind of the point... the reason those 150s are so quiet is because they effectively removed the hiss band (including the desirable kind from cymbals, unfortunately). Ah well, you can't win on everything. I've checked the frequency response, and we appear to be back to normal, although it's sometimes hard to tell because the sound of the amp is different from anything else I have. I ran a test signal generator into it, and can hear up to the limit of my ears (about 19kHz - I'm still young!), so it must be ok.
Because the desoldering pump was working so well I decided to fit my stop-gap 85ºC non-polar 10uFs on the input while I was in there. I don't want to risk zero-ohm links because I use my X-Can with a variety of sources, and one day I'll forget to check something for DC.
The 3300uFs in the power supply appear to be working well, and if anyone wants my 4 spares at a reduced rate, send me a PM or something. I would say "for free", but that's about £7.50 of caps!
Now if only Panasonic and RS would get their acts together so I can do the rest!

I'll run the valves in over the weekend and see how it's sounding. Looks promising from early impressions.

Thanks for the encouragement Mike. I presume you can't think of a technical reason why the 3300s would be a problem in the PSU? I'll be using 1500uFs in place of the 1000uFs as well when they turn up, which I assume will be ok because you used 2200s there! I can't use the remaining 3300s there because they physically won't fit (fat things that they are!).

It's nice to have it back in "working" order - I didn't feel like listening to it sans-treble and didn't want to open it up until the valves arrived because it's such a faff.

[thegreatroberto]

Are we missing something here ?

The power supply section of 3 other bits of my HiFi ( power amp,pre-amp and active sub) uses 6800uF caps (of suitable voltage).

Originally my V1 X-cans had (of 35V rating) 6 off 1000uF and 2 470uF caps.........Which if configured correctly is rather close to................................

Yes, 6800uF.So given that caps can act as filters is this capacitance acting also as a mains filter ?Remembering that we are after a nice flat DC.And also remembering that caps are used as High pass filters,i.e. they don't let DC flow through them, but do appear to allow AC.

Here we all are, packing as much capacitance into our aluminium boxes as space allows but is it the correct thing to do ? Is there something magical about 6800uF, or is it just a convienient size ? Me thinks it is more than a coincidence ?

I'm sure there must be an electronics person out there that has designed good quality power supplys for HiFi just waiting to tell us !

[Stormy]

You have to take into account the fact that the X-Can is running in stereo, and therefore would have half that 6940uF per side, whereas I presume there is more than one 6800uF cap in each of your other components! The X-Can V3 has 4 1000uF caps in the power section (I think they all are, but two of them are a little way away from the others) as well as 6 2200uFs and then 4 220uFs on the output, adding up to nowhere near 6800uF.
The X-Can V3 has bigger caps than the V1 (except on the output where we have dropped to 440uF total per side, probably for lower ESR from parallel caps or better high frequency response from multiple smaller caps), and is widely regarded as an improved version of the design - but there may well be an optimum balance somewhere (assuming MF haven't already found it with the V3). Why would one of the techies at MF say something about wanting to fit Farads of capacitance in that case? I don't really know enough about electronics to be of much use here (see the blunder with the filter caps for proof), but I am aware that different values of caps have differing frequency responses and filter effects because as they become larger they tend to get "slower" and consequently can't keep up with the high frequencies.
I'm assuming this is where bypass caps come into play - the big caps provide the raw grunt for bass, but the small value caps in parallel maintain the higher frequency response without affecting the total capacitance value too much because they are orders of magnitude smaller. With smoothing caps, you are supposed to design in extra bypass caps in such a way that their resonant frequency fills in a trough in a linearity graph, but how this affects the audio I have no idea.
Is there any risk of treble loss from fitting 1000uF caps on the output without bypassing them? Is my idea of two 470uFs per side better if there is a risk?

[PinkFloyd]

Quite a high wank factor in fitting uber large capacitors into the PSU but there are also significant gains to be had. Some people say that fitting more capacitance is pretty much like polishing a turd whereas others swear by it.... I, personally, swear by it (fitting more capacitance, "not" turd polishing) extra capacitance allows longer transient current ability, the maximum current which can be supplied will be limited by the internal resistance of the filter capacitors this is where low ESR (Equivalent Series Resistance) caps come into their own. High quality, low ESR caps employed in a large array of small capacitors will be able to supply more peak current.

The V3 employs 2200uF caps where the V2 employed 1000uF types. The extra 1200uF per cap has improved the sound considerably and fitting 2200uF into the V2 has the same effect (if you can squeeze them in!) Obviously there is a sensible limit to the amount of capacitance you can fit without having to fit higher ampage diodes etc. and 3300uF would be pretty much as far as I'd go with the stock wallwart / diodes that are on the board.

If any of you ring the tech guy at MF and ask him what he thinks he'll tell you to fit more capacitance (make sure you ask him about the V2 as he's not too happy to discuss modding the V3 for obvious reasons ie: it's currently in production) I believe his own X-Can V2 has been rehoused in a different enclosure and he has fitted some serious amount of capacitance on board.

There's plenty of technical bumph about capacitance on the net and you'd end up going mad if you read all the differing opinions..... one thing I do know is that fitting more capacitance and / or replacing the Jamicons with quality low ESR types makes a significant improvement to the sound quality of both the V2 and V3 amps.. of this there is no doubt..... if it didn't sound good / was a waste of money then I wouldn't bother doing it and I certainly wouldn't recommend it.

The fact is, it does improve the sound at a relatively low cost and the people who have tried it are pretty impressed with the overall improvement... as long as you don't go totally bonkers and replace the 2200uF with 1 farad caps you'll be perfectly ok.

[Stormy]

You'd need a pretty big case for Farad caps. The ones you can buy for car audio (which will granted be bigger than necessary because they are rated at a much higher voltage) are about 4 inches in diameter and 8 inches tall!
So 4 x 470uF could/should be technically better than 2 x 1000uF?

[thegreatroberto]

Damm, you boys stole my thunder.I was going to get 6x 1F caps the size of a V2 and house them in a large biscuit tin ! At least that way you can take the top off to keep it cool.
It's good to see that Pink Floyd has contact with an insider at MF.What does he have to say about Jamicon caps ? Maybe he has modded his with some really big jamicons!
Fitting bigger caps allows more current to be drawn with less Dc ripple.I was concerned about any unwanted side effects ( in the PSU) section.But it is starting to look like their isn't one.

Well the V1 is history and I have a V2 on its way.The Panasonic FMs that are due in march ( from RS)will find their into the V2 now.So another Alps pot and new knob are on order.And a blue led.........................

[Stormy]

Sorry!
They're also about £80 apiece, at which point I seriously would be wondering why you weren't buying an all-valve amp instead, lol. I suppose Black Gates are getting that way anyway. Argh - how much would a 1 Farad cap from Black Gate cost (if they actually made one)?!