'AK Mods' SC Class A 15W Amplifier

[Deleted]

Feb 20, 2012 10:03:32 GMT drymdrum said:

Alex,
with regard to the air cored coil how much leeway do we have in the value around 4.7mh?
I'm heading out to get the coils I've wound measured and see if I'm in or out of the ballpark.
Regards,
Alan

Hi Alan
They aren't that critical. If you use the correct formers just wind about 20 1/2 turns of enamelled copper wire on them.
If anyone has the coils from the Jaycar HA, it would be O.K. too, as they are actually 4.7uH .
Regards
Alex

[Deleted]

Hi Alex,
that's fine then, I don't have the specified Philips type former, they are available from Jaycar but postage makes them a tad expensive for what they are.
However I've wound four to the approximate size on DIY formers and popped them in for measurement at a friends transformer winding factory.
Regards,
Alan

[Deleted]

some history..In the early days amps had little parts (most discrete designs still have) and the parts that were available had little gain back then, more gain nowadays.In these days overall feedback wasn't much 'feedback' at all and certain amplifier designs clearly showed the points, where the components that did the positive half and the negative half power part (push-pull design), took over from each other. Class-B amplifiers.Nasty sounding distortion and was often even visible on an oscilloscope

Hi Frans

interesting points

I'm sure that you know that some people are still building those early Single Ended Class A circuits. most modern tube amp ''designs'' are usually based on 1930/40/50 circuits from companies like Western Electric. I've built quite a few of those myself and found that with the right speakers that they can sound pretty good. i think that the simplicity of the circuits that you mentioned may have something to do with the goodness of the sound but that's just my opinion so i may be wrong on that.

by the way

some of those early transistor designs sounded truly horrible even when compared to the valve amps they replaced. i had a leak stereo 20 and 30 and the transistor 30 was just plain awful.

IMHO i suspect that the early rush to transistorize had as much to do with production costs and the ability to mass produce more easily.

I've built a few Push Pulls also which to my ears loose some of that goodness IMHO. maybe that loss of SQ has something to do with crossover distortion that i can't hear.

''True that most people don't go above 1W but to go twice as loud 10W is needed.
Everyone that has had amps with power meters noticed when turning up the volume the power increases rapidly.
To go twice as loud as 10W you need 100W.
If you need to play loud and still have short peaks to be properly played back (say demonstrations in larger rooms to impress people or drive inefficient speakers) you will need 500W to 1000W''.

yup good points which broadly match the measurements we made at collage.

the comment about loudspeaker efficiency is also a good point and why i was asking about that earlier in the thread.

''So having huge amounts of Watts can be handy, though normally I did not exceed the 1W limit''.

yes having huge amounts of Watts can be handy but not really needed if the speakers/amp are well matched. I've played 1.5W into 98dB 8ohm (Quazar OB's) speakers in a 25M x 40M room (to impress people) and it filled the space nicely.

Building speakers yourself (especially own design filters) can give strange results where certain spots have impedances as low as 1 Ohm while using 8 Ohm drivers. This can happen when someone has made a filter 'by ear' that is 2nd, 3rd or 4th order or accidentally used a wrong value without knowing.
Building a 'flat impedance' speaker takes a lot of effort with complex filter designs with lots of compensation circuits.
Even first order filter speakers where the tweeter and woofer take over with a single inductor and cap can give lower than expected impedance values if for instance the tweeter and woofer have overlapping area's with a wrongly configured filter, or filter that has been altered by someone to sound better 'by ear'.

I've never built anything just by ear in my life but most commercial speakers are designed by using calculation and then voiced by ear. seems to work well (in some cases). could be that only designing by calculation is just as bad as only using ones ears.

I'm a firm believer in using both methods.

my DIY speakers sounded bad due mostly to lack of ability on my part which is why i tend to buy or build other peoples designs. ditto amplifiers but it does not get in the way of my musical enjoyment so I'm happy.

take care

[Deleted]

Feb 20, 2012 10:03:32 GMT drymdrum said:

Alex,
with regard to the air cored coil how much leeway do we have in the value around 4.7mh?
I'm heading out to get the coils I've wound measured and see if I'm in or out of the ballpark.
Regards,
Alan

Hi Alan

i used the Jaycar formers for my build and to be honest i have not a clue how they measure as i have now way of doing that.

i used 20 1/2 turns of 1mm and locked the wire in place on the former with some 25mm heat shrink.

mmmmm

winding

you must be getting closer

well done HFSM

take care

[Deleted]

Feb 20, 2012 11:18:54 GMT @sandyk said:

Not to put too fine a point on it, but Silicon Chip magazine and Frans are not in agreement on rated power. The attached panel is from the later 20W Class A amplifier which runs at + and -22V rails and is slightly more efficient due to no emitter resistor in the VAS.
Rather than degenerate into a war of words, both viewpoints should be left as they are and no further correspondence entered in to.

This is after all a construction thread, NOT a discussion thread.
Those interested, may wish to learn more about this area from Douglas Self's books.
Alex

Hi Alex

fair point

sorry for going a little OT.

take care

[Deleted]

Hi All
Ok back to business

I completed my final set up today which is nice and at the same time a little sad.

I’ve enjoyed building /listening to this am very much and also Alex’s tutoring and support when things came up unexpected.

Thanks again Alex

Yup sad that this build and the fun I’ve had from it is at an end.

So how I did my set up.

I used the mini clip leads that Alex recommended for the HA/PRE build on long leads to allow the lid to go on whilst monitoring the front end balance and voltage across the 4 1ohm resistors on the OP.

I also found that having two multi meters was useful as it allowed me to monitor and adjust one channel at a time.
So patience is needed for the set up process so allow plenty of time to get things right.

Ok first off I roughly set the front end balance to zero and put 240mv across the 4x 1ohm resistors with the lid off.

then i popped the lid on and let things warm up a little.

Then I played about a CD’s worth of music to allow things (especially the front end) to settle.

I’m using IN5819 which from memory take longer to settle down than the SF12 but it pays to allow plenty of time for the front end balance. When you first switch on and listen try to hold that SQ in your head.

Sooo after a nice session with Mazy Star and Hope Sandoval I whipped the lid off and reset the front end to 0mv and reset the OP V to 240mv which is a little less than the 250 I found by ohms law but it’s cold here so allows a little fore changes in room temperature.

did a quick check with the lid back on.mmmmm looking good.

I just did the same for the other channel (Stereolab-Emporor Tomato Ketchup) and screwed the lid in place once i'd rechecked that all was A OK.

Job jobbed.

Take care

Oh yes that sound we heard at switch on

Now compare to the sound you have now that it’s all warmed up with the front end balance sorted.

Marvel at the extra detail/focus and clarity.

I have

it's just sounding superb

[Deleted]

Hi Shaun,
everything you've done so far has helped us all to a better understanding of the nitty grittys we all have to get to grips with on this quest for the HIGH SQ so well done on firing ahead and blazing the trail.
Of course without a certain AK's mighty tuppence worth on this forum most of us would be scratching our but ends wondering what to do next and I know that had I not happened on The Grotto by chance .............all hail a certain Mike ............I'd certainly not have the HIGH SQ up here in the box room I call The Den.
By the way Shaun it's exactly 12 foot by 8 foot and fits me like a glove.
I enjoyed your chat with Frans despite the moderators comments..............seems it's hard to keep a DIY thread squeaky clean.
As to my build it's slow but getting there.
I've two boards cut and almost fully populated just two caps to get as I missed them out on my major buy from RS, thought I'd get them from Maplins but they don't have the values.
Next up will be the speaker protection boards so I'll have to look at the circuit and suss out a layout get the boards cut and made up.
Do you have an Idea of the type and size of speaker cabinets you'd go for if building them yourself, I know you said woodworks not your thing but if you had detailed drawings of a desired design B&Q will cut the MDF to your requirements, there after power screwdriver and drill are about all the tools you will need.
Regards,
Alan

[Deleted]

Feb 21, 2012 0:13:52 GMT drymdrum said:

Hi Shaun,
everything you've done so far has helped us all to a better understanding of the nitty grittys we all have to get to grips with on this quest for the HIGH SQ so well done on firing ahead and blazing the trail.
Of course without a certain AK's mighty tuppence worth on this forum most of us would be scratching our but ends wondering what to do next and I know that had I not happened on The Grotto by chance .............all hail a certain Mike ............I'd certainly not have the HIGH SQ up here in the box room I call The Den.
By the way Shaun it's exactly 12 foot by 8 foot and fits me like a glove.
I enjoyed your chat with Frans despite the moderators comments..............seems it's hard to keep a DIY thread squeaky clean.
As to my build it's slow but getting there.
I've two boards cut and almost fully populated just two caps to get as I missed them out on my major buy from RS, thought I'd get them from Maplins but they don't have the values.
Next up will be the speaker protection boards so I'll have to look at the circuit and suss out a layout get the boards cut and made up.
Do you have an Idea of the type and size of speaker cabinets you'd go for if building them yourself, I know you said woodworks not your thing but if you had detailed drawings of a desired design B&Q will cut the MDF to your requirements, there after power screwdriver and drill are about all the tools you will need.
Regards,
Alan

Hi Alan

thanks for the nice words it's appreciated

but rather than blazing a trail I'm just following the well worn path set out by Alex.

if i might add that my methodology may be wrong or i may have missed something.

I'm hoping that if I've made mistakes that they will be picked up on and then we all learn something.

so the above set up was exactly how i set the HA/PRE just with different OP bias.

what caps are you stuck for?maybe i have some spare

your build sounds like it's coming on nicely and you are wise to include speaker protection. I'm looking forward to your company in the HQ audio club.

just as an aside

i changed the offset corrector IC's in my pre from TL071 to AD744 and the noise floor seems to have dropped noticeably. now even with my ear inside the speaker cone i can't tell if the amp is switched on or not and likewise the tweeter. no no no noise whatsoever. pretty impressive.

on loudspeaker building

yes I've build a few pairs all of which where rejected bu the master of domestic aesthetics. ;D ;D

so fare I've had a go at 1/4 wave TL and a few ported jobs but not had to much luck with the SQ.

I've found Loudspeaker building to be a bit of a black art with all of the lining/stuffing/tuning involved not to mention trying to get the veneers right.

if only building speakers was just about calculation/build and bobs your uncle they sound great. but it's not so in my limited experience.

so it's idiot proof for me I'm afraid.

sorry for the slight drift OT

take good care Mr soon to be very luck indeed SQ wise HFSM

[Deleted]

Hi Alex ,
the 68pf poly caps...............
what voltage for these?
I see on Ebay a lot of 25 @ £8.40..... 68pf 30v are these ok?
..................................................................................
RS have them @ £10.90 plus vat pack of 10
Polystyrene axial capacitor,68pF 160Vdc
.......................................................................
Or a little further afield(Melbourne Au.)
Polystyrene Capacitor 68pf 125v NOS - x 5 @ approx half the RS price.
If the 30v ones are okay then I'll go for them, if not looks like an Aussie deal.
Alan

[Deleted]

Feb 21, 2012 15:54:21 GMT drymdrum said:

Hi Alex ,
the 68pf poly caps...............
what voltage for these?
I see on Ebay a lot of 25 @ £8.40..... 68pf 30v are these ok?
..................................................................................
RS have them @ £10.90 plus vat pack of 10
Polystyrene axial capacitor,68pF 160Vdc
.......................................................................
Or a little further afield(Melbourne Au.)
Polystyrene Capacitor 68pf 125v NOS - x 5 @ approx half the RS price.
If the 30v ones are okay then I'll go for them, if not looks like an Aussie deal.
Alan

Hi Alan

i used the same polystyrene caps listed in the original HA/PRE build.

from Farnell

i think that some may have used polypropylene but as i understand it polystyrene is slightly better at high frequencies which is why they may have been specified.

as usual i may be wrong on that

LCR COMPONENTS - FSC 160V 68PF 2.5% - CAPACITOR, 68PF, 160V

160V is pretty standard for Polystyrene so I'm guessing that is why they where available at Farnell without to much fuss.

expensive yessss but nice.

IMHO with caps especially polystyrene it's about build quality (the caps are caps brigade are going to disagree on that which is fair enough). also polystyrene has a tendency to melt (like ice cream in the sun) if to much heat is applied which could cause problems.

so be careful with those 30V ers they may have a pretty thin dielectric (kid gloves). the 160V are more likely to withstand a little more heat from the iron sooooo if it where me I'd go for those.

sorry I've no spares of those

but

you could go for the RS ones take what you need and i'll have the rest for the up and comming GB. Pay Pal is at the ready.

all of the above comes with the usual Government fool warning and i may be wrong .

anyway lets see what the boss recommends.

looking good Main Man

take care

[Deleted]

Shaun and Alan
I would only use 63V or higher rated caps there.The main thing there is not to use ceramic types. Silver Mica is also suitable.

Regards
Alex

[Deleted]

Hi Shaun,
I've gone with RS for both 68pf and 100pf as I'll have 'em tomorrow............have to say I like the service RS have been giving for some time now................... so that completes the main board builds component wise............'errr LS devises not withstanding.

Had a call from my friend regarding my 4.7 inductors.
It went something like this.........................
Ray....
Alan, I reckon you need about sixty turns on your formers.............

Alan....
C'mon Ray you must be joking that can't be right, the guru quotes 20.5

Ray....
Not according to my calculations and the meter are you sure it's 4.7 milly henrys?.............
pause for Alan to clear his throat................as the penny drops
Ray asks..........
what symbol before henry?.................................(((inside Alan's head a little blighter is saying over and over......Bugger!!)))

Alan............
Ummm' ahhh' I hate to say this Ray but it is similar to a "u"

Ray.....
Okay Alan leave it with me.............
........................................................
Need I say more.
Alan

[Deleted]

Hi Alex,
thanks for the info, just saw your post was typing my screed as you were yours.
The deed is done so got the right ones.
Regards,
Alan

[Deleted]

i changed the offset corrector IC's in my pre from TL071 to AD744 and the noise floor seems to have dropped noticeably. now even with my ear inside the speaker cone i can't tell if the amp is switched on or not and likewise the tweeter. no no no noise whatsoever. pretty impressive.

Shaun
Very interesting comments .
I had hoped that the 100nF capacitor to earth at the junction of the 2 output resistors of the offset corrector would have completely killed any residual HF noise fdue to the opamps high gain, but apparently not. I had better check that all mine are using either OPA134 or preferably AD744, which also normally results in very low DC offset.
Regards
Alex

[Deleted]

Hi Shaun,
AD744 will be used here also.
I'll wing the spares of Polystyrene axial capacitor,68pF 160Vdc from RS your way, there should be six as I need four. I'll probably have some SF12 spare as well and I'll look at what else there might be.
Regards
Alan

[Deleted]

Alan
Don't give all your spare SF12 away, as they are ideal in small low noise PSUs that you might want to make a little further down the track. They are also faster than the UF4002 etc.
Kind Regards
Alex

[Deleted]

Feb 21, 2012 20:45:01 GMT drymdrum said:

Hi Shaun,
I've gone with RS for both 68pf and 100pf as I'll have 'em tomorrow............have to say I like the service RS have been giving for some time now................... so that completes the main board builds component wise............'errr LS devises not withstanding.

Had a call from my friend regarding my 4.7 inductors.
It went something like this.........................
Ray....
Alan, I reckon you need about sixty turns on your formers.............

Alan....
C'mon Ray you must be joking that can't be right, the guru quotes 20.5

Ray....
Not according to my calculations and the meter are you sure it's 4.7 milly henrys?.............
pause for Alan to clear his throat................as the penny drops
Ray asks..........
what symbol before henry?.................................(((inside Alan's head a little blighter is saying over and over......Bugger!!)))

Alan............
Ummm' ahhh' I hate to say this Ray but it is similar to a "u"

Ray.....
Okay Alan leave it with me.............
........................................................
Need I say more.
Alan

Hi Alan

good move on the RS order and as i say any you may have over i'll pick up for the GB but i'm gonna pay and pass on the cost so lets do that

no rush this end.

Yup it would need about 60 odd turns for 4.7mH but we a talking Micro (MU - Greek but looks like a U). i used two air core inductor calculators i found on the net to calculate the turns ratio for the size of bobbin-thickness of wire . both came up 20 1/2 turns for 19AWG wire. you can imagine how delighted i was when a few days later Alex recommended the same thing.

so a 3 way check.

Alex mentioned that the value is not critical and TBH looking at the reason for using the inductor outlined in the original build article i believe him.

you could pop in BC560 BC550 (or MPSA18 correctly orientated ) for the front end just to get her up and running. then when the LS arrive you know you've won the jackpot. just leave them on loooong legs and snip them out to replace later.

that would let you carry on the good work

Oh yes it's tool of the week time. (errr no not me)

i grabbed a pair of those surgical forceps from Maplino's and boy have they been useful. grip and lock on component leads for the tear free removal of said component. it's quick,easy and saves over heating the board whilst struggling to grip the little blighters..

I'm sure glad i had a pair

so Alan lets not let a few LS get in the way

i have 550 at the ready which although are a little HFE challenged will do for the get go.

take care HFSM

Hi Alex

TBH all that i was expecting when changing the IC's in the PRE/HA was slightly better offset correction. so i was pretty surprised i was to find better offset correction (expected) and a drop in the noise floor (unexpected) i had assumed that this was because the 744 was a better spec'd device but apart from that i just don't know why the above should be.

OK so now my DIY foible is out

yes i always have a quick listen up close to my speakers at switch on always have done as it points to any small problems when the amp is most stressed.

well that's my excuse

mad mad mad

it's the quietest amp I've ever heard or built with most of the class A stuff having that characteristic Hisssssss.

what I've also found with your Class A is that it does a great job of avoiding that euphoric over warm sound that most of these types of amp seem to go for. you know that solid state for Valve lovers sort of sound.

it also seems to have real speed and impact which I'm thinking is going to have more than a little to do with the regs/JLH.

good power supplies yes I'm a firm believer .

take care

[Deleted]

Hi Shaun,
boards are built up just the 550/560's to go in I have them to hand
but got diverted by Daffy Duck
Regards,
Alan

[Deleted]

Hi anyone thinking of building Alex's mod'ed 15 Watt as per this great thread.
Daffy Duck suggested today this simple way to obtain the
inductors for the build.

Here's a pic.

[Deleted]

Feb 22, 2012 23:06:08 GMT drymdrum said:

Hi anyone thinking of building Alex's mod'ed 15 Watt as per this great thread.
Daffy Duck suggested today this simple way to obtain the
inductors for the build.

Here's a pic.

Hi Alan

great work especially the imaginative former less coil winding.

rock solid construction

well done HFSM

nearly there

take care

[Deleted]

Hi All
So I’ve had Alex’s Class A for a while now so I thought that it was about time to

Talk a little Turkey

IMHO In My system to my ears

Yup I was going to do a nice big review with all the trimmings but with this set up it’s really not so easy.

I could just say errr this is great and that’s fantastic but this combination does it all so well I think that it may be more useful to talk in more general terms.

OK every time I listen to it I just go back to the same thoughts that I had when listening to the HA.

It’s so clear and revealing but not clinical. It lets you know when the source material is not up to scratch but never complains about it, making the best of whatever I play through it.

But play some high quality recordings and it rewards with a level of sumptuousness that’s hard to buy at any price.

It seems to have a knack of giving you the heart and soul of a performance and just lets the music happen without getting in the way or imposing on it.

Had me dancing around the room (dogs bemused) to OTT-Scylon and really knows how to party when it needs to but also has a light touch when needed.

So to sum up

Alex’s amps offer top to bottom transparency, musicality and control that’s just fantastic.

What price would I expect to pay for an equivalent commercial PRE/HA PA?

I’ll let you know when I find something that gets anywhere near the quality of Alex’s amps.

Priceless.

I’m just blown away

All you people about to build this Amp I’d say this (to quote Monty Python)

You luck, lucky B*****rds

Take care

[Deleted]

Hi Alex,
link below to RS UK
uk.rs-online.com/web/p/general-purpose-relay/5116964/
is this a suitable relay for the SPK protection circuit?
Regards,
Alan

[Deleted]

Feb 24, 2012 23:19:16 GMT drymdrum said:

Hi Alex,
link below to RS UK
uk.rs-online.com/web/p/general-purpose-relay/5116964/
is this a suitable relay for the SPK protection circuit?
Regards,
Alan

Hi Alan

is there a PCB mount equivalent? may be easier to work with.

so just the one board to go

take care

[Deleted]

Feb 24, 2012 23:31:58 GMT @shaun said:

Feb 24, 2012 23:19:16 GMT drymdrum said:

Hi Alex,
link below to RS UK
uk.rs-online.com/web/p/general-purpose-relay/5116964/
is this a suitable relay for the SPK protection circuit?
Regards,
Alan

Hi Alan

is there a PCB mount equivalent? may be easier to work with.

so just the one board to go

take care

Hi Shaun
I haven't checked that out, but this looks like a lot of work, and would need flying leads.Add to that the cost of the terminal blocks and the rest of the bits.
Why not the real kit? It's not exactly expensive, and seems excellent VFM
Regards
Alex

[Deleted]

Hi Shaun,
I'm working up to the whole thing slowly. PSU built and working, main boards built but not fired up yet. I'll build the SPK board.....working on it now full DIY version, then the case as it will have the heatsinking for sides and back. I'm not inclined to hang big chunks of metal off of the OPTs or fiddle around jury rigging. I can wait for the whole to come together.
I've a fair idea now what size the case will be so order up the metal bits soon.
Regards,
Alan