(AK) classA HA/Preamplifier.

[Deleted]

Hi All

from Alex's SCHA tweaks on page 1

''NOTE. BC547C, BC548C and BC549C may be used if BC550C isn't readily available.

BC557C ,BC558C and BC559C may be used if BC560C isn't readily available.''

I'm finding the bc550c / bc560c combo a bit hard to source. 550c £15 P.P from Farnell (US order).

i think that for the JLH in the classA some of the above are going to be ruled out due of their low voltage capabilities but how about the bc549c and bc559c at 30v they look possible.

I'm new to transistors so maybe I've missed something really important.

i'm here to learn so any thoughts would be much appreciated.

take care

[Deleted]

Nov 27, 2010 11:30:15 GMT @shaun said:

Hi All

from Alex's SCHA tweaks on page 1

''NOTE. BC547C, BC548C and BC549C may be used if BC550C isn't readily available.

BC557C ,BC558C and BC559C may be used if BC560C isn't readily available.''

I'm finding the bc550c / bc560c combo a bit hard to source. 550c £15 P.P from Farnell (US order).

i think that for the JLH in the classA some of the above are going to be ruled out due of their low voltage capabilities but how about the bc549c and bc559c at 30v they look possible.

I'm new to transistors so maybe I've missed something really important.

i'm here to learn so any thoughts would be much appreciated.

take care

Hi Shaun
They are as cheap as chips from RS Australia, so I presume RS U.K. too ?
Alex

australia.rs-online.com/web/search/searchBrowseAction.html?method=retrieveTfg&Nty=1&Ntx=mode%2bmatchallpartial&Ntk=I18NAll&Ns=stockPolicy_au%7c1%7c%7cnew_au%7c1&Ne=4294967294&Nr=AND%28avl%3aau%2csearchDiscon_au%3aN%29&N=4294624261+4294864882&Ntt=BC550C&multiselectParam=&selectAttribute=ON Semiconductor#breadCrumb
australia.rs-online.com/web/search/searchBrowseAction.html?method=searchProducts&searchTerm=BC560C&x=30&y=15

[bizzie]

Nov 27, 2010 11:30:15 GMT @shaun said:

Hi All

from Alex's SCHA tweaks on page 1

''NOTE. BC547C, BC548C and BC549C may be used if BC550C isn't readily available.

BC557C ,BC558C and BC559C may be used if BC560C isn't readily available.''

I'm finding the bc550c / bc560c combo a bit hard to source. 550c £15 P.P from Farnell (US order).

i think that for the JLH in the classA some of the above are going to be ruled out due of their low voltage capabilities but how about the bc549c and bc559c at 30v they look possible.

I'm new to transistors so maybe I've missed something really important.

i'm here to learn so any thoughts would be much appreciated.

take care

They have them at Cricklewood at 10p each with postage at I think £5 for small orders, great to deal with too I use them a lot.

www.cricklewoodelectronics.com/Cricklewood/home.php?cat=183&sort=orderby&sort_direction=0&page=10

[Deleted]

Hay Alex

You still up. hows the weather?

thanks for the reply.

unfortunately i don't have an RS account so it looks like i need to set one up.

I'm on a roll with the stupid questions so

when i built regs in the past (6080 tube based) I've tended not to let Mr power supply get connected to the audio circuit until he has been tested for good manners. i usually put a load resistor across the output just to make sure all is well and weed out any problems before they get to serious.

i intend to build the psu first (test)and then add the JLH (test again). if the PSU checks out connect to the audio circuit. i was thinking of pulling 25ma ish (load resistor) per side through the reg/JLH just to wake them up and check all is well. i like to build a bit at a time and test plenty as i go.it's just a habit I've picked up along the way

is the above going to be OK or have i missed something (as i often do).

anyhow take care and don't enjoy the sun to much

take care

[Deleted]

Nov 27, 2010 12:05:24 GMT bizzie said:

Nov 27, 2010 11:30:15 GMT @shaun said:

Hi All

from Alex's SCHA tweaks on page 1

''NOTE. BC547C, BC548C and BC549C may be used if BC550C isn't readily available.

BC557C ,BC558C and BC559C may be used if BC560C isn't readily available.''


I'm finding the bc550c / bc560c combo a bit hard to source. 550c £15 P.P from Farnell (US order).

i think that for the JLH in the classA some of the above are going to be ruled out due of their low voltage capabilities but how about the bc549c and bc559c at 30v they look possible.

I'm new to transistors so maybe I've missed something really important.

i'm here to learn so any thoughts would be much appreciated.

take care

They have them at Cricklewood at 10p each with postage at I think £5 for small orders, great to deal with too I use them a lot.

www.cricklewoodelectronics.com/Cricklewood/home.php?cat=183&sort=orderby&sort_direction=0&page=10

Thanks Bizzie i'll check that out

take care

[Will]

Hi Shaun,

Testing the PSU/JLH out as you suggest is a very good idea, and good practice.

After it's checked at 25mA, you could go to 100mA-ish or 200mA-ish to do a full load test, depending on whether you are going single or dual supply.

Somwthing to remember is that the voltage out of the JLH is also a slightly down on what's fed in if you use the CL, so may need a bit of tweaking.

[Deleted]

Hi Shaun
That sounds like a plan !
After you check out the PSU voltages, then connect it to the JLHs and do the quick checks as mentioned in the links for the JLH PCB as given by Jon/Alan. There should only be a couple of a hundred mV drop through the JLH when loaded.The output voltages from the JLH may hunt over a range of several mV until the 2,200uF capacitors are fully formed.
If the JLH checks out O.K. under a 100mA or so load, then leave it working unloaded for 48 hours or so before hooking it up to the main PCBs.
That way, after you have finished adjusting the bias and front end balance of the L and R PCBs, the HA should be close to it's best performance within minutes of switch on.
Alex

P.S. Fart sack time !

[Deleted]

Hi Will

thanks for the reply.

with me old habits (good and bad) die hard. and i just like to have things straight in my mind before getting the iron hot.

i was thinking that it could be a good way of allowing the big JLH caps to form whilst building the audio boards. I'm using some trim pots on the psu as i like to get the voltage spot on (just one of my many obsessions).

i have my TX's and most of the parts for the PSU so now I'm really cooking with gas.

take care

[Deleted]

Nov 27, 2010 12:27:01 GMT @sandyk said:

Hi Shaun
That sounds like a plan !
After you check out the PSU voltages, then connect it to the JLHs and do the quick checks as mentioned in the links for the JLH PCB as given by Jon/Alan. There should only be a couple of a hundred mV drop through the JLH when loaded.The output voltages from the JLH may hunt over a range of several mV until the 2,200uF capacitors are fully formed.
If the JLH checks out O.K. under a 100mA or so load, then leave it working unloaded for 48 hours or so before hooking it up to the main PCBs.

That way, after you have finished adjusting the bias and front end balance of the L and R PCBs, the HA should be close to it's best performance within minutes of switch on.
Alex

P.S. Fart sack time !

Hi Alex

i did read the thing about the voltage hunting and was a bit bemused by this.

just to be clear about what this means.

is it an aperiodic movement of the voltage around the required fixed output or just over voltage. i.e not frequency related but er random?

ha! I've seen lots of that type of thing with tube regs. so for the lower voltage stuff (300v ish)i usually use solid state regs as they are by comparison SOLID AS A ROCK.


sleep well

[Deleted]

Nov 27, 2010 12:39:03 GMT @shaun said:

Nov 27, 2010 12:27:01 GMT @sandyk said:

Hi Shaun
That sounds like a plan !
After you check out the PSU voltages, then connect it to the JLHs and do the quick checks as mentioned in the links for the JLH PCB as given by Jon/Alan. There should only be a couple of a hundred mV drop through the JLH when loaded.The output voltages from the JLH may hunt over a range of several mV until the 2,200uF capacitors are fully formed.
If the JLH checks out O.K. under a 100mA or so load, then leave it working unloaded for 48 hours or so before hooking it up to the main PCBs.

That way, after you have finished adjusting the bias and front end balance of the L and R PCBs, the HA should be close to it's best performance within minutes of switch on.
Alex

P.S. Fart sack time !

Hi Alex

i did read the thing about the voltage hunting and was a bit bemused by this.

just to be clear about what this means.

is it an aperiodic movement of the voltage around the required fixed output or just over voltage. i.e not frequency related but er random?

ha! I've seen lots of that type of thing with tube regs. so for the lower voltage stuff (300v ish)i usually use solid state regs as they are by comparison SOLID AS A ROCK.


sleep well

Hi Shaun
It is a slow cyclic hunting of ONLY several mV around the measured voltage as seen on a decent Digital meter or a CRO. It stops when the large electros are fully formed. It is more evident when using 2 x 4,700uF electros instead of 2 x 2,200uF electros.The larger value electros take longer to fully stabilise as you would expect.
Alex

[XTRProf]

Yeah, curious. Any idea what's the input sensitivity of this SC Class A HA?

Thanks.




Chong

[XTRProf]

Nov 27, 2010 9:32:28 GMT @solderdude said:

Nov 26, 2010 7:07:17 GMT FritzS said:

Alex,
whats recommend spare parts for
2SA1930 & 2SC5171

Which new BJT's comes close to 2SA1930 & 2SC5171 in performance?

2SB1274 / 2SD1913 (Farnell: 1790447 / 1790446 )
easier to get Sanyo transistors that will perform very well too.
Isolated TO-220ML housing.
Can also be used in Panda output stage should you want 'modern' devices that reach 100MHz.
Not that the Panda or any other amp with these output transistors will get any faster as the 'speed' of the amplifier is determined by the feedback being at a much lower frequency.

data sheet: semicon.sanyo.com/en/search/property.php?clcd=124&prod=2SD1913

the 2SA1930 & 2SC5171 reach 200MHz and have higher Vce voltages

Hi Fritz,

Overall, as Fran had described, the 2SA1930 and 2SC5171 should be better specs than the Sanyo 2SB1274 / 2SD1913. But there is no harm in trying the latter as well. If our systems are highly resolving, we may hear small (but so called audiofools or nuts ) differences.

Btw, Fran, do we need to modify a lot of this SC Class HA if the 2SB1274 / 2SD1913 are plugged in? Or it's just plug and play as all the current and voltage requirements are already met with the present resistor network setup for the 2SA1930 and 2SC5171?

Thanks.



Chong

[Deleted]

Chong
You would need to get the "R"version for best results. That may not be so easy.
Alex

: The 2SBB1274 / 2SD1913 are classified by 0.5A hFE as follows :
Rank Q R S
hFE 70 to 140 100 to 200 140 to 280

[XTRProf]

Nov 28, 2010 0:30:41 GMT @sandyk said:

Chong
You would need to get the "R"version for best results. That may not be so easy.
Alex

: The 2SBB1274 / 2SD1913 are classified by 0.5A hFE as follows :
Rank Q R S
hFE 70 to 140 100 to 200 140 to 280

Ok, thanks Alex. So plug and play with at least the R version. I think there is a S version as well with hFE 140 to 280 which is technically better than the R version hFE 100 to 200 from your description.

[Deleted]

Nov 28, 2010 0:48:39 GMT XTRProf said:

Nov 28, 2010 0:30:41 GMT @sandyk said:

Chong
You would need to get the "R"version for best results. That may not be so easy.
Alex

: The 2SBB1274 / 2SD1913 are classified by 0.5A hFE as follows :
Rank Q R S
hFE 70 to 140 100 to 200 140 to 280

Ok, thanks Alex. So plug and play with at least the R version. I think there is a S version as well with hFE 140 to 280 which is technically better than the R version hFE 100 to 200 from your description.

Chong
Those versions are usually hard to come by.
I would also be wary of the possibility of excessive gain at that location.CFP output stages are occasionally a little hard to stabilise.
I think it is best to stick with the consistency of the other Toshiba devices.

[Deleted]

The 2SB1274 / 2SD1913 transistors are drop in replacements for the original ones IF you can't easily get the ones Alex prescribed.

Farnell only sells the 'S' version b.t.w. which has the highset amplification range (140 to 280)
The 2SC5171 is specified to vary between 100 and 320 acc to their data sheet.

No need to 'try' them or change anything in the amp/schematics, just put them in IF you can't get hold of the Toshiba but can get your hands on the Sanyo devices.
The 2SB1274 / 2SD1913 have the same BW as the used BC327/BC337 (100 MHz)
having more then 100MHz bandwidth output devices is not needed nor won't do anything as it can't be driven faster by the BC3x7 in front of it anyway.
The amp itself is probably limited to around 1MHz... so the speed of the output stage is not of any importance anything above 10MHz will follow the input just as well.
No influence in the audible range or distortion amount e.t.c. between the transistors.

I was wondering.. since the amplification factor of the output transistors is top priority if, in the parts list, the amplification factor range of the BC327/337 are specified as in the schematics only BC327/BC337 is mentioned while the other BC5xx do have their amplification ranges mentioned ( -b, -c suffix).

Since the amplification of the compound output transistor is hfe BC3x7 X hfe output transistor it may be an idea to include the amplification range of the BC3x7 in the schematics as well.
The BC3x7 transistors have -16, -25, -40 gain ranges.
The BC3x7 without suffix can vary between the (not specified) -10 to -25 suffix range which is kind of 'wide' ranged.
Otherwise having closely matched output transistors is of no use as the BC3-x7 spread will completely defeat the effort of finding closely matched/spread devices ?

[Deleted]

Nov 28, 2010 8:57:54 GMT @solderdude said:

The 2SB1274 / 2SD1913 transistors are drop in replacements for the original ones IF you can't easily get the ones Alex prescribed.

Farnell only sells the 'S' version b.t.w. which has the highset amplification range (140 to 280)
The 2SC5171 is specified to vary between 100 and 320 acc to their data sheet.

No need to 'try' them or change anything in the amp/schematics, just put them in IF you can't get hold of the Toshiba but can get your hands on the Sanyo devices.
The 2SB1274 / 2SD1913 have the same BW as the used BC327/BC337 (100 MHz)
having more then 100MHz bandwidth output devices is not needed nor won't do anything as it can't be driven faster by the BC3x7 in front of it anyway.
The amp itself is probably limited to around 1MHz... so the speed of the output stage is not of any importance anything above 10MHz will follow the input just as well.
No influence in the audible range or distortion amount e.t.c. between the transistors.

I was wondering.. since the amplification factor of the output transistors is top priority if, in the parts list, the amplification factor range of the BC327/337 are specified as in the schematics only BC327/BC337 is mentioned while the other BC5xx do have their amplification ranges mentioned ( -b, -c suffix).

Since the amplification of the compound output transistor is hfe BC3x7 X hfe output transistor it may be an idea to include the amplification range of the BC3x7 in the schematics as well.
The BC3x7 transistors have -16, -25, -40 gain ranges.
The BC3x7 without suffix can vary between the (not specified) -10 to -25 suffix range which is kind of 'wide' ranged.
Otherwise having closely matched output transistors is of no use as the BC3-x7 spread will completely defeat the effort of finding closely matched/spread devices ?

Hi Frans'
I have already mentionrd to several members the desirability of matching the HFE of the BC327/BC337 to that of the O/P devices, especially with the 15W Class A amplifier , due to the often large differences in HFE between the PNP and NPN output devices.
In that scenario, a lower gain PNP output device can be helped by selecting a higher gain driver transistor on that side, than on the other side. Unfortunately, we often have to take what we can get as regards the suffix of the BC327 and BC337 . Most of the BC337/25 that I have appear to have an HFE average of around 250, although some others without the 25 or 40, have a HFE as high as 380 in my own stash.I only have 1 BC337/16, and it has an HFE of 120. BC327/BC337/40 aren't always easy to obtain, although they would be my personal choice.
This current project is actually Will's baby, although there is quite a bit of liason between us.
Will is possibly even more fanatical with matching than I am !!!
Also, there is often much more discussed in emails between interested members than appears in posts in the thread !
Alex
P.S.
Point taken though. I shall try and make time to post additional information for the benefit of the interested visitors to the thread . :-
Incidentally, Farnell is now Element 14.

[Deleted]

Farnell is still Farnell.
Actually a few years back they changed Farnell to Farnell-in-one.
Now they are back to Farnell again it seems.
Expensive for most parts, cheap for some parts and a bit difficult to order from as an individual but possible.
They have a huge stock.

Element 14 is a community/Forum for engineers and is linked with Farnell

Maybe in the schematics/BOM just mentioning BC3x7/25 is a good idea.
Don't know about what's listed in the parts list (didn't look for it)

[Deleted]

Nov 28, 2010 10:08:46 GMT @solderdude said:

Farnell is still Farnell.
Actually a few years back they changed Farnell to Farnell-in-one.
Now they are back to Farnell again it seems.
Expensive for most parts, cheap for some parts and a bit difficult to order from as an individual but possible.
They have a huge stock.

Element 14 is a community/Forum for engineers and is linked with Farnell

Maybe in the schematics/BOM just mentioning BC3x7/25 is a good idea.
Don't know about what's listed in the parts list (didn't look for it)

Frans
In Australia, Farnell recently changed their name to Element 14. As a customer of theirs I receive regular updates, including the name change
Alex

element14 Australia | world-leading distributor of electronic and ...element14 Australia is a world-leading marketer and distributor of electronic and maintenance, repair and operations (MRO) products.
au.element14.com/ - Cached

[Deleted]

So it seems for Australia, China, Hong-Kong, India, Korea, Malaysia, New zealand, Philippines, Singapore, Taiwan and Thailand Farnell 'Asia-Pacific' is now called element 14.

All other countries it's still named Farnell and element 14 is only a community/forum.

didn't know that as I only look in the Dutch and UK Farnell site.

[FritzS]

On Digikey I found this:
search.digikey.com/scripts/DkSearch/dksus.dll?vendor=0&keywords=2SA1930
search.digikey.com/scripts/DkSearch/dksus.dll?vendor=0&keywords=2SC5171

search.digikey.com/scripts/DkSearch/dksus.dll?vendor=0&keywords=2SB1274
search.digikey.com/scripts/DkSearch/dksus.dll?vendor=0&keywords=2SD1913

[Will]

Hi Frans'I have already mentionrd to several members the desirability of matching the HFE of the BC327/BC337 to that of the O/P devices, especially with the 15W Class A amplifier , due to the often large differences in HFE between the PNP and NPN output devices.In that scenario, a lower gain PNP output device can be helped by selecting a higher gain driver transistor on that side, than on the other side. Unfortunately, we often have to take what we can get as regards the suffix of the BC327 and BC337 . Most of the BC337/25 that I have appear to have an HFE average of around 250, although some others without the 25 or 40, have a HFE as high as 380 in my own stash.I only have 1 BC337/16, and it has an HFE of 120. BC327/BC337/40 aren't always easy to obtain, although they would be my personal choice.This current project is actually Will's baby, although there is quite a bit of liason between us.Will is possibly even more fanatical with matching than I am !!!Also, there is often much more discussed in emails between interested members than appears in posts in the thread !Alex P.S. Point taken though. I shall try and make time to post additional information for the benefit of the interested visitors to the thread . br]Incidentally, Farnell is now Element 14.

This is pretty much what I found when I bought my bag of 327/337. Using the advice above. I matched the toshibas parts to a similat hFE, and the 327/337 to a similar range. I can't remember the values know.

It may be better to buy the bc327/337-25, but looking at the spec in places like farnell, some 3xx-25 can sometimes be listed as having greater hFE, so looking at the range a 3xx-25 part is in (which can also be dependant on supplier it seems) and then comparing the part to be bought is the best way forward. The BOM link posted has always had a link to a ON semi 3xx-25 part.


Nearly ready to order the PCB's, just waiting for one more payment. ***coughchongcough***

[Will]

Will is possibly even more fanatical with matching than I am !!!

And for a good reason

I love DIY because you build something that you could either not buy, or if you could, you would never be able to afford it.

Matching all the components (well, OK not the capacitors, I don't have a meter) is something that I feel would be financially un-do-able commercially, as no company would be able to take the time to do it, unless the buyer understood that was part of the (huge) price. As a DIYer, our time is free, so we can take this time to make the things we build as good as possible.

[Deleted]

Farnell specs cannot be trusted.
They are merely there as an indication.
For the real specs you should always check the manufacturers datasheet.
Also always check what they sent... experience learns they just as easily deliver 'alternatives'.

[Deleted]

Nov 28, 2010 14:02:40 GMT Will said:

Will is possibly even more fanatical with matching than I am !!!

And for a good reason

I love DIY because you build something that you could either not buy, or if you could, you would never be able to afford it.

Matching all the components (well, OK not the capacitors, I don't have a meter) is something that I feel would be financially un-do-able commercially, as no company would be able to take the time to do it, unless the buyer understood that was part of the (huge) price. As a DIYer, our time is free, so we can take this time to make the things we build as good as possible.

Hi Will

That sums up why i became interested in DIY very nicely.


Hi all

i like to accompany my builds with a good book. and being new to this solid state thing i was wondering if anyone could suggest a good electronics book. i was thinking of one of the JLH books might be quite topical but I've not really read his work and don't want to end in up over my head.

in valve terms I've found the Morgan Jones books great , easy to follow with a good balance between diy, maths ans electronics

so if anyone has read any of the JLH books a nod in the right direction would be grrrrreat .

any (clean) idea's welcome

take care