Hi,
Arcam Alpha amp:-
Preamp stage -47K input with 1uf electrolytic input NE5532 / NE5534 with 7815/7915 PSU
Power amp stage:
- BC547 diffrential input, MPSA56 stage
- FST239 / TIP3055
- FST240 / TIP2955
PSU
-1N5403 rectifiers
-torroidal tx (upgraded by me)
Sounds quite nice but seems to take a bit of time (about 40mins) to warm up. Not sure if it's the amp or my cd player !
I don't think there's much I can do to the power section but would it be worthwhile improving the preamp section ?
Arcam Alpha amp:-
Preamp stage -47K input with 1uf electrolytic input NE5532 / NE5534 with 7815/7915 PSU
Power amp stage:
- BC547 diffrential input, MPSA56 stage
- FST239 / TIP3055
- FST240 / TIP2955
PSU
-1N5403 rectifiers
-torroidal tx (upgraded by me)
Sounds quite nice but seems to take a bit of time (about 40mins) to warm up. Not sure if it's the amp or my cd player !
I don't think there's much I can do to the power section but would it be worthwhile improving the preamp section ?
regulate the power supply. not many improvments come close to regulating a preamp. i just did it for my nad 302. just wonderful!
I'd replace the 7815/7915 with LM317/337. Be aware these are not drop-in replacements.
You could try replacing the 1uF electrolytic input caps with some MKS2 polyester parts if they'll fit.
By "warm up", do you mean for it to sound at it's best, or for it to turn on?
You could try replacing the 1uF electrolytic input caps with some MKS2 polyester parts if they'll fit.
By "warm up", do you mean for it to sound at it's best, or for it to turn on?
jaycee said:
To sound it's best it needs some time. It's not that nice to listen to for the first 40mins, and then seems to come alive.
As for the electrolytic caps in the signal path if I go smaller than the schematic I will lose the bass ? (in the schematic 1uf, and 10uf coupling caps are used)
LM317 / LM337 circuit shouldn't be too hard to build on a little board. I probably have enough parts lying around here! +33 /-33V at input of the regulators on the schematic. I have some LT1806, and LT1085 regulators but no negative versions!
Looking at the preamp section there seems to be some kind of FET circuit used after the op amps.
Yeah, a lower capacitance will raise the lower cutoff frequency.
The issue with taking 40 mins to sound it's best could be related to the bias current. If you have the service manual maybe you should consider re-adjusting this. It might also be a good idea to replace the bias adjust trimmers.
The issue with taking 40 mins to sound it's best could be related to the bias current. If you have the service manual maybe you should consider re-adjusting this. It might also be a good idea to replace the bias adjust trimmers.
I do not know the Arcam Alpha so my remarks must be general.
First, if I understand correctly, then with an input impedance of 47K and a 1 uF capacitor, the -3 dB frequency is only 3.4 Hz! That means that you could lower the value to 220 nF for a 15 Hz low frequency turning point, which places it well within obtainable polyester or polycarbonate capacitors.
But that should not affect your main problem. Here I would agree with Jaycee that it sounds like a bias problem in the sense of warm-up time. You should check there with a multimeter, if you have an idea what it is supposed to be. (You do not mention if this is a problem that developed recently or whether it was always like that.) But it might not be as easily solved as just increasing the bias current. You might also have to check whatever temperature compensating device there is (sensing thermistor/transistor on the heat sink?), and in the end need a circuit diagram with specifications.
As a suggestion, I never switch my own amplifier off. My previous amp remained on for 14 years, barring the odd power interruption (it did not give up after that, it was stolen). The amount of power consumed is negligible and it gives things time to really stabilise.
Of course, if you also suspect your CD player, you should determine whether that is OK.......
Regards.
First, if I understand correctly, then with an input impedance of 47K and a 1 uF capacitor, the -3 dB frequency is only 3.4 Hz! That means that you could lower the value to 220 nF for a 15 Hz low frequency turning point, which places it well within obtainable polyester or polycarbonate capacitors.
But that should not affect your main problem. Here I would agree with Jaycee that it sounds like a bias problem in the sense of warm-up time. You should check there with a multimeter, if you have an idea what it is supposed to be. (You do not mention if this is a problem that developed recently or whether it was always like that.) But it might not be as easily solved as just increasing the bias current. You might also have to check whatever temperature compensating device there is (sensing thermistor/transistor on the heat sink?), and in the end need a circuit diagram with specifications.
As a suggestion, I never switch my own amplifier off. My previous amp remained on for 14 years, barring the odd power interruption (it did not give up after that, it was stolen). The amount of power consumed is negligible and it gives things time to really stabilise.
Of course, if you also suspect your CD player, you should determine whether that is OK.......
Regards.
Ashley, email Arcam for a service manual. They are very helpful, and a full schematic will make modding so much easier.
pinkmouse said:
Arcam have sent me the service manual! Fantastic service 🙂
I have the bias figures so I will take a look but I have been working recently on my digital stuff (QED digit dac) . The manual details the bias adjust procedure in full.
I bought it as faulty (dead transformer with paperclip connections
) so it would be a modding project but I've been messing around with digital for ages so haven't looked at the Arcam one bit.
I would imagine the pre-amp is for vinyl records only and the line level inputs go straight to the volume control and power amp. So modding the pre-amp will have no influence on the sound. Unless of course the amp has tone controls and you select them to be in circuit.
Check bias setting for your warm up issue.
Check bias setting for your warm up issue.
richie00boy said:
Just checked the service manual.
-CD input into single NE5532 amplifier
-this then feeds a NE5534 (one per channel) with tone control section.
-then a balance section
-then a mute Jfet section
When you come from another place, 40 minutes is the time that your hearing takes to get used to the acoustics of the room and the frequency response of the speakers.
Fortunately, our eyes are faster when it comes to accomodate to changes in ambient light intensity and dominant colours.
Fortunately, our eyes are faster when it comes to accomodate to changes in ambient light intensity and dominant colours.
Now there is a deep thought!
One seldom considers the characteristics of the hearing faculty: Like our other senses self-compensting, etc. (..... and not desiring to go off-thread, nor open this very subjective topic!)
Can it take that long, Eva? (Asking, not disagreeing) Have you any references to this kind of acousto-medical research? I have often read indirect reference but have not been able to trace any direct research reports - I believe the Scandinavian folks have done some work here.
Regards.
One seldom considers the characteristics of the hearing faculty: Like our other senses self-compensting, etc. (..... and not desiring to go off-thread, nor open this very subjective topic!)
Can it take that long, Eva? (Asking, not disagreeing) Have you any references to this kind of acousto-medical research? I have often read indirect reference but have not been able to trace any direct research reports - I believe the Scandinavian folks have done some work here.
Regards.
Hi,
I have replaced the preamp coupling capacitors with rubycon zl 100uf caps, and switched the tone op amps to OPA627. Bass is much improved.
Now need to look at the NE5532 input stage.
I have replaced the preamp coupling capacitors with rubycon zl 100uf caps, and switched the tone op amps to OPA627. Bass is much improved.
Now need to look at the NE5532 input stage.
Change TIP2955 / TIP3055 for 2SC5200 / 2SA1943
Hi,
Can I drop in 2SC5200 / 2SA1943 transistors to 'improve' the amplifier ? (I don't need the extra power just improved HF behaviour).
I've looked at the driver stage, and it uses FS239 / FS240. I have no datasheet for the FST part but a google group post suggests it's a ZTX653 / 753 driver. http://groups.google.com/group/sci....tronics.components/browse_frm/month/2001-10?&
TIP3055 SC5200
CE voltage 60V 230V
CB voltage 100V 230V
EB voltage 7V 5V
C current (DC) 1.5A 15A
B current ( DC) 7A 1.5A
Power diss(W) 90W 150W
Hfe 20-70 (4A IC 4v CE) 55-160 (1A IC 5v CE)
Transition fs 2.5Mhz 30Mhz
C Capacitance 200pf
Hopefully I have correctly taken the figures from the datasheets for comparison. I'm a little worried about the Hfe for the potential replacement !
Hi,
Can I drop in 2SC5200 / 2SA1943 transistors to 'improve' the amplifier ? (I don't need the extra power just improved HF behaviour).
I've looked at the driver stage, and it uses FS239 / FS240. I have no datasheet for the FST part but a google group post suggests it's a ZTX653 / 753 driver. http://groups.google.com/group/sci....tronics.components/browse_frm/month/2001-10?&
TIP3055 SC5200
CE voltage 60V 230V
CB voltage 100V 230V
EB voltage 7V 5V
C current (DC) 1.5A 15A
B current ( DC) 7A 1.5A
Power diss(W) 90W 150W
Hfe 20-70 (4A IC 4v CE) 55-160 (1A IC 5v CE)
Transition fs 2.5Mhz 30Mhz
C Capacitance 200pf
Hopefully I have correctly taken the figures from the datasheets for comparison. I'm a little worried about the Hfe for the potential replacement !
This is very unlikely to work as you hope, you will likely need to redo the compensation and possibly the feedback network, not a simple job unless you know what you are doing and have access to decent test equipment.
pinkmouse said:
Hi,
Unfortunately I don't have any test equipment, and I don't have the spice models. Nor would I have much confidence in the model results anyway!
Would it be an idea to increase heatsink size, replace the bias pot's, and increase the bias (currently 30ma) ? See http://www.tnt-audio.com/clinica/bias_e.html
To be completely honest, I wouldn't bother. With the money you would spend on upgrading, why don't you think about building a completely new amp? Something simple like Rod's P3a or MikeB's Symasym shouldn't be expensive and will likely provide a world of improvement over your Alpha. 😉
Replacing pots with nice multiturn cermet ones is not a bad idea. As for increasing the bias, I'm not for this mod. The bias should be set to give best distortion performance, if you increase it above this level you suffer from gm doubling. Doug Self has documented this well if you want to read more into it.
Hi there,
I own this amp also, but not your knowledge...
So can you tell me wich opamp can be changed and some other easy and real improvment ?
Thanks
I own this amp also, but not your knowledge...
So can you tell me wich opamp can be changed and some other easy and real improvment ?
Thanks
transistors
Ashley,
The 2955/3055 are slow transistors, it could be that the designer didn't rely on the slowness and has compensated the amp well enough to work with much faster transistors, but maybe not. There are a whole host of better parts out there that are only a little faster than the originals, but offer better gain, linearity, power dissipation etc.
Take a look at the mj15003/4 or better yet the mj21193/4 they are quite a lot superior in most ways but only twice as fast, minimizing the risk of gross instability.
I recently replaced the outputs of a NAD 4030 receiver that had been overloaded, the mj15003/4 sounded better and have twice the power handling.
edit: they aren't using the to3 versions of the 2955/3055 transistors, so you'd want to use the mjl21193/4 which are night and day better than the originals, but in a slightly larger package.
Stuart
Ashley,
The 2955/3055 are slow transistors, it could be that the designer didn't rely on the slowness and has compensated the amp well enough to work with much faster transistors, but maybe not. There are a whole host of better parts out there that are only a little faster than the originals, but offer better gain, linearity, power dissipation etc.
Take a look at the mj15003/4 or better yet the mj21193/4 they are quite a lot superior in most ways but only twice as fast, minimizing the risk of gross instability.
I recently replaced the outputs of a NAD 4030 receiver that had been overloaded, the mj15003/4 sounded better and have twice the power handling.
edit: they aren't using the to3 versions of the 2955/3055 transistors, so you'd want to use the mjl21193/4 which are night and day better than the originals, but in a slightly larger package.
Stuart
- Status
- Not open for further replies.