Hi,
I have decided to try to decipher the Alchemist Forseti Power Amp ADP20, from some photos found in the net 😉. Below I attatch the schematic I have made. I would appreciate any help in finding any mistakes. Originally it worked in class AB it had 153W in to 8ohms, i don't know if the power voltage 36V that i had counted is OK. So I need your help for this project 🙂.
I have decided to try to decipher the Alchemist Forseti Power Amp ADP20, from some photos found in the net 😉. Below I attatch the schematic I have made. I would appreciate any help in finding any mistakes. Originally it worked in class AB it had 153W in to 8ohms, i don't know if the power voltage 36V that i had counted is OK. So I need your help for this project 🙂.
Attachments
Hello bassil
I have the same amplifier : ADP20 - MK2
I have now replaced all the output drivers , and need to tune the bias current .
Do you have any idea , how should i do it ?
Thanks
I have the same amplifier : ADP20 - MK2
I have now replaced all the output drivers , and need to tune the bias current .
Do you have any idea , how should i do it ?
Thanks
For reference:
First, thanks for the schematic
However I think T13, T15 and T17 should have their base connections similar to T11.
A bit odd design regarding the driverstage, since it seems to be single ended. Very long time ago since I worked with those produces.
If I recall correct supply voltage is around +/- 55VDC, I dont have any figures of quiescent current - but pretty high. Approx 50mA for each transistor. Best place to measure is over a resistor in the PSU separating different capacitorbanks. 0.47ohm 5W or something like that I think it is.
First, thanks for the schematic
However I think T13, T15 and T17 should have their base connections similar to T11.
A bit odd design regarding the driverstage, since it seems to be single ended. Very long time ago since I worked with those produces.
If I recall correct supply voltage is around +/- 55VDC, I dont have any figures of quiescent current - but pretty high. Approx 50mA for each transistor. Best place to measure is over a resistor in the PSU separating different capacitorbanks. 0.47ohm 5W or something like that I think it is.
Hi Basill.
Time to ressurect this thread.
Enkie is correct with the correction. The bias protect diodes are also back to front.
I have a Forseti AP15A in for repair so this diagram is useful (very similar).
The 15 does use 55Volt rails but is slightly lower power.
Time to ressurect this thread.
Enkie is correct with the correction. The bias protect diodes are also back to front.
I have a Forseti AP15A in for repair so this diagram is useful (very similar).
The 15 does use 55Volt rails but is slightly lower power.
there is also a basic schematic in their web page which is comfortable with all models i think .
If information about the principal of operation ( which is slightly different than any other amplifier ) and bias information is needed please leave a note here
Kind regards
Sakis
If information about the principal of operation ( which is slightly different than any other amplifier ) and bias information is needed please leave a note here
Kind regards
Sakis
Can I please ask for it. Euxariste. 🙂
I was asked to fix such an amp with relatively high whiss during normal operation (hearable at ~40cm from the tweeter!) and loud buzz when turning off. Generating...
Here is a link to the general Forseti schematic at the Alchemist website, referred to by Sakis. Specifically, this is APD15 but mainly only varies from other models in the number of output devices (Output transistor + driver sets) according with each model.
http://www.alchemisthifi.info/ranges/forseti/documents/APD15-Forseti-schematic-1500x935.gif
http://www.alchemisthifi.info/ranges/forseti/documents/APD15-Forseti-schematic-1500x935.gif
Thank you Ian, however mine is quite different. I'm not banned at google (grin) and have seen this initial schematic, but am so lucky to have encountered a "tdp signature series" device, with at least 14 transistors in the driver stage of each amp.
The output stage is generally the same, but I'm pretty sure that it's not the source of this noise.
Will recap it anyway, there are some junk capxons in there.
Noise like that should be a transistor failure small ones will be the suspects trace it with heat or freeze ..
then it could also be a soldering issue or capacitor issue which is more rare ..
Regards
Sakis
then it could also be a soldering issue or capacitor issue which is more rare ..
Regards
Sakis
As Alchemist went bust in 2001, all Alchemist amplifiers should be in need of new caps at this age, especially as the originals were not the best available. I just re capped my 2 power amps and Forseti pre (all TDP versions), with mostly top line Vishay and a few Nichicon FG. I am enjoying the benefits. These amps are really something special 🙂
Hi there,
I hope somebody might be able to advise me.
I am servicing my own Alchemist Forseti power amplifier and it shows a wearied problem.
It switches ON after a time delay (which is correct) but then one of the channels switches OFF then ON again.
This might happen to either channel and sometimes it stops once the amp is warm.
I believe it has something to-do around an IC4098 IC, however the ICs are OK (I changed them on both channels)
I also replace the Tantal elcos which determines the time constant, same same.
Does anyone have an idea what exactly the function of this chip is,
I doubt that the designers used such a complex design for just a power on delay.
Cheers
Bjoern
I hope somebody might be able to advise me.
I am servicing my own Alchemist Forseti power amplifier and it shows a wearied problem.
It switches ON after a time delay (which is correct) but then one of the channels switches OFF then ON again.
This might happen to either channel and sometimes it stops once the amp is warm.
I believe it has something to-do around an IC4098 IC, however the ICs are OK (I changed them on both channels)
I also replace the Tantal elcos which determines the time constant, same same.
Does anyone have an idea what exactly the function of this chip is,
I doubt that the designers used such a complex design for just a power on delay.
Cheers
Bjoern
you need to determine if the fault is real or not
it might be something wrong with the protection circuit so protection clicks with no obvious reason
or the fault is real and both of your amps present excessive offset at some point and trip the protection
so observe offset values before the protection at the time of fault
CFP amplifiers with that type of design and that amount of active parts in the output often are a nightmare to stabilize...
Remember that depending on the period of manufacture , available parts . serial number and so on Alchemist people installed various patches in the circuit to keep it stable
yet again depending in the all above Alchemist people might missed the patches depending on availability ....
Paravichini is totally responsible for that because he knew that the amp had stability issues and he only focus on the exterior design and the power supply after hired from Alchemist to finish somebody else's project ....
The total stability problem forced them to use locomotive slow transistors like 3055 and some times a bit faster like 1047-817 but still with dead slow drivers like Tip 31-41 and so on and still full of patches to keep it stable well known as "sonic killers " ...All this was done to avoid oscillations...
i had an idea on how to solve that and eventually go with much faster transistors like 200 mhz for drivers and at least 30 for outputs and without patches ....So i had a chat with Tim about that issue but he didn't care much about it ...He only wanted my money ...
He said about it : ""if you want to hire me for consulting i am available """
Imagine i have already repaired 30-40 of his mistakes and he was offering his services to me ...No way Hose ....
in reality i never spend the time to find out if my idea is working properly ...Truth is if u manage to make one Alchemist with fast transistors and no patches to keep it stable, then you will have a world class unbeatable machine .....
it might be something wrong with the protection circuit so protection clicks with no obvious reason
or the fault is real and both of your amps present excessive offset at some point and trip the protection
so observe offset values before the protection at the time of fault
CFP amplifiers with that type of design and that amount of active parts in the output often are a nightmare to stabilize...
Remember that depending on the period of manufacture , available parts . serial number and so on Alchemist people installed various patches in the circuit to keep it stable
yet again depending in the all above Alchemist people might missed the patches depending on availability ....
Paravichini is totally responsible for that because he knew that the amp had stability issues and he only focus on the exterior design and the power supply after hired from Alchemist to finish somebody else's project ....
The total stability problem forced them to use locomotive slow transistors like 3055 and some times a bit faster like 1047-817 but still with dead slow drivers like Tip 31-41 and so on and still full of patches to keep it stable well known as "sonic killers " ...All this was done to avoid oscillations...
i had an idea on how to solve that and eventually go with much faster transistors like 200 mhz for drivers and at least 30 for outputs and without patches ....So i had a chat with Tim about that issue but he didn't care much about it ...He only wanted my money ...
He said about it : ""if you want to hire me for consulting i am available """
Imagine i have already repaired 30-40 of his mistakes and he was offering his services to me ...No way Hose ....
in reality i never spend the time to find out if my idea is working properly ...Truth is if u manage to make one Alchemist with fast transistors and no patches to keep it stable, then you will have a world class unbeatable machine .....
I believe its nothing to do with the amp circuit as such but more with the CD4098 chip. For instance originally it had 4.7uF Tantalum capacitor at pin 2 to set the timing. I replaced that with a normal electrolytic of the same value but that didnt work at all the time never triggered. I dont quite have a clue why an electrolytic wont work .
Then I changed back to a Tantalum still had spontaneous switching, yesterday I added a 100k ohm resistor to the 1.7M ohm existing one in series with the cap and so far it seems to work. I have no idea why that circuit is so sensitive to the RxCx combination.
Then I changed back to a Tantalum still had spontaneous switching, yesterday I added a 100k ohm resistor to the 1.7M ohm existing one in series with the cap and so far it seems to work. I have no idea why that circuit is so sensitive to the RxCx combination.
Historically, Tantalum is lower-leakage than Aluminum electrolytic. This matters most for long timing networks.
The Al cap industry worked hard to take this corner of the market. A modern Low Leakage Al electro will normally replace Tantalum. A normal Al electro may not.
The Al cap industry worked hard to take this corner of the market. A modern Low Leakage Al electro will normally replace Tantalum. A normal Al electro may not.