I have a friend who has built 2 of the studio 350 amps featured in Jan/Feb 2004 Silicon Chip magazine. They both have the same bias problem, where the output is at rail volts. All components have been double-checked, esp. transistor pinouts. The only way I could get the output voltage close to zero was to reduce the value of resistors in Q4 & Q5 collectors. This causes high currents, & tends to overheat these transistors.
Has anyone had a similar problem with this amp?
Howard
Has anyone had a similar problem with this amp?
Howard
I have had the same problem. I also note that another user (Silicon Chip issue 194, Nov 04) encountered this problem. Silicon Chip advised him:
"Our guess is that you have swapped a pair of transistors or you have an open-circuit solder connection somewhere. Replace the fried resistors and power up the board again, with the resistors across the fuseholders, and check each transistor for a base-emitter voltage of about 0.7V. An incorrect reading indicates a fault in the transistor or its associated components".
In my case, I have exhaustively checked component positioning. I have also checked all resistors and most other components by multimeter. I have checked all solder joints. In the end I replaced diodes D1, D2, D3 and transistors Q1- Q9. Didn't make any difference. In my case, the problem is focussed on 3 voltages which are hopelessly out of spec:
- across 6.8k 1W resistor near base of Q6 (123V, should be 60V)
- +70V to collector of Q6 (132V, should be 69V)
- -70V to collector of Q5 (4V, should be 69V)
"Our guess is that you have swapped a pair of transistors or you have an open-circuit solder connection somewhere. Replace the fried resistors and power up the board again, with the resistors across the fuseholders, and check each transistor for a base-emitter voltage of about 0.7V. An incorrect reading indicates a fault in the transistor or its associated components".
In my case, I have exhaustively checked component positioning. I have also checked all resistors and most other components by multimeter. I have checked all solder joints. In the end I replaced diodes D1, D2, D3 and transistors Q1- Q9. Didn't make any difference. In my case, the problem is focussed on 3 voltages which are hopelessly out of spec:
- across 6.8k 1W resistor near base of Q6 (123V, should be 60V)
- +70V to collector of Q6 (132V, should be 69V)
- -70V to collector of Q5 (4V, should be 69V)
Hi all
Here are some notes and errata from the Silicon Chip web site.
Studio 350 Power Amplifier Module, January 2004: The 470ìF 100V electrolytic capacitor connected to the -70V rail (adjacent to fuse F2) is shown reversed on the circuit diagram (Fig.7). (02/04)
Studio 350 Power Amplifier Module, February 2004: The ±70V power supply wiring colours shown on the overlay diagram (Fig.1) disagrees with the wiring colours shown on the power supply wiring diagram (Fig.6). In all cases, the +70V wiring should be red and the -70V wiring blue. The ±70V markings on the PC board and diagrams are correct. The wiring diagram for the Silicon Chip Online website is correct. (04/04)
Hope this is of some help
Cheers
Rob
Here are some notes and errata from the Silicon Chip web site.
Studio 350 Power Amplifier Module, January 2004: The 470ìF 100V electrolytic capacitor connected to the -70V rail (adjacent to fuse F2) is shown reversed on the circuit diagram (Fig.7). (02/04)
Studio 350 Power Amplifier Module, February 2004: The ±70V power supply wiring colours shown on the overlay diagram (Fig.1) disagrees with the wiring colours shown on the power supply wiring diagram (Fig.6). In all cases, the +70V wiring should be red and the -70V wiring blue. The ±70V markings on the PC board and diagrams are correct. The wiring diagram for the Silicon Chip Online website is correct. (04/04)
Hope this is of some help
Cheers
Rob
Hi guy!
Hey I got a new PCB layout for studio 350 power amp and I just wondering if the polarity of the cap close to F2 is correct before I order a couple of boards?
Hey I got a new PCB layout for studio 350 power amp and I just wondering if the polarity of the cap close to F2 is correct before I order a couple of boards?
An externally hosted image should be here but it was not working when we last tested it.
Hola
The capacitor orientation is correct. The (+) must be in the lower side...
Last edited:
The Altronics 500VA transformers state "Note: Must be used with a 7.5A external AC fuse." I imagine those from Jaycar would be similar.
Both their catalogue and website are not very helpful there.
SandyK
Silicon Chip 350W amplifier.
Thanks for that. The amplifier I am actually building is the L20 V9 which I purchased on Ebay. For a bit of amusement see my posting under L20. It actually sounds pretty good. Now I have the toroids I hope I don't have any more issues. Still, if it has been working on +/-63V I suppose it can survive +/-70V.
Thanks for that. The amplifier I am actually building is the L20 V9 which I purchased on Ebay. For a bit of amusement see my posting under L20. It actually sounds pretty good. Now I have the toroids I hope I don't have any more issues. Still, if it has been working on +/-63V I suppose it can survive +/-70V.
Famous last words
I'm sure there are good reasons for rating the supply as it is, since every supplier wants to offer the highest power per dollar they can get away with. If you would like assurances, email the designer who is a forum member too.
There is a great deal of difference in the power output possible with a total 14V or ~12% supply increase. Think this, the component dissipation and voltage ratings through first.
I'm sure there are good reasons for rating the supply as it is, since every supplier wants to offer the highest power per dollar they can get away with. If you would like assurances, email the designer who is a forum member too.
Silicon Chip 350W amplifier.
"Famous last words", how often have I heard that! Thanks for the post. I will never need to use the full output of the L20's so at least I won't stress the power supply. But I'm sure some other problems will come out of the woodwork before long
. The seller of the L20's and the designer have been helpful but it's like drawing teeth sometimes. Still I suppose they are busy answering questions from other mugs like me. Regards.
"Famous last words", how often have I heard that! Thanks for the post. I will never need to use the full output of the L20's so at least I won't stress the power supply. But I'm sure some other problems will come out of the woodwork before long
. The seller of the L20's and the designer have been helpful but it's like drawing teeth sometimes. Still I suppose they are busy answering questions from other mugs like me. Regards.
The problem is really a safety issue.
If, for any reason, the signal level (and that should include accidental noise we experimenters are very likely to make when faffing about) exceeds the input maximum level, the amplifier will not just clip at the previous safe level but take everything to a higher level.
Sure, I've used amps on higher supplies quite safely but I have also stupidly done so with other people's gear even, purely for the same convenience reason but due to noise from bad connections, blew the whole output stage. It cost me more than the repair was worth too. Though a Mosfet output stage is likely more robust, the input/VAS transistors at least should have suitably high ratings. Then, there are heatsink and resistor ratings, cap and semi voltages etc.
You need cap ratings at least 15% above the nominal voltage with modern parts and resistors are rated at much higher temperatures than you would like to touch so plenty of latitude is the order of the day if you expect reasonable life, avoiding burns. and damaged boards.
I responded here because of the scepticism expressed in the thread discussing the design as it was presented. True, most people never use more than 10% of the capability of their gear but there are several unintentional ways to greatly exceed it. I confess I haven't built an L20 but I think it runs close to the edge already, with insufficient output pairs for the power rating so my inclination is to take it very carefully. and limit rather than overkill on the power supply.
As you can't get a transformer from the retailers, why not get a local product from Tortech. or Harbuch? They are not that much more expensive in single quantity than the Chinese imports in large quantity.
If, for any reason, the signal level (and that should include accidental noise we experimenters are very likely to make when faffing about) exceeds the input maximum level, the amplifier will not just clip at the previous safe level but take everything to a higher level.
Sure, I've used amps on higher supplies quite safely but I have also stupidly done so with other people's gear even, purely for the same convenience reason but due to noise from bad connections, blew the whole output stage. It cost me more than the repair was worth too. Though a Mosfet output stage is likely more robust, the input/VAS transistors at least should have suitably high ratings. Then, there are heatsink and resistor ratings, cap and semi voltages etc.
You need cap ratings at least 15% above the nominal voltage with modern parts and resistors are rated at much higher temperatures than you would like to touch so plenty of latitude is the order of the day if you expect reasonable life, avoiding burns. and damaged boards.
I responded here because of the scepticism expressed in the thread discussing the design as it was presented. True, most people never use more than 10% of the capability of their gear but there are several unintentional ways to greatly exceed it. I confess I haven't built an L20 but I think it runs close to the edge already, with insufficient output pairs for the power rating so my inclination is to take it very carefully. and limit rather than overkill on the power supply.
As you can't get a transformer from the retailers, why not get a local product from Tortech. or Harbuch? They are not that much more expensive in single quantity than the Chinese imports in large quantity.
I have in fact purchased 2 x 500VA toroids for this project and they will give me +/-70V in lieu of the temporary single transformer I've been using which gives +/-63V. The PS electros are rated at 80V so should be OK. I'm not so sure about the on board discrete rectifiers though and may fit 35A block rectifiers. These amps are being constructed as two separate monoblocks so everything on the temporary lash up is duplicated. Each rail to each amp has 2A fast blow fuses at present but in the final version will be fitted with 5A slow blow fuses. I have a single 'stereo' speaker protection board in the temporary build and will have one in each of the final monoblocks. I have never connected this amp up to my 'expensive' speakers before I was satisfied that it was working OK and I am paranoid about not turning it on until all connections are made. In any case I will probably sell the whole lot when it's finished as so far I much prefer the sound of my 28W 845B monoblocks! Building solid state amps keeps me sane as I chase that 'valve sound'. I don't think I will ever get it with silicon. 
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.