Sure, this is the problem. Just inspect the schematic and you will find an uggly ground loop through an AC coil. Also, there must be four separate connections at the secondarry side. Avoid center-tap configuration with just three wires. Left and right channel should touch in a single point at RCA connectors, not in transformer. Also use shielded wires at the input. I use LiYCY cable.
Good luck!
Last edited:
Be carefull with A60+ version. I have found a possible problem with the larger version: speaker protection seems to malfunction. Could be a bad component, but it doesn't respond properly to a sudden large pulse signal. It doesn't cut the output. Fortunately, I test just with dummy load. For the time being I am using a separate speaker protection board untill I rectify the malfunction.
What kind of board are you using A60+? In my Chinese one everything is fine. I'm waiting for your feedback. Absolutely right. The channels must be completely separated, connected to the minus RCA input through the case with a 10 ohm resistor and nowhere else. It has already been explained here many times.
Hello Berlusconi. Board version is actually V1.1 ++. I am also using an off board speaker protector circuit. The chip used for the on board circuit does not seem to be readily available. In more thorough testing I have uncovered a stability issue where the amp will ring terribly at 1mhz on a 1khz square wave when driving a 4 ohm load with as low as .015uf capacitance across the load resistor. It will drive a resistive load with a perfect square wave. I also have issues with parts layout in the chassis and things have gotten so congested in there that I have decided to redesign the layout of the PS and other circuits to get a better ground structure and a more servicable layout. Have you seen this sort of instability issue?
I have, among others, A60+ V1.1. It appears that upc1237 chip is faulty, but I have several chips left for repair. I hope this will solve the problem. Also, on 60+ there is a 47uF. Perhaps this is a problem too.
Regarding the connection at the RCA terminals I have an elegant explanation. I will get back to this later on.
Guys, I have just ordered from Mouser 30 pairs of Toshiba 2SA1587/2SC4117 to substitute 2SA970/2SC2240. I will report after testing.
The reason is that the existing transistors are CDIL products and they do not list all important parameters, and also do not guarantee . Toshiba has listed all parameters and are identical with the original transistors. The only difference is that new transistors are SMD, but I have adaptors.
The reason is that the existing transistors are CDIL products and they do not list all important parameters, and also do not guarantee . Toshiba has listed all parameters and are identical with the original transistors. The only difference is that new transistors are SMD, but I have adaptors.
| 2SA970 | 2SA1587 | 2SC2240 | 2SC4117 | ||
| VCBO | -120 | -120 | 120 | 120 | |
| VCEO | -120 | -120 | 120 | 120 | |
| VEBO | -5 | -5 | 5 | 5 | |
| IC | -100 | -100 | 100 | 100 | |
| IB | 20 | -20 | 20 | 20 | |
| PC | 300 | 200 | 300 | 200 | |
| ICBO | -0,1 | -0,1 | 0,1 | 0,1 | |
| IEBO | -0,1 | -0,1 | 0,1 | 0,1 | |
| VBR | -120 | 120 | |||
| HFE | 200-700 | 200-700 | 200-700 | 200-700 | |
| VCESAT | -0,3 | -0,3 | 0,3 | 0,3 | |
| VBE | 0,65 | 0,65 | |||
| FT | 100 | 100 | 100 | 100 | |
| COB | 4 | 4 | 3 | 3 | |
| NF | 3…6 | 1…10 | 2…6 | 1…10 |
Last edited:
@Wolfgang Felber
Was it preassembled?
I have assembled that version by using all mine components and everything including protection works.
I have also purchased pairs of the larger A60+ version, assembled, but without the output transistors and capacitors and that works flawless, except the speaker protection.
Both versions measure the same.
Was it preassembled?
I have assembled that version by using all mine components and everything including protection works.
I have also purchased pairs of the larger A60+ version, assembled, but without the output transistors and capacitors and that works flawless, except the speaker protection.
Both versions measure the same.
I wish I know and now I'm not ready to spend my time to solve this. Extra speaker protection costs about 10 bucks. Time is all we have.
Thanks @sajti for the suggestion but I've already done that. Inadequate. In this particular case CDIL 2240/970 was better choice.
I wanted to replace with SMD equivalents just to see if this could further reduce the noise floor, which is already low.
Mental inertia can be indeed persistent. For decades production has migrated to the former colonies, knowledge too. “Originals” are replaced by generic parts a long time ago. “Originals” that we purchase from “trustworthy” sources are produced by the usual suspects of producing fake products, just re-branded.
Last edited:
Yes, for instance a voltage of 25mv across your .25 ohm emitter resistors will indicate a current of 100ma/ output transistor.I am still working through my hum issues on an 8 pair version. I have been doing some shielding to get rid of the hum. I’m using 1 large toroid to power both channels and this is not ideal in terms of hum. Good luck on your build! I’m almost ready to go live with mine.
There are more alternatives how to connect one or multiple channels.
@Paroxod4 has presented a simple but very effective approach. He used just the on-board power supply and he has connected channels at the input RCAs. Nowhere else, just at that point.
It is of the utmost importance to understand that there should be a SINGLE connection between channels. It should be of the lowest possible resistance and the best point to connect two channels is between grounds at RCAs.
I have seen very ellaborate explanations, even booklets, but it is simple, there is no need for complicated explanation. K.I.S.S.
The best is this thread by @digi01
Digi01's thread about grounding
Please note that this thread has degenerated into grounding police yada yada and went into unnecessary lenght. Read just seversal posts of Digi01 and thats it.
Last edited: