Pleased to hear its working now, as I bet you are 
I would reserve judgment on what has happened and whether he has really found the fault.
Very very rare for a power diode to fail open circuit and even if it did, the rails would still be 'correct' although one side would be operating with a half rather than full wave rectifier. That means the voltages would be correct at light loading but would show increased ripple as current demand got higher.
I would reserve judgment on what has happened
and whether he has really found the fault.Very very rare for a power diode to fail open circuit and even if it did, the rails would still be 'correct' although one side would be operating with a half rather than full wave rectifier. That means the voltages would be correct at light loading but would show increased ripple as current demand got higher.
OK++++++ update help still needed!
Ok so the issue in the beginning was blown output transistor. Which worked after replacing bias was fine.
Then power diode failed somehow which got replaced along with all others (in main board) and now the bias is 2mv? I made sure I measured it correctly, I can confirm the there is no heat and everything is as cool as ice even at semi loud volumes.
I think the problem is not the main power diodes which got replaced by normal types, I had a huge pack of UF4001's I bought 4 years back which I thought would be good replacement for the Cyrus since one had failed I advised to replaced all with "superior" type but now the bias is super low!
The Cyrus works perfectly fine! Perfectly and it was only noticed after a few weeks of "fixing" it that it was realised there was no heat. I thought that is also sounded like it had less drive but then I handed the Cyrus over weeks back and had not heard it since the capacitors were replaced and now the diodes.
From memory 10 diodes were replaced from the main board, my mate is not sure if the bias was fine before the capacitor replacement or not he thinks it was fine.
Can diode have such an influence as to lower the bias so much! I think it was 8mv before, I would have to double check earlier post to confirm actual reading.
Any advice? I already tired what the manual stated about the bias last week but it didn't make any difference, I rather just replace the resistor to fix bias then remove all the diodes. Well actually I wont be doing it lol. Thanks in advance
Ok so the issue in the beginning was blown output transistor. Which worked after replacing bias was fine.
Then power diode failed somehow which got replaced along with all others (in main board) and now the bias is 2mv? I made sure I measured it correctly, I can confirm the there is no heat and everything is as cool as ice even at semi loud volumes.
I think the problem is not the main power diodes which got replaced by normal types, I had a huge pack of UF4001's I bought 4 years back which I thought would be good replacement for the Cyrus since one had failed I advised to replaced all with "superior" type but now the bias is super low!
The Cyrus works perfectly fine! Perfectly and it was only noticed after a few weeks of "fixing" it that it was realised there was no heat. I thought that is also sounded like it had less drive but then I handed the Cyrus over weeks back and had not heard it since the capacitors were replaced and now the diodes.
From memory 10 diodes were replaced from the main board, my mate is not sure if the bias was fine before the capacitor replacement or not he thinks it was fine.
Can diode have such an influence as to lower the bias so much! I think it was 8mv before, I would have to double check earlier post to confirm actual reading.
Any advice? I already tired what the manual stated about the bias last week but it didn't make any difference, I rather just replace the resistor to fix bias then remove all the diodes. Well actually I wont be doing it lol. Thanks in advance
Assuming that the main voltage rails are correct then nothing in the power supplies can affect the bias.
The bias is determined solely by the voltage across Q39 which is used as a voltage to turn on the output stage and drivers.
The circuit showed a fixed bias arrangement which means it is very dependent on the transistors used... I think we covered all this way back in the thread at some point. So if the amp has low bias then the combined value of R81 and R83 need to be lowered slightly. Use the bulb tester (and no speakers connected).
A preset could be fitted the value of which depends on the value of the resistors fitted now. The preset value needs to at least equal the already present value of the series resistors which would mean a 1k based on the circuit.
The bias is determined solely by the voltage across Q39 which is used as a voltage to turn on the output stage and drivers.
The circuit showed a fixed bias arrangement which means it is very dependent on the transistors used... I think we covered all this way back in the thread at some point. So if the amp has low bias then the combined value of R81 and R83 need to be lowered slightly. Use the bulb tester (and no speakers connected).
A preset could be fitted the value of which depends on the value of the resistors fitted now. The preset value needs to at least equal the already present value of the series resistors which would mean a 1k based on the circuit.
Thanks Mooly.
R81 looks like a resistor but its not! It looks completely different from everything else. It does look like what's marked on the boards as X and a number next to it i.e. X1
From some reading its some kind of noise remover component and not a resistor?!?
R83 is a 680r. There must be some voltage as it did spike upwards to like 3.5mv when music was playing.
Should I only lower or increase that? To 1k? I did get that part.
You say I should lower the value should I lower R83 to something less than 680r?
Also would you know why the parts don't match up to the description adding a resistor across the terminal didn't do anything! Thanks
R81 looks like a resistor but its not! It looks completely different from everything else. It does look like what's marked on the boards as X and a number next to it i.e. X1
From some reading its some kind of noise remover component and not a resistor?!?
R83 is a 680r. There must be some voltage as it did spike upwards to like 3.5mv when music was playing.
Should I only lower or increase that? To 1k? I did get that part.
You say I should lower the value should I lower R83 to something less than 680r?
Also would you know why the parts don't match up to the description adding a resistor across the terminal didn't do anything! Thanks
R81 and it counterpart in the other channel are resistors alright but probably of a type that could be trimmed by hand or selected on test after manufacture so it's likely that it does look different if its been doctored to 130R (a non-standard value) or whatever appropriate value was required for the correct bias of that particular amplifier with its particular set of semis in the output stage.
The manufacturer could have fitted a standard 220R bias trimpot instead but these can be unreliable and don't tolerate repeated twiddling. Measure R81 out of circuit if you are inquisitive. It probably reads somewhere around nominal 130 ohms anyway.
The manufacturer could have fitted a standard 220R bias trimpot instead but these can be unreliable and don't tolerate repeated twiddling. Measure R81 out of circuit if you are inquisitive. It probably reads somewhere around nominal 130 ohms anyway.
I did measure it in circuit last week. My meter is worth £5 but it said 00.5 ohms when set in the 200 setting. This was for R81 and R83 was something like 800 ohm or 820 / 850 cant remember exactly.
I could take it out of circuit on Sunday or ask mate to do it earlier.
As stated earlier it all seems to be working fine but those UF4001's seemed to have done something to the bias. They are suppose to be Ultra Fast and improve sound, I cannot comment if that's true or not but given a 50 pack was less a couple quid you cant go wrong.
A lot of work has already gone into this amplifier, transistors, capacitors and diodes between me and my mate 30+ parts replaced!
So far the only lead I have is t lower R83 and other side which I think is R84. Its currently 680r (so the manual states) so I will try find 600?
but it said 00.5 ohms when set in the 200 setting. This was for R81 and R83 was something like 800 ohm or 820 / 850 cant remember exactly. I could take it out of circuit on Sunday or ask mate to do it earlier.
As stated earlier it all seems to be working fine but those UF4001's seemed to have done something to the bias. They are suppose to be Ultra Fast and improve sound, I cannot comment if that's true or not but given a 50 pack was less a couple quid you cant go wrong.
A lot of work has already gone into this amplifier, transistors, capacitors and diodes between me and my mate 30+ parts replaced!
So far the only lead I have is t lower R83 and other side which I think is R84. Its currently 680r (so the manual states) so I will try find 600?
You can't adjust bias in a hit and miss fashion. Changing the 680 ohm to 600 ohm could be enough to alter the bias 10 to 20 fold, it is that critical on value. Worst case and the amp could go into thermal runaway and destroy the output stage before you realise what is going on.
The recommended value is 8 to 20mv across the 0.22 ohm which is a current of 36ma to 90ma. Toward the lower end is going to be safer to avoid thermal issues.
If you want to stick with fixed resistors then I would connect your meter securely across the 0.22 ohm so that the voltage reading is permanently displayed and then add parallel resistors across the 680 ohm. Begin by dabbing a 15k across the 680 ohm and see what effect that has. Then work down, 12k, 10k, 8k2 and so on.
The recommended value is 8 to 20mv across the 0.22 ohm which is a current of 36ma to 90ma. Toward the lower end is going to be safer to avoid thermal issues.
If you want to stick with fixed resistors then I would connect your meter securely across the 0.22 ohm so that the voltage reading is permanently displayed and then add parallel resistors across the 680 ohm. Begin by dabbing a 15k across the 680 ohm and see what effect that has. Then work down, 12k, 10k, 8k2 and so on.
13mv equates to around 60 ma in the 0.22 ohm and that would give a dissipation in each output transistor of around 1.8 watts. It would feel cool at that, particularly if run for just a few minutes.
I would just neatly attach the resistor across the existing one. If the original is 680 ohm then you would need a 650 ohm replacement. Much easier just to add the 15k to what is there.
I would just neatly attach the resistor across the existing one. If the original is 680 ohm then you would need a 650 ohm replacement. Much easier just to add the 15k to what is there.
Adding a resistor can be done really neatly, either on top or underneath.
Replacing diodes in the PSU can't really affect the bias, in fact the bias will be pretty stable against even major changes in supply voltage (like changes in voltage of -/+50%). A different type of diode might alter a rail value by a tiny fraction of one percentage point at most.
The only other thing I would say is that bias will depend on the amp being in a thermally stable state which means leaving it on for quite a while. If you checked it when the amp (weather) was cold then it is quite possible to get an initially low result.
Replacing diodes in the PSU can't really affect the bias, in fact the bias will be pretty stable against even major changes in supply voltage (like changes in voltage of -/+50%). A different type of diode might alter a rail value by a tiny fraction of one percentage point at most.
The only other thing I would say is that bias will depend on the amp being in a thermally stable state which means leaving it on for quite a while. If you checked it when the amp (weather) was cold then it is quite possible to get an initially low result.
Ah yes. The diode could have changed voltage slightly I guess. I did read something about voltage drop in diodes.
It's 28 degrees at the mo! Last night was very hot also. It sounds like it need increasing but ever so slightly. I know from some experience they should at least be slightly warm. They were ice cold last time and also reported cold last night.
Will update bit later as kids are loving the 30 degrees heat!!
It's 28 degrees at the mo! Last night was very hot also. It sounds like it need increasing but ever so slightly. I know from some experience they should at least be slightly warm. They were ice cold last time and also reported cold last night.
Will update bit later as kids are loving the 30 degrees heat!!
Ended up fitting a 12k on top, lucky I had it!
After 15 minutes the bias measured 18.5mv after 30 minutes it did drop to around 17mv and the transistors were ever so slightly warm.
I refuse to do any more work on this Cyrus LOL
The dc offset was a bit weird one was minus, I could have had the probes in wrong way round but both channels measured less than 10mv.
Did not sound too bad at all! New capacitors and diodes, bias adjusted,decent output transistors. Sounded very informative and involving if a little flat for my liking.
Job done! Again! Thanks Mooly! Again!
I wonder what happened to the other guy was looking forward to his results!?
After 15 minutes the bias measured 18.5mv after 30 minutes it did drop to around 17mv and the transistors were ever so slightly warm.
I refuse to do any more work on this Cyrus LOL
The dc offset was a bit weird one was minus, I could have had the probes in wrong way round but both channels measured less than 10mv.
Did not sound too bad at all! New capacitors and diodes, bias adjusted,decent output transistors. Sounded very informative and involving if a little flat for my liking.
Job done! Again! Thanks Mooly! Again!
I wonder what happened to the other guy was looking forward to his results!?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.