Hi, The 470Ω resistor has had a couple of 1.5K strapped across it and the capacitor in the 470uf position is a 330 with a smaller cap strapped across it to give the measured value. It is these last two components that I replaced as they were blown, like with like without consulting the diagram. Someone has deliberately made these changes I was hoping someone could suggest why
I do tend to agree with wiseoldtech that the poor amp had some real bodging. The 1k5's strapped across the 470R will have been an attempt at increasing output bias current and hence power. Probably best cause of action would be to find the original schematic and start from there. I suspect you will need new KT88's anyway.
So thought you mite like to be brought up to date with this project. Following advice above I have mades some changes to the amp. I have reverted the resistor and capacitor on the KT88 cathodes to the original schematic values and changed the grid resistor to 220K from 560K. This brought the amp back to some semblance of life , however all the valves were FUBAR. There was lots of distortion, crackling and channels dropping out. I have changed them out for a set of Chinese/Russian valves I got with the amp that I guess were probably the ones originally supplied. The amp now works and sounds ok at low listening levels but if you turn the input up it al gets very distorted.
I tried the triode as well as the UL settings with no discernible difference. The different feedback settings also made little difference to the distortion though did change the sound level.
I also experimented lifting the diode in the powers supply and re-connecting it to the resistor feeding the 12AX7. This made little difference, dropping a couple of volts and the supply to the 12AU7s lifting by a couple of volts, so I put it back where it came from for now.
I really would appreciate any suggestions of what to do next.I would like to get the amp working as well as I can and hopefully have it working so that it doesn't trash valves!
Thanks for your help!
I tried the triode as well as the UL settings with no discernible difference. The different feedback settings also made little difference to the distortion though did change the sound level.
I also experimented lifting the diode in the powers supply and re-connecting it to the resistor feeding the 12AX7. This made little difference, dropping a couple of volts and the supply to the 12AU7s lifting by a couple of volts, so I put it back where it came from for now.
I really would appreciate any suggestions of what to do next.I would like to get the amp working as well as I can and hopefully have it working so that it doesn't trash valves!
Thanks for your help!
Other thing to check is the output transformer. To do this disconnect from unit. Apply 1KHz sinewave from sig generator through 1Kohm to primary of transformer. Measure voltage across primary with scope. Should be virtually same as voltage across sig gen output. Measure secondary voltage that should be sqrt(8/5000) times as big. Short secondary. Voltage on primary to drop to virtually nothing.
Last edited:
I did a bit of 'scoping today and found a seriously dry joint, now that is resolved the random channel drop out and distortion is fixed. I tracked through with the signal being fed from the input. Apart from the input to the first stage where the signal seemed noisy that I would think was the most sensitive setting on the 'scope. All the rest of the traces looked clean and undistorted right up to the transformer outputs.
So now the amp is working there are a few things that I don't like so much. Generally the sound is a bit sibilant. Any suggestions as to where I could make changes to improve this ?
So now the amp is working there are a few things that I don't like so much. Generally the sound is a bit sibilant. Any suggestions as to where I could make changes to improve this ?
The X10 setting and capacitor input settings are recommended.
And worthy scope used properly should not experience a problem.
A 10X probe (9 Meg Ohm in series) and a 1 Meg Ohm scope input.
If you use the scope internal AC coupling, most scopes rate that combination at 300VDC max.
Using the same 10X probe and a series capacitor does not change the transient voltage that will appear at the scope input.
There is a 100X probe, 9 Meg series, 110k parallel, and the scope input 1 Meg.
That keeps the scope input safe for much higher B+, plate circuits, etc.
How do I know that you can blow out most scope inputs with a 10X probe, AC coupling, and more than 300V?
I have heard of lots of such things from a major world service center for scopes.
Also, read the manual (if the scope manufacturer is not afraid to bring that subject up).
Some data sheets will list that truth.
Analog, digital, it does not matter. The input circuits start out as analog, where they go after that is up to the scope design.
We are talking about scope designs at least as early as the 1950s and up.
Use improperly at your own risk. Your mileage may vary.
Some of the old scopes used 600V AC coupling capacitors. But the transient voltages and currents could still be an issue.
On some modern scopes I have seen a baseline rise (steady rise, not just the temporary rise during charging the cap) with only 300V, and set to AC coupling, using a 9 Meg series R in the 10X probe, and 1 Meg in the scope (1 Meg after the AC coupling cap, but that connects to the attenuator and/or directly to the input amplifier).
Tube grids take lots of transients.
Solid state inputs take lots of transients too (well, at least one time).
It is not always an issue of positive voltages, sometimes it is an issue of negative voltages (and reverse breakdown of a junction).
Charge up that cap to 300V, maybe it is ok. But then ground the probe tip, or probe a low positive voltage, and now the voltage goes Negative 300V to the scope input circuitry.
If you use the scope internal AC coupling, most scopes rate that combination at 300VDC max.
Using the same 10X probe and a series capacitor does not change the transient voltage that will appear at the scope input.
There is a 100X probe, 9 Meg series, 110k parallel, and the scope input 1 Meg.
That keeps the scope input safe for much higher B+, plate circuits, etc.
How do I know that you can blow out most scope inputs with a 10X probe, AC coupling, and more than 300V?
I have heard of lots of such things from a major world service center for scopes.
Also, read the manual (if the scope manufacturer is not afraid to bring that subject up).
Some data sheets will list that truth.
Analog, digital, it does not matter. The input circuits start out as analog, where they go after that is up to the scope design.
We are talking about scope designs at least as early as the 1950s and up.
Use improperly at your own risk. Your mileage may vary.
Some of the old scopes used 600V AC coupling capacitors. But the transient voltages and currents could still be an issue.
On some modern scopes I have seen a baseline rise (steady rise, not just the temporary rise during charging the cap) with only 300V, and set to AC coupling, using a 9 Meg series R in the 10X probe, and 1 Meg in the scope (1 Meg after the AC coupling cap, but that connects to the attenuator and/or directly to the input amplifier).
Tube grids take lots of transients.
Solid state inputs take lots of transients too (well, at least one time).
It is not always an issue of positive voltages, sometimes it is an issue of negative voltages (and reverse breakdown of a junction).
Charge up that cap to 300V, maybe it is ok. But then ground the probe tip, or probe a low positive voltage, and now the voltage goes Negative 300V to the scope input circuitry.
Last edited:
Well I managed not to blow my scope up and still need some help with why this amp should be sibilant. It didn’t used to sound like this but I have made some changes and also swapped out the kt88s and AU7s as they sounded really distorted. Before I blow money on new valves, which should I change first and what valves should I buy?
Thanks
Thanks
It is hard to figure out what has happened without knowing:
1. We need the original circuit diagram and particular parts that you started with and liked the sound of.
2. We need the new circuit diagram, and all the old versus new changed parts that you are now using (the present amp that you do not like the sound of).
What post #s are 1. and 2.?
1. We need the original circuit diagram and particular parts that you started with and liked the sound of.
2. We need the new circuit diagram, and all the old versus new changed parts that you are now using (the present amp that you do not like the sound of).
What post #s are 1. and 2.?
Hi again! Post 1 has a pdf of where I started with a not working amp somewhat modified marked up over what I believed to be a manufacturers schematic. Post 13 has a diagram of the actual power supply connections and a photo of the insides. Post22 is where I am now. Along the way I have binned the golden lion KT88s as they were definitely dud. I also changed the AU 7s as they were also causing audible distortion . I am currently running no name Chinese valves that came with the amp and I suspect were the originals.
Today I had some crackle from the right channel. I thought I had fixed that when I found a dry joint.
Before the amp got sick it sounded ok but lacking a bit of clarity in the upper frequencies, now it is distinctly sibilant on vocals and probably still shy of clarity in the high frequencies.
I would like to improve the amp and definitely get rid of the sibilance and intermittent crackles!
Thanks for your ideas[emoji2]
Today I had some crackle from the right channel. I thought I had fixed that when I found a dry joint.
Before the amp got sick it sounded ok but lacking a bit of clarity in the upper frequencies, now it is distinctly sibilant on vocals and probably still shy of clarity in the high frequencies.
I would like to improve the amp and definitely get rid of the sibilance and intermittent crackles!
Thanks for your ideas[emoji2]
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.