Some current source circuits are quite noisy, but those are usually noisy from the very moment they are turned on.
Carbon composition resistors are not particularly stable, so it would be interesting to see if replacing them with carbon film or metal film changes anything.
Carbon composition resistors are not particularly stable, so it would be interesting to see if replacing them with carbon film or metal film changes anything.
I have a scope and a signal generator but not a lot of experience using them. How can I test for oscillation? I installed 1uf ceramic caps on the heater pins as that was reported to help.
Member
Joined 2009
Paid Member
Audible hiss sometimes means oscillations at above-audible frequency. Sometimes hard to detect, contact with the scope probe may stiffle the oscillations. The answer is a small cap on the CCS which will kill any oscillations and once you’ve done that you can more or less rule it out.. Wish I could share a reference to where to put said cap but I’ve seen it around on the net.
I doubt an oscillating CCS is the culprit as it wouldn’t appear after a few months. Something changes over time if you hear the hiss only after a period of time. Probably something changing under heat stress or electrical stress. Tubes and CC resistors are often suspects; it’s well known that CC gets noisy over time (Best to overate them for power disipation).
I doubt an oscillating CCS is the culprit as it wouldn’t appear after a few months. Something changes over time if you hear the hiss only after a period of time. Probably something changing under heat stress or electrical stress. Tubes and CC resistors are often suspects; it’s well known that CC gets noisy over time (Best to overate them for power disipation).
Last edited:
I have a scope and a signal generator but not a lot of experience using them. How should I test for oscillation?
First isolate the noise to the CCS.
If the noise is only in the left channel, swap the CCS between channels.
Does the noise then switch channels?
If the noise is only in the left channel, swap the CCS between channels.
Does the noise then switch channels?
Try substituting a resistor for the CCS and see if the problem persists.
The dual-DMOS (DN2540) CCS may have some issues (see the "His Master's Noise" article") but I haven't experienced this. If you have an IR thermometer, make sure that the devices are not over-heating as well.
Measure the DC voltage drop across the CCS.
Divide that by the value of current that you chose for the CCS.
Use that resistance value.
Multiply the DC voltage drop by the CCS current.
The wattage of the resistor should be at least 3 times that value.
The voltage gain with the resistor instead of the CCS will be less.
Divide that by the value of current that you chose for the CCS.
Use that resistance value.
Multiply the DC voltage drop by the CCS current.
The wattage of the resistor should be at least 3 times that value.
The voltage gain with the resistor instead of the CCS will be less.
If the hiss developed over time it can be an oscillation in one of the tubes. Try to short inputs of the second stage to check if it is the first or second stage causing it.
LED bias is inherently noisy. Put a big capacitor (like 220uF or more) in parallel with the LED bias. Try both stages, you can just tack it in to see if it helps.
A resistor already went bad. It’s probably happened again. Dump the carbon comp.
Member
Joined 2009
Paid Member
Or if it's important for whatever reason that you use CC just try larger wattage and parallel them up, they don't go bad so often when they're not stressed.
FYI - to be honest here, I've never used CC myself. But I've just bought CC parts to build a phono and line amp using almost exclusively CC resistors. I also plan to use CC in a power amp. I'm going to find out for myself what kind of hassles it gives! - no technical reason to use CC, just historical reasons. Technically, they're a poor choice except in very specific uses.
FYI - to be honest here, I've never used CC myself. But I've just bought CC parts to build a phono and line amp using almost exclusively CC resistors. I also plan to use CC in a power amp. I'm going to find out for myself what kind of hassles it gives! - no technical reason to use CC, just historical reasons. Technically, they're a poor choice except in very specific uses.
Last time it seemed to be the second stage. However, since that cuts the gain significantly, it is hard to be sure. Thanks for the help!
I'll give that a try. Can I do the same for the single LED tube (D3A) as well as the 4 LED string(5867)?
In response to @Bigun and @astouffer, I thought Carbon Comp were the preferred resistor for grid stoppers? What's the chances of the same resistor failing within such a short time?
Member
Joined 2009
Paid Member
I think there is some 'lore' that has built up around CC resistors, not unusual in our hobby. There's almost no place for them in modern electronics but I have read that they are still used in applications where resistors maybe exposed to pulsed power overloading as they have the ability to absorb this power throughout the whole body of the resistor whereas modern film resistors would suffer. Generally, CC are less stable over time, have high temperature dependence, are rarely found with 5% tolerance so not intended for precision applications, and they are one of (if not THE) noisiest resistors you can get. I'm planning to use them because I'm building a copy of an old design, a case of 'industrial archeology' rather than best possible performance.
Last edited: