What colour are the 10 ohm resistors on the pre-drivers supposed to be?
All the others big ones in my amp are brown. The 10 ohms resistors are black. One was touching some silk insulation and regular insulated wire. The silk insulation is burnt, the regular is melted.
I suspect that these were originally brown, but overheated and changed colour.
Can anyone verify what colour they are supposed to be?
I suspect that changing them to 5W parts might be a good idea -- judging by size, they appear to be 2 watt parts.
Are they originally 2 watt parts? The schematic does not say.
Thanks.
All the others big ones in my amp are brown. The 10 ohms resistors are black. One was touching some silk insulation and regular insulated wire. The silk insulation is burnt, the regular is melted.
I suspect that these were originally brown, but overheated and changed colour.
Can anyone verify what colour they are supposed to be?
I suspect that changing them to 5W parts might be a good idea -- judging by size, they appear to be 2 watt parts.
Are they originally 2 watt parts? The schematic does not say.
Thanks.
Last edited:
Thanks Jaycee! Why metal film?
I tried 2x12 ohm shinkoh in the output zobel, and while detailed, the bass was soft and there was EMI from the SMPS in the digital gear on the scope -- made for a hazy background. Not a lot of depth either. Nice details, but boring.
I went back to the original carbons and the music came to life again.
I've replaced the other 0.22 ohm carbon emitter resistors with 5W mills wirewound. They sound fine.
I am a bit hesitant to use any kind of metal film. I'm considering either the 5W mills or the 2 or 5 W kiwame or 2 watt Amtrans here.
But I am most concerned of the power, since there was burnt insulation where a wire was touching the resistor body.
I tried 2x12 ohm shinkoh in the output zobel, and while detailed, the bass was soft and there was EMI from the SMPS in the digital gear on the scope -- made for a hazy background. Not a lot of depth either. Nice details, but boring.
I went back to the original carbons and the music came to life again.
I've replaced the other 0.22 ohm carbon emitter resistors with 5W mills wirewound. They sound fine.
I am a bit hesitant to use any kind of metal film. I'm considering either the 5W mills or the 2 or 5 W kiwame or 2 watt Amtrans here.
But I am most concerned of the power, since there was burnt insulation where a wire was touching the resistor body.
I tried the mills as emitter resistors in the phase linear.
They sound far less lively than the stock carbons.
I can find only vishay/dale wire wound as an alternative. Are there any other 0.22R 2 watt + resistors that should be considered?
Caddock does not make 0.22R, only 0.2. and I believe 0.02 - that's an expensive proposition.
They sound far less lively than the stock carbons.
I can find only vishay/dale wire wound as an alternative. Are there any other 0.22R 2 watt + resistors that should be considered?
Caddock does not make 0.22R, only 0.2. and I believe 0.02 - that's an expensive proposition.
You're probably experiencing the effects of changing the quiescent bias more than anything else. If you just swap resistors out it's likely to change. It can be set anywhere from a full class B to a moderate AB and still work properly, and it changes the basic "sound" (and the stability or lack thereof).
The quiescent bias does not change with a different *type* of emitter resistor. It measures exactly the same with either resistor type in place. I make sure after each and every change that the bias is exactly as it was.
The Mills are more detailed and rounder sounding than the carbons, which are inaccurate and diffuse, but... they have tone, life and ambience that is absent with the Mills. Transient attack is blunted, while with the original carbon it is not -- drum strike for eg.. Also reverb trails are much shorter with the Mills.
I can see why Mills would be popular in a studio environment. They separate instruments and layers really well, on a precisely defined stage. Smaller than the carbons, but very precise. OTOH, it could make for too much reverb.
Oh sure, it can be argued that the mills is showing what is there and there may be problems elsewhere in the chain, however, I don't think a resistor can add transient attack. But it can reduce HF, and LF response, which is what I hear.
As for the carbons, yes, very inaccurate, but pleasing.
What I am looking for is both accuracy and musicality.
The Mills are more detailed and rounder sounding than the carbons, which are inaccurate and diffuse, but... they have tone, life and ambience that is absent with the Mills. Transient attack is blunted, while with the original carbon it is not -- drum strike for eg.. Also reverb trails are much shorter with the Mills.
I can see why Mills would be popular in a studio environment. They separate instruments and layers really well, on a precisely defined stage. Smaller than the carbons, but very precise. OTOH, it could make for too much reverb.
Oh sure, it can be argued that the mills is showing what is there and there may be problems elsewhere in the chain, however, I don't think a resistor can add transient attack. But it can reduce HF, and LF response, which is what I hear.
As for the carbons, yes, very inaccurate, but pleasing.
What I am looking for is both accuracy and musicality.
Like marginal stability. Not necessarily outright oscillation, but trying to - with an above-band gain peak and lots of ringing. This results in a bright, almost artificially detailed sound. I wouldn't trust a Phase Linear not to do this.The Q-point has a huge impact here, but there are other ways of doing it. Play with the value of the lead compensation capacitor and you'll see exactly what I mean. Some resistors are more inductive than others, and some (like carbon) increase in value radically at RF frequency and still have low inductance. This will tend to stabilize it. If this is what is happening, it is easily measureable.
You may need to look at a different amp. The PL is a good compromise with what was available at the time, but they can be made better now.
This sounds like the Phase linear with the carbon resistors. There is a lot of ringing, especially with instruments like the vibraphone. Also, the space/ambience seems almost "echo-like". Like artificial space.
I have a 'scope now. Can you please let me know how to measure this stability?
The mills wirewound non-inductive does not have the ringing and spaciousness of the carbon. Is it possible that the mills is stable and the carbon is not?
It is possible.
MOX would be different, not necessarily better... or worse. They tend to be low-ish inductance, but rise in value significantly with temperature. You can't burn 'em out, partly because R skyrockets when you get them glowing orange. I would expect their HF behavior to be more similar to wirewound than carbon, however.
To see if there is anything really going on, look at the *small signal* square wave response, with the scope probe directly at the feedback pickoff point and normal loading of the amp. More overshoot or ringing means less stable. Use a 1k or so test signal and zoom way in on the leading or trailing edge. This should separate stability-related artifacts from slewing-induced ones.
Marginally stable emitter followers really hate inductance in the emitter leg. That's why zobels are used in the first place. The usual 'fix' for all this is to use base stoppers. Back in those days, the low beta made it impractical. If you've replaced the outputs with MJ15024's you can add them and see if it helps anything.
To see if there is anything really going on, look at the *small signal* square wave response, with the scope probe directly at the feedback pickoff point and normal loading of the amp. More overshoot or ringing means less stable. Use a 1k or so test signal and zoom way in on the leading or trailing edge. This should separate stability-related artifacts from slewing-induced ones.
Marginally stable emitter followers really hate inductance in the emitter leg. That's why zobels are used in the first place. The usual 'fix' for all this is to use base stoppers. Back in those days, the low beta made it impractical. If you've replaced the outputs with MJ15024's you can add them and see if it helps anything.
Last edited:
If I'm not mistaken the output emitter resistors in Phase Linears that LOOK like carbon comps are actually wirewounds. If you compare an actual 2 Watt carbon comp resistor to those in the PL the resistors in the PL are quite a bit smaller. 1 Watt carbon comps are about the same size as the 2 Watt emitter resistors in the PL. Might want to check out some more Phase Linear sites.
Craig
Craig
Thank you!
All the outputs have been replaced with MJ15024G.. two most recently with non-G versions.
What value and power rating would you suggest for the base stoppers on the MJ15024's?
Does the type of resistor matter? eg. metal film/carbon film etc....
I may be able to borrow a fn generator, to check out if there is any ringing, and I'll definately keep you posted.
All the outputs have been replaced with MJ15024G.. two most recently with non-G versions.
What value and power rating would you suggest for the base stoppers on the MJ15024's?
Does the type of resistor matter? eg. metal film/carbon film etc....
I may be able to borrow a fn generator, to check out if there is any ringing, and I'll definately keep you posted.
I can post photos later, but I believe they are carbons -- if only because of the radical shift in sound when non-inductive WW was used. They are quite large.
2.2 ohms per output device, give or take, as long as they're the same. Non-inductive types are best. I tend to use MOX because they're flameproof. It doesn't NEED to be more than 1/2W, but I go larger in case of a circuit fault. PLs have been known to do this...
Look at the 10 Ohm resistor, see how smooth the surface is and how square the corners are. Now look at the .22 Ohm resistors, there should be a seam running down the sides 180 degrees apart and the corners aren't so square. Looks like a completely different manufacturing process. My first PL experience was back in the late 70s or early 80s and I'm almost positive the emitter resistors were WW. I've worked on several since but never had to replace any of those particular resistors. Have you tried looking for any carbon comps under 1 Ohm, I couldn't find any. I have drawers full of CC resistors, none are below 1 Ohm. I have a box full of the "seamed" ones around here somewhere, once I find it I'll sacrifice one and crack it open and see what I get.
Craig
Craig
The 150 ohm Rbe serves 2 purposes - to bleed charge off the base during turn-off, and to set the idle bias current in the pre-driver. It's probably less sensitive to what type of resistor is used there than for the emitter resistors. But with these old finicky things I wouldn't trust it till I tried. The original transistors had less fT and gain than anything modern you'd put in - and things that you wouldn't expect to matter can have unintended consequences.
Base stopper is something else entirely - a low ohm resistor in series with the base. Its purpose is to cancel any negative resistance at the base that ends up being generated for a variety of reasons. Back then, intrinsic resistances were so high to make them unnecessary. Check out the curves on the old RCA1B05 (used as a replacement for PL909 way back when) if you can find the data sheet some time. They probably didn't even need emitter resistors.
Base stopper is something else entirely - a low ohm resistor in series with the base. Its purpose is to cancel any negative resistance at the base that ends up being generated for a variety of reasons. Back then, intrinsic resistances were so high to make them unnecessary. Check out the curves on the old RCA1B05 (used as a replacement for PL909 way back when) if you can find the data sheet some time. They probably didn't even need emitter resistors.
Thanks!
OK, so the 10 ohm needs to be 2 watts minimum, with 3 better. I won't replace it.
It sounds like the 150 ohm can be 0.75 watts, but won't do much. Since I have them I may as well try it, and see if there is any difference. I am doubtful, but it maybe a bit quieter?
I can always check that with the scope.
But today, I'm just enjoying the gear... not changing it. After putting the amp directly on the concrete floor, as opposed to a makeshift rack, it is sounding better.
OK, so the 10 ohm needs to be 2 watts minimum, with 3 better. I won't replace it.
It sounds like the 150 ohm can be 0.75 watts, but won't do much. Since I have them I may as well try it, and see if there is any difference. I am doubtful, but it maybe a bit quieter?
I can always check that with the scope.
But today, I'm just enjoying the gear... not changing it. After putting the amp directly on the concrete floor, as opposed to a makeshift rack, it is sounding better.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.