You guys are just way too kind here.
@Anatech: I was actually thinking the same thing on the next upgrades, so i will try to do so next week and see how much i can improve the overall sound.
BTW there's a story behind the degrading transistors, i never gave this much thought some years back either. I used to have a part-time job in an hi-fi store where i worked in the used equipment department. Our job was to test and sell all equipment that was traded-in on new equipment. I loved all that vintage stuff that came trough the shop.
We also had a senior technician in the shop who did nothing but repair and service equipment all day long, i really wanted to learn more about those amplifiers so whenever i had time i would talk to him about repairs.
One day after closing the shop, he was still working on an amplifier (NEC). I was curious and asked what the problem was and if he already found something that was broken.
It's oscillating... (and showed me on his scope while he was running a test signal through it)
Hmm ok, is that bad? It works like it should, right?
It sounds old and tired to me. The transistors are probably worn and out of spec so i'll just replace the drivers and the OP's.
But how do you know, you've only tested the output right?
It's a hunch...
Indeed after he replaced the transistors the oscillations were gone and it sounded much fresher. So from that day i realized you must take age of any part in consideration. Perhaps it were the drivers that were the problem, but the technician didn't care about cost of parts. He really did not want to get anything back he repaired. "Why not overhaul something if you have the chance and get the best out of something so you can enjoy it to the fullest?" That remark always stuck with me ever since.
And he was right about the parts, they really don't cost that much.
@Anatech: I was actually thinking the same thing on the next upgrades, so i will try to do so next week and see how much i can improve the overall sound.
BTW there's a story behind the degrading transistors, i never gave this much thought some years back either. I used to have a part-time job in an hi-fi store where i worked in the used equipment department. Our job was to test and sell all equipment that was traded-in on new equipment. I loved all that vintage stuff that came trough the shop.
We also had a senior technician in the shop who did nothing but repair and service equipment all day long, i really wanted to learn more about those amplifiers so whenever i had time i would talk to him about repairs.
One day after closing the shop, he was still working on an amplifier (NEC). I was curious and asked what the problem was and if he already found something that was broken.
It's oscillating... (and showed me on his scope while he was running a test signal through it)
Hmm ok, is that bad? It works like it should, right?
It sounds old and tired to me. The transistors are probably worn and out of spec so i'll just replace the drivers and the OP's.
But how do you know, you've only tested the output right?
It's a hunch...
Indeed after he replaced the transistors the oscillations were gone and it sounded much fresher. So from that day i realized you must take age of any part in consideration. Perhaps it were the drivers that were the problem, but the technician didn't care about cost of parts. He really did not want to get anything back he repaired. "Why not overhaul something if you have the chance and get the best out of something so you can enjoy it to the fullest?" That remark always stuck with me ever since.
And he was right about the parts, they really don't cost that much.
Hi Roger,
It doesn't take much help to cool things off sometimes. When you get a chance, look up these parts to see what the case style is designed to dissipate. Most TO-220 transistors are approx. 2 watt. That isn't much higher than a TO-126, or even a TO-92M (the 1 watt version). The case temperature also depends on the ambient temperature. That can easily be 40 °C. Hotter if there are a bunch of hot transistors all huddled in the same place. My favorites are those upside down power supply PCBs that Pioneer loves. Everything is usually toast in those.
Hi jooch,
In service, you have to have some justification to replace stuff. Normally outputs and drivers are fine, except in the HK 730 receiver where everything needs an upgraded replacement ... with higher breakdown voltages. It is not normal for transistors to affect sound quality unless they throw the quiescent conditions way off. So, when you do change something, have a solid reason why you are doing that.
I do agree on one thing you said. If you are into a unit, give it all it needs the first time. That way the customer pays for all of the job, and is much happier in the end. You can not afford to redo work or replace what used to be good parts. Buy from the distribution network. Ebay can be attractive, but experienced techs will fully test those parts before committing them to a repair or prototype. If they are bad, they get rid of the parts as they lost the gamble. A customer unit never has questionable parts installed. See my first comment.
I bought from a source that was a touch more expensive when I ran my shop. One day he asked me why, and why I didn't complain. Well, I told him that the difference wasn't much, I knew exactly where his parts came from, and it was the same percentage markup anyway. My customers paid less for parts from me, and a little more in labor. I only ever received three defective parts over 14 years dealing with him. I was cheaper for me to pay more for the real part - for certain. One distributor I knew had a part marking machine in New York somewhere. He was not someone I respected, and I didn't buy from him. No names here, everything sorted itself out over the years.
Only certain parts from long ago actually age poorly. Once the manufacturing processes were brought under control, variability was less of a problem. Same for reliability.
-Chris
It doesn't take much help to cool things off sometimes. When you get a chance, look up these parts to see what the case style is designed to dissipate. Most TO-220 transistors are approx. 2 watt. That isn't much higher than a TO-126, or even a TO-92M (the 1 watt version). The case temperature also depends on the ambient temperature. That can easily be 40 °C. Hotter if there are a bunch of hot transistors all huddled in the same place. My favorites are those upside down power supply PCBs that Pioneer loves. Everything is usually toast in those.
Hi jooch,
In service, you have to have some justification to replace stuff. Normally outputs and drivers are fine, except in the HK 730 receiver where everything needs an upgraded replacement ... with higher breakdown voltages. It is not normal for transistors to affect sound quality unless they throw the quiescent conditions way off. So, when you do change something, have a solid reason why you are doing that.
I do agree on one thing you said. If you are into a unit, give it all it needs the first time. That way the customer pays for all of the job, and is much happier in the end. You can not afford to redo work or replace what used to be good parts. Buy from the distribution network. Ebay can be attractive, but experienced techs will fully test those parts before committing them to a repair or prototype. If they are bad, they get rid of the parts as they lost the gamble. A customer unit never has questionable parts installed. See my first comment.
I bought from a source that was a touch more expensive when I ran my shop. One day he asked me why, and why I didn't complain. Well, I told him that the difference wasn't much, I knew exactly where his parts came from, and it was the same percentage markup anyway. My customers paid less for parts from me, and a little more in labor. I only ever received three defective parts over 14 years dealing with him. I was cheaper for me to pay more for the real part - for certain. One distributor I knew had a part marking machine in New York somewhere. He was not someone I respected, and I didn't buy from him. No names here, everything sorted itself out over the years.
Only certain parts from long ago actually age poorly. Once the manufacturing processes were brought under control, variability was less of a problem. Same for reliability.
-Chris
I know what you mean, especially the 1943/5200 pair fell victim to that. Right now i order from Polida2008 on ebay, so far he hasn't disappointed me.
Yesterday i ordered some 2SA970's from him to replace those 979's and 2SC3421/2SA1358 to replace the drivers.
I already replaced C1 to 4.7/50, C5 to 220/35 and C7 to 100/50 on both channels.
This afternoon i will be putting in 1/2w metal film resistors, that's twice the original rating.
I also have two other upgrades planned:
- Replace the two 10.000uF/50V power supply caps with four 6.800uF/63V.
- Modify the stereo/mono or loudness button to provide a source direct function.
Bypassing the preamp will mess up the volume control so i probably need an additional preamp, right?
Hi jooch,
Your main filter caps are probably fine. Increasing capacitance is not recommended, but the higher voltage buys you a better capacitor. There is nothing wrong with 10 KuF. When you install multiple capacitors for one, the wiring actually becomes an electrical problem Put whatever you buy into the same holes. If the new part doesn't fit, don't buy it to begin with.
Physically, multiple capacitors represent an electrical danger. Electrical or friction tape types are not suitable for mounting stuff like this. In fact, short of making another circuit board to mount the new capacitors on, or mechanical fastening, there is nothing you can do to make such an arrangement safe or suitable. Don't be a hacker (or "plumber" as hacks are known by).
-Chris
Your main filter caps are probably fine. Increasing capacitance is not recommended, but the higher voltage buys you a better capacitor. There is nothing wrong with 10 KuF. When you install multiple capacitors for one, the wiring actually becomes an electrical problem Put whatever you buy into the same holes. If the new part doesn't fit, don't buy it to begin with.
Physically, multiple capacitors represent an electrical danger. Electrical or friction tape types are not suitable for mounting stuff like this. In fact, short of making another circuit board to mount the new capacitors on, or mechanical fastening, there is nothing you can do to make such an arrangement safe or suitable. Don't be a hacker (or "plumber" as hacks are known by).
-Chris
Hi jooch,
I don't know what the gain of your tone control section is. One op amp can replace that circuit, but why? Upgrade the caps, then leave that circuit alone. If you really want a different amp, just build it. You actually decrease its value by doing modifications like that.
-Chris
I don't know what the gain of your tone control section is. One op amp can replace that circuit, but why? Upgrade the caps, then leave that circuit alone. If you really want a different amp, just build it. You actually decrease its value by doing modifications like that.
-Chris
Don't be negative, just trust me! I might not have gained a degree in electronics yet but i make that up in crafting skills. Here's a JVC i upgraded some years back:
HELP? JVC JA-S5G Amp? - Page 2 - AudioKarma.org Home Audio Stereo Discussion Forums
You made me smile when you mentioned hacker and plumber. I do hack things all the time, but that mainly concerns computer stuff. When i modify things i really put my best foot forward and take my time.
I think that also makes the difference between hacking and crafting, taking your time and being passionate. Although it doesn't mean i won't use plumbing to reach my goals 😉
I know most vintage lovers fear most above anything else equipment getting lost into the wrong hands. But really thats the fate of such things if it cannot make it to the living room.
I firmly believe one should always try to improve something if possible, although not many will actually try or seek hard enough to do so.
In my opinion the AKAI already sounds better than most audio surround sets for which ignorant consumers today will happily pay four figure prices for. Whenever someone happens to pick up some music playing from my setup at home, i always get reactions like: "That sounds so tasty, can i buy that somewhere?"
That's exactly my point! You can't just build something like this. I don't have enough funds to let someone CNC mill a front plate, knobs and what not.
The thing is these amps were forgotten, abandoned, binned, because they were malfunctioning or did not meet the the requirements of this day anymore.
I got this AKAI (including speakers) for a bottle of wine because the previous owner found it unreliable and inconvenient not having a remote. He now listens to music on a plastic surround receiver with five small speakers and a subwoofer, now that truly is a waste of music and time.
In financial terms i am indeed likely to decrease the value of the AKAI, but i think that's good. Keeps collectors and greed away, those aren't the kind that actually listen to music anyway.
In musical terms i believe the AKAI is gaining in value.
Anyway here's the current preamp, but i don't want to replace it. For vintage purposes it is useful, however when i want to properly listen to music i don't want any tonal change to the music.
But if i simply bypass the preamp the volume control will be completely different in scale, that's why i thought of a simple opamp gain stage to compensate a little.
It is important that i know the current gain, but how does one calculate that?
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Hi jooch,
2) keep your work reliable
From experience working on equipment that has been modified by folks who only know what they read on the internet, many are more concerned with the "modification" and saying they did it than actually improving the equipment. You are spending time and money to create a better sounding bit of kit. Wouldn't you love to know where to spend your money and time where it really brings results? Wouldn't you rather have something that lasted years rather than a unit that was moved, then went poof!
The most dangerous change people make is to strap on extra capacitors somewhere and wire them into the existing (or new) capacitors. Mechanical mounting is not secure in 100% of the cases I have seen. This was even done to a Carver M 1.5t. That unit has supply voltages of +/- 125 VDC! The added caps were just flopping around against electrical tape. Some idiot even removed the meters on a Marantz 500, filed the openings and stuck two more caps in, bottoms hanging out. That amplifier is a classic and can put many other amps to shame, and some moron hacked it. No common sense.
I am not lumping you in that group (or I wouldn't even comment). I am trying to pass on the knowledge. Whenever you plan to do something, have a think on it. Capacitance is not a cure-all, and you most certainly can have too much of it. If you increase the capacitance a lot, you have to still charge the same amount of energy, but in a shorter length of time. This is referred to as "conduction angle" after how many degrees in 360 ° are the diodes conducting. Since the time is less, that means the peak current is higher, and since that is true, so are the IR losses in the transformer higher. Everything runs hotter, and you generate high frequency noise bursts, or make them worse.
So if the design has enough capacitance, and adding more has negative effects, why would a rational person do that? Just always keep in mind the common sense aspects and you won't waste your money. Also remember to make the physical mounting as good as it was from the factory. Do not add electrical length to high current paths (= radiating antenna).
I do recognize you are trying to do a good job. Just trying to help.
-Chris
Actually, I wasn't being negative at all. What I am attempting to do is to let you know a better way to 1) spend your money
2) keep your work reliable
From experience working on equipment that has been modified by folks who only know what they read on the internet, many are more concerned with the "modification" and saying they did it than actually improving the equipment. You are spending time and money to create a better sounding bit of kit. Wouldn't you love to know where to spend your money and time where it really brings results? Wouldn't you rather have something that lasted years rather than a unit that was moved, then went poof!
The most dangerous change people make is to strap on extra capacitors somewhere and wire them into the existing (or new) capacitors. Mechanical mounting is not secure in 100% of the cases I have seen. This was even done to a Carver M 1.5t. That unit has supply voltages of +/- 125 VDC! The added caps were just flopping around against electrical tape. Some idiot even removed the meters on a Marantz 500, filed the openings and stuck two more caps in, bottoms hanging out. That amplifier is a classic and can put many other amps to shame, and some moron hacked it. No common sense.
I am not lumping you in that group (or I wouldn't even comment). I am trying to pass on the knowledge. Whenever you plan to do something, have a think on it. Capacitance is not a cure-all, and you most certainly can have too much of it. If you increase the capacitance a lot, you have to still charge the same amount of energy, but in a shorter length of time. This is referred to as "conduction angle" after how many degrees in 360 ° are the diodes conducting. Since the time is less, that means the peak current is higher, and since that is true, so are the IR losses in the transformer higher. Everything runs hotter, and you generate high frequency noise bursts, or make them worse.
So if the design has enough capacitance, and adding more has negative effects, why would a rational person do that? Just always keep in mind the common sense aspects and you won't waste your money. Also remember to make the physical mounting as good as it was from the factory. Do not add electrical length to high current paths (= radiating antenna).
I do recognize you are trying to do a good job. Just trying to help.
-Chris
Last edited:
Hi jooch,
Of course if you really want a head ache, the gain can be calculated. Too many places to make a mistake, but by measuring you have reality. The answer with no doubt.
-Chris
You can easily make an error calculating the gain. Why not just measure it? Set your frequency down to 120 Hz and feed that into the aux jack. Measure the level in on terminals 1 and 3, then measure the output with controls centered on pins 5 and 8. Divide the output voltage by the input voltage for each channel. Now you have the ratio, or gain. You could figure out in dB, but that is not useful information for what you are doing.
Of course if you really want a head ache, the gain can be calculated. Too many places to make a mistake, but by measuring you have reality. The answer with no doubt.
-Chris
Well i can't argue with that, that's why i'm sharing my idea's so i can filter out as much mistakes before any smoke rises! 🙂
That must've definitely felt like a crime scene, if i could afford something like that i wouldn't have bothered looking for upgrades.
Good info, i wasn't aware of the conduction angle. Also interesting you mention current paths becoming antenna's. What sort of problems would that create? More background noise?
In any case i'm not trying to increase capacitance at all but lower the ESR, i just happen to have a large stash of those 6800 caps. 😉
I read much i 'know' about caps on Rod's website, concerning this matter:
Last edited:
Indeed a better idea, i will run that test tomorrow. I have just finished up putting in the new resistors.
As has already been pointed out increasing the supply capacitance will cause a proportional increase in the charging currents and consequently for the saw tooth pattern (ripple) on the supply capacitors to be worse.
It is difficult to imagine how you would be able to fit the paired capacitors without compromising the layout through the length of wire needed to hook them up.
While pairing 6800 uF capacitor would decrease ESR the milliohm resistance in any wire length would mitigate against that. At the same time the ripple factor on the supply rails has been increased.
You should bear in mind that capacitor charging currents are not just what the transformer can deliver continuously which might be a few amperes. These charges occur in brief bursts and are magnitudes more than that. Heavy currents have implications for milliohm resistances.
Furthermore, a changing electrical field in any wire induces a magnetic field around it and charging currents rise and fall rapidly.
Magnetic fields are a significant consideration in the chosen route thickness and orientation of wires and PCB traces and of amplifier components.
As to wiring there are leads or traces to the capacitors and the bridge rectifier and from the transformer which is a wound component. If the current demands on these are increased then the magnetic fields will increase.
I think you should stick to the layout you have since this will have been chosen to minimize these influences.
Replace the main filter capacitors with new ones of the same value. The old capacitors will probably be down on capacitance through years of use. You would need access to instruments to measure this and any increase in ESR .
Taking a direct approach I expect fitting 10,000 uF 63 volt capacitors with their lower ESR should be possible since modern parts tend to be smaller.
You asked a question about an increase in background noise. In short YES - this could be electrical or acoustic from mechanical sources.
From what I can see, you have treated your project as more or less of an exercise in restoration. It is not a bad approach to take - it should give you the desired results without entering uncharted waters.
I usually try not to use more than 10,000uf.
You can get to the point where the 13 amp mains fuse blows due to the power supply capacitors being a short on power up.
I also got caught out a few years back with a irs2092 based class d amp.
I uprated the power supply to 2 off 20,000uf and the amplifier went haywire on power down. I would get siren type noises and then a huge thump through the speaker.
I emailed IR and they said the power supply was discharging too slowly and causing the problem.
In the end I designed a low voltage detector to hold the irs2092 in reset if the voltage was too low.
You can get to the point where the 13 amp mains fuse blows due to the power supply capacitors being a short on power up.
I also got caught out a few years back with a irs2092 based class d amp.
I uprated the power supply to 2 off 20,000uf and the amplifier went haywire on power down. I would get siren type noises and then a huge thump through the speaker.
I emailed IR and they said the power supply was discharging too slowly and causing the problem.
In the end I designed a low voltage detector to hold the irs2092 in reset if the voltage was too low.
Hi Nigel,
Elegant solution that sounds like something that should have been designed in. Good thinking!
Hi jooch,
mjona pretty much said everything I was preparing to say, and more. You can increase or decrease capacitance a little since the tolerances are pretty wide. I was only going to add that I often install some resistors in series before the bridge rectifier on my builds. You get a minor drop in voltage under power, but little bad current spikes compared to new low ESR parts. You are also very wise to have a look at ripple current values. This is also close to the maximum continuous current draw. But be aware that you have to balance the features off against other characteristics.
Best, Chris
Elegant solution that sounds like something that should have been designed in. Good thinking!
Hi jooch,
mjona pretty much said everything I was preparing to say, and more. You can increase or decrease capacitance a little since the tolerances are pretty wide. I was only going to add that I often install some resistors in series before the bridge rectifier on my builds. You get a minor drop in voltage under power, but little bad current spikes compared to new low ESR parts. You are also very wise to have a look at ripple current values. This is also close to the maximum continuous current draw. But be aware that you have to balance the features off against other characteristics.
Best, Chris
There is a Service Manual for Akai AM-2450 available from "Hi Fi Engine" and it costs nothing to set up an account and download material to be found there.
I don't know where you sourced your circuit diagram from however the manual describes how to set up the output stage standing current on page 19. The manual setting is 20 milliamps and adjustment to that level should have been possible using VR1 without altering R10.
The value of each emitter resistor for the output pair of transistors is 0.47 ohms and to restore to that setting, you needed to do a measurement across the pair in series with no speaker connected and the volume control fully anti-clockwise - this to read 19 millivolts when cold, and adjusted at intervals periodically until the amplifier had warmed up and the value is constant.
If your output stage transistors have higher current gain that will lessen the demand on driver transistors, nonetheless you would still not want these heating up.
The 2SC3421 and 2SA1358 substitute drivers need to have a similar current gain to the original parts to avoid having the output section increase the loading of the voltage amplifier - this would reduce both the loop gain and available feedback.
You would want the ones having the highest current gain - ones with the "Y" suffix.
It is normal to reset the standing current in an output stage whenever a replacement part is installed and if you tried the amplifier without doing that and it produced some oscillation the bias setting would have moved too low.
You can stick with a standing current around 50 milliamps. Although I cannot tell from the manual if the bias transistor is sited on the heat sink near the output transistors it apparently has a separate board which suggests this is so.
Last edited:
Hi mjona,
One more point. You often do not want too high a gain. Save high gain parts for current mirrors and current sources (with a red LED).
Did I get your attention?
If you use a distortion analyzer, run the amp into a 4 ohm load at 10 KHz and adjust for the minimum distortion, you should see your current sitting around 5 mA or lower. Distortion increases either way from this point. It increases more slowly as bias current is increased. To avoid the unpleasant low bias distortion, they set the bias current high enough so it sounds good shortly after turn on, and also for the worst case combinations of parts. That is a far cry from sound being better at higher standing currents though.
Keep this in mind. Semiconductor failure increases exponentially with temperature, x run time. That is true of almost all components I think. So the best thing you can do is make sure things only run as warm as they must. Increase ventilation too.
I am not picking on you, just trying to correct misinformation that exists on the open web.
-Chris
Not really. Only if you are driving the amp so hard that you are or near clipping on a 4 ohm load. Even then, it is not a critical issue unless the replacement parts have a much, much lower Hfe. In other words, those will be close enough.
One more point. You often do not want too high a gain. Save high gain parts for current mirrors and current sources (with a red LED).
I'm sorry, but I do not see the connection between low bias current and oscillation. Not even damped ringing throughout a 35 + year span as a tech.
NO! DO NOT RUN THE BIAS HIGHER THAN RECOMMENDED!
Did I get your attention?
If you use a distortion analyzer, run the amp into a 4 ohm load at 10 KHz and adjust for the minimum distortion, you should see your current sitting around 5 mA or lower. Distortion increases either way from this point. It increases more slowly as bias current is increased. To avoid the unpleasant low bias distortion, they set the bias current high enough so it sounds good shortly after turn on, and also for the worst case combinations of parts. That is a far cry from sound being better at higher standing currents though.
Keep this in mind. Semiconductor failure increases exponentially with temperature, x run time. That is true of almost all components I think. So the best thing you can do is make sure things only run as warm as they must. Increase ventilation too.
I am not picking on you, just trying to correct misinformation that exists on the open web.
-Chris
1.In terms of current to drive the output transistors that is true enough but that is not the point of my argument. I will save my breath on that for now.
2.There are no current mirrors or current sources here.
3. I referred to "bias setting" in relation to the voltage multiplier TR15 and VR1 consistent with the title of this thread "Modify bias circuit..." as distinct from the current flow through the output devices "standing current" which you call "bias current".
Referring back to Post #1 it was reported that the driver transistors had been changed and "some oscillations" were seen on the scope - no images were provided. In the next two posts discussion was about adjusting the bias to give a "standing current" of 55-60 milliamps.
In Post #4 the comment "no more oscillations" was made. I think there is a distinction to be made between a low standing current and a situation where the bias is not enough to bring the output stage beyond the cusp of conduction - in the last circumstance the problem could have been some low level incipient distortion or crossover spikes - perceived to be oscillation. It is easier to reply in the like term than go into a problem which has been resolved.
4.The recommendation to follow this track was made by "Mooly" the other moderator on this thread. I came on the scene when " the train" was leaving the station seemingly with everyone aboard. I drew attention to the recommended "standing current " level of 20 milliamps because that would suit an unmodified amplifier. I deferred to the apparent preferences of others in relation to an transistor upgrades and can understand their reasons for choosing to run a slightly higher current.
5. My pockets are not deep enough to justify the purchase of a distortion analyzer. I would set the standing current to according to the manufacturers recommended level and not fix anything that was not broken. Music is for enjoyment and one can spend far too much time playing around with equipment and it can become all consuming. I don't intend spending what remaining years I have ahead in this way.
6.My approach to the perceived problem with this amplifier would have been to consider the age and service life of the capacitors with a view to replacement. ESR and associated heating problems would be a primary concern as the electronics would have to live in an unnecessarily warmer environment.
7. I have a collection of trustworthy books on electronics that I can check back to.
Last edited:
I designed a 250WRMS bipolar output amplifier.
With very little bias it worked well.
However when I turned the bias up it oscillated badly.
In the end I had to add 10 ohm base stopper resistors to the output transistors to stop the oscillation.
It seems higher bias levels can push the output transistors into oscillation.
With very little bias it worked well.
However when I turned the bias up it oscillated badly.
In the end I had to add 10 ohm base stopper resistors to the output transistors to stop the oscillation.
It seems higher bias levels can push the output transistors into oscillation.
Just a quick update after being away for the weekend:
I replaced all the resistors with metal film 0.6W 1% types (except the larger ones and the FS's) and to my surprise noticed a big difference in performance.
Offset with the old resistors would level out in about 2 minutes to 60/19mV, with metal film under a minute to 19/2mV 😎
Noise wise there wasn't a noticeable difference, which probably means most is generated elsewhere. However i did find a noticeable reduction in sharpness (distortion?) and a better stereo image. I assume this is because of the tighter tolerances and thus better performance of the circuit?
I also experimented on the illumination of the VU meters. The old light bulb had burned out and finding a new one is difficult. From my experiences with the other one i own, the bulb is a bit too bright and the other problem is the VU meters are primarily lit In the top region, see this picture.
I used a big led which i had laying around an made some reflectors from aluminum foil, which ended up quite nice (i'll make some pics later on).
I replaced all the resistors with metal film 0.6W 1% types (except the larger ones and the FS's) and to my surprise noticed a big difference in performance.
Offset with the old resistors would level out in about 2 minutes to 60/19mV, with metal film under a minute to 19/2mV 😎
Noise wise there wasn't a noticeable difference, which probably means most is generated elsewhere. However i did find a noticeable reduction in sharpness (distortion?) and a better stereo image. I assume this is because of the tighter tolerances and thus better performance of the circuit?
I also experimented on the illumination of the VU meters. The old light bulb had burned out and finding a new one is difficult. From my experiences with the other one i own, the bulb is a bit too bright and the other problem is the VU meters are primarily lit In the top region, see this picture.
I used a big led which i had laying around an made some reflectors from aluminum foil, which ended up quite nice (i'll make some pics later on).
Hi jooch,
I like your light solution. I used a similar idea, but with a shipping noodle to diffuse the light.
-Chris
I like your light solution. I used a similar idea, but with a shipping noodle to diffuse the light.
Now, if you think that's cool, match some transistors. Matching output transistors can decrease open loop THD by 10X (from Motorola / On Semi application note).
Partially, yes. By changing some capacitors, and those further upstream you can really change the character of your amp for the better. Use Mica, polystyrene or NP0-C0G type ceramic capacitors. The other Poly type caps are better than electrolytic or tantalum capacitors. Doing that will put a big smile on your face.
-Chris
Hi mjona,
I'm not arguing here, I only want the facts to remain unclouded. By all means, enjoy the music. But if you will comment on the nuts and bolts, expect to have those comments evaluated by those who do this for a living.
Everything I have written here is intended to be factual. Things you can write down and refer to later in time.
-Chris
Okay, what's your point? I was simply giving you a complete answer as to where you would want to use high Hfe parts. This was for your benefit, and anyone else who is reading along.
Believe it or not, I'll answer something even though it might not be in strict definition of how you interpret a thread title. In fact, the idea that the bias circuit needs to be modified and was erroneous. This was mentioned earlier by you, see:
This is correct, and pretty much ends discussion about modifying the bias control circuit, no matter what people want to call it.
Higher bias current in any transistor tends to increase transconductance, and high frequency response. In fact, higher bias current tends to allow for oscillation or damped oscillation. This is one area of replacing one device type for another may deliver this nasty surprise. My comment to this was attempting to show you what the actual numbers for bias and crossover distortion would be. See:
I am telling you this because you are not likely to read this information from a book. You learn this from experience. Again, trying to give you some real world facts from the end of a soldering iron.
Moderators are just members with extra work to do. Unless we make it clear, we most often post as a member. We have to answer to the moderating team for our actions. There is some latitude in answering posts, as you can see from reading most threads. Our main goal is to give information that is accurate for our members.
My pockets were not either, but given that my vocation demanded one ... These days I am proud of the excellent grouping of equipment I have acquired over the years. With that equipment I have direct experience with most of the questions I answer, and if I don't, I say so. Audio service has been my vocation for over 35 years now. My beliefs have been verified over the course of my life and corrections made where they were needed.
ESR is only one failure mode. I use an HP LCR meter, hellishly expensive, to verify or condemn capacitors, inductors or resistors at higher frequencies (As opposed to DC values). Capacitors often lose capacitance too. ESR only tells part of the story, and now you know more about that,
Ahhh, but are the authors trustworthy? Some are. Some do not approach from the direct observation type empirical knowledge and depend on simulation or theory. Nasty problem, math can only describe and account for known variables. Not all variables are taken into account when some studies are made. Likewise, authors who do not directly work on the subject at hand may not be qualified to speak on it. Also, purely practical folks may also come to incorrect conclusions without a background that involves a teaching institution. It sure is nice when observations agree with theory.
I'm not arguing here, I only want the facts to remain unclouded. By all means, enjoy the music. But if you will comment on the nuts and bolts, expect to have those comments evaluated by those who do this for a living.
Everything I have written here is intended to be factual. Things you can write down and refer to later in time.
-Chris
- Status
- Not open for further replies.