Hi there folks
I'm a newbie electroniciker type chap and looking at playing with my old denon amp (which I can't find any diagrams for).
I'm a mechanical engineer and fairly handy with my hands (no pun intended) and want to start looking at the biasing of the amp and replacing the preset pots etc.
I was reading some stuff regading biasing/checking/adjusting the bias settings on the output stages but after having a look inside my amp I'm getting a bit confused. Let me lay it down:
DENON 480r 50 W amp A/B design (I think). Not a particularly good amp in it's day, and hence a good start for my endevours.
Now looking at the output stage, we have (for each channel):
2SC4208a small transistors that are positioned in a hole drilled in the heatsink (not physically attached). The only preset pots in the whole amp seem to be attached to the collector of these transistors and the current thus measured is miniscule (microamps) with no input signal. Also seems to be a bit different for each side.
Then we have 1 each of 2SA1490 and 2SC3854 power transistors each side screwed onto the heatsink.
There is not any pot for DC offset control, guess it is done elctronically. (offset is negligable anyway, so I guess it is working as well).
So I guess my questions would be what to look for/do when adjusting the bias?
Should the current through the pots for each small transistor 2SC4208a be identical? Or has it been set this way to finetune another part of the circuit?
When adjusting the bias for the small transistors via the pots, am I only adjusting their outputs or does it cascade down the line to the Power transistors? At present the power transistors are both showing very identical idle voltages for each channel.
Any other advice or assistance much, much appreciato.
Cheers
Matt
I'm a newbie electroniciker type chap and looking at playing with my old denon amp (which I can't find any diagrams for).
I'm a mechanical engineer and fairly handy with my hands (no pun intended) and want to start looking at the biasing of the amp and replacing the preset pots etc.
I was reading some stuff regading biasing/checking/adjusting the bias settings on the output stages but after having a look inside my amp I'm getting a bit confused. Let me lay it down:
DENON 480r 50 W amp A/B design (I think). Not a particularly good amp in it's day, and hence a good start for my endevours.
Now looking at the output stage, we have (for each channel):
2SC4208a small transistors that are positioned in a hole drilled in the heatsink (not physically attached). The only preset pots in the whole amp seem to be attached to the collector of these transistors and the current thus measured is miniscule (microamps) with no input signal. Also seems to be a bit different for each side.
Then we have 1 each of 2SA1490 and 2SC3854 power transistors each side screwed onto the heatsink.
There is not any pot for DC offset control, guess it is done elctronically. (offset is negligable anyway, so I guess it is working as well).
So I guess my questions would be what to look for/do when adjusting the bias?
Should the current through the pots for each small transistor 2SC4208a be identical? Or has it been set this way to finetune another part of the circuit?
When adjusting the bias for the small transistors via the pots, am I only adjusting their outputs or does it cascade down the line to the Power transistors? At present the power transistors are both showing very identical idle voltages for each channel.
Any other advice or assistance much, much appreciato.
Cheers
Matt
Hi Matt,
Actually, the 480r receiver was better than most mid range units. The small transistors are bias (heat sensing) transistors. The pots set the current through the output transistors. Do not replace the bias controls unless they are bad. The bias current can be measured by looking at the voltage across the emitter resistors. Do not run the amps above the suggested bias current. It will not buy you anything and the heatsink is only so big. Just check and correct the bias if you must and enjoy the unit. Be very careful not to slip with your probes.
The current is not equal through the controls except by chance and the DC offset is corrected by a servo circuit. Given that you measure idle currents that are close in each channel, the bias currents are most likely within spec. for that unit.
-Chris
Actually, the 480r receiver was better than most mid range units. The small transistors are bias (heat sensing) transistors. The pots set the current through the output transistors. Do not replace the bias controls unless they are bad. The bias current can be measured by looking at the voltage across the emitter resistors. Do not run the amps above the suggested bias current. It will not buy you anything and the heatsink is only so big. Just check and correct the bias if you must and enjoy the unit. Be very careful not to slip with your probes.
The current is not equal through the controls except by chance and the DC offset is corrected by a servo circuit. Given that you measure idle currents that are close in each channel, the bias currents are most likely within spec. for that unit.
-Chris
Thanks for the reply Lowlevel - sorry for my long hiatus.
Yes, still keen - wherabouts?
Also I've found out that the L R channels (at speaker outputs) are slightly different in voltage levels when the amp is supplied with a constant frequency input. This could account for my little imaging problem. Tried to adjust the pots but it didn't make any difference to the levels.
Matt
Yes, still keen - wherabouts?
Also I've found out that the L R channels (at speaker outputs) are slightly different in voltage levels when the amp is supplied with a constant frequency input. This could account for my little imaging problem. Tried to adjust the pots but it didn't make any difference to the levels.
Matt
With the amp off..... Gently turn the pots with a metal screwdriver. You may be able to feel the dents in the carbon track. Centre the pot in the dent and you will be close to the current before you turned the controls. Then, adjust following the manufacturer's proceedure.
Many an amp has come my way after being at the "twiddlers" house first. This trick works well but no alchohol, no excessive tea, chocholate or coffee. You need to feel the motion gently.
Hope this helps you out.
-Chris
Many an amp has come my way after being at the "twiddlers" house first. This trick works well but no alchohol, no excessive tea, chocholate or coffee. You need to feel the motion gently.
Hope this helps you out.
-Chris
anatech said:
Lamp in series with the amp also works nicely if you have no clue where the bias pot should be. For a class-AB amplifier, choose a bulb thats 1/2 rated the one channel of your amplifier. So for a 50wpc amplifier choose a 25 watt bulb.
First you need to turn down the bias off, i.e. the side where the bulb is not lit . Then begin turning up each channels bias such that turning it up will make the bulb go a very dull (barely visible) dull amber with both channels biased.
Needless to say, You also want to make a note of the bias pots for reference before you twiddle...
So you will need to nudge each channel's bias pot a little till you get this level.
This is not the optimal level but will get you close so that when you turn on your amp, you do not blow the OP devices. As always choose the manufacurer's specs on what bias level is optimal.
PS: I do not agree with Chris's assertion that cranking up the bias buys you nothing...
but thats me. Chris is trying to make sure you have a reliable amplifier. You can "usually" crank up the bias till the heatsinks idle at no more than 43C-45C or 25C above ambient. With the amp playing at normal levels, the heatsinks should not go over 55C or 60C in short bursts.Fortunately...
I measured the Bias settings accurately before I twiddled so I can return them exactly to the same setting!
The problem is that adjusting them doesn't seem to do a bloody thing. I'm using a constant freq sine wave as input to measure output levels etc, but nothing seems to change when I tweak - no output level changes or anything, which means I can't even tell if the "new" settings are even balanced. As in my post above, the bias settings are slightly different to start with, so the only thing I can do is possibly use the same resistance ratio??
And I don't have the manufacturers instructions vis a vis this post in the first place.
Thanks Lowlevel, I will try and contact them via email (I'm actually living in Europe!).
Thanks again guys, and girls....
Matt
I measured the Bias settings accurately before I twiddled so I can return them exactly to the same setting!
The problem is that adjusting them doesn't seem to do a bloody thing. I'm using a constant freq sine wave as input to measure output levels etc, but nothing seems to change when I tweak - no output level changes or anything, which means I can't even tell if the "new" settings are even balanced. As in my post above, the bias settings are slightly different to start with, so the only thing I can do is possibly use the same resistance ratio??
And I don't have the manufacturers instructions vis a vis this post in the first place.
Thanks Lowlevel, I will try and contact them via email (I'm actually living in Europe!).
Thanks again guys, and girls....
Matt
I think you're measuring the wrong thing. It sounds as though you are just sticking a voltmeter on the output and measuring the DC offset. This is completely unrelated to what you cause when you turn the bias pot.
Unfortunately if I'm right and you measured the DC offset then you might as well bin the reading you wrote down.
You need to measure the current flowing through the output stage. Most amps have power resistors in the output stage and you should be measuring the voltage developed across one of these at a time, then use ohm's law to calculate the current. You should find that the voltage reading is quite close across 2 of the resistors and the other two 'pair up'. The paired resistors are on one channels output stage. THIS MUST BE DONE WITH NO MUSIC OR SIGNALS INPUT and NO SPEAKERS CONNECTED.
Of course if I'm wrong and you did the above then great
Unfortunately if I'm right and you measured the DC offset then you might as well bin the reading you wrote down.
You need to measure the current flowing through the output stage. Most amps have power resistors in the output stage and you should be measuring the voltage developed across one of these at a time, then use ohm's law to calculate the current. You should find that the voltage reading is quite close across 2 of the resistors and the other two 'pair up'. The paired resistors are on one channels output stage. THIS MUST BE DONE WITH NO MUSIC OR SIGNALS INPUT and NO SPEAKERS CONNECTED.
Of course if I'm wrong and you did the above then great
Hi richie00boy,
Exactly, except I measure across opposing pairs. I measure each one to get an idea of gain match without having to work too hard. Measuring across two will be more accurate with some of the less expensive meters.
K-amps,
The effect of bias current depents a lot on the circuit design and power supply. Most heatsinks in consumer gear are undersized and design marginal for listening. Higher bias current isn't a big help in these situations.
-Chris
Exactly, except I measure across opposing pairs. I measure each one to get an idea of gain match without having to work too hard. Measuring across two will be more accurate with some of the less expensive meters.
K-amps,
The effect of bias current depents a lot on the circuit design and power supply. Most heatsinks in consumer gear are undersized and design marginal for listening. Higher bias current isn't a big help in these situations.
-Chris
Chris,
I won't get into a debate on if high bias makes the sound better or not, needless to say it's always safer to heed manufacturer's ratings.
However experience has shown me that (other things notwithstanding) an output stage operating at 40C-60C sounds better than one operating 20c-30c. I do not know the reason, some people muse that the constant current helps, some say it is the optimal operating temperature of the devices, lastly high bias means some operation into class-A which has benefits too...
However having said that, who ever is cranking up the bias needs to be cognizent of the flip-side of doing so. I am pretty confident that is he keeps his temperatures below 45C (regardless of size) the OP devices will be fine. Now if Yamaha deliberately undersized the Trafo then that is another issue to be considered...
I won't get into a debate on if high bias makes the sound better or not, needless to say it's always safer to heed manufacturer's ratings.
However experience has shown me that (other things notwithstanding) an output stage operating at 40C-60C sounds better than one operating 20c-30c. I do not know the reason, some people muse that the constant current helps, some say it is the optimal operating temperature of the devices, lastly high bias means some operation into class-A which has benefits too...
However having said that, who ever is cranking up the bias needs to be cognizent of the flip-side of doing so. I am pretty confident that is he keeps his temperatures below 45C (regardless of size) the OP devices will be fine. Now if Yamaha deliberately undersized the Trafo then that is another issue to be considered...
Pretty much evey mass market manufacturer made transformers and heatsinks as small as they could get away with. Yamaha lost a TON here in Canada because the CSA thermal was 10°C lower than the UL rating. I'm sure they weren't the only ones. This amp came from around that time era.
Even now, look at what Pioneer calls an output stage! Small fan cooled heat tunnel with an SMT PCB on the bottom. Normally found blown to .. the devil.
Larger, more substantial amps like Adcom I will agree with you on. Not the other things out there.
-Chris
Even now, look at what Pioneer calls an output stage! Small fan cooled heat tunnel with an SMT PCB on the bottom. Normally found blown to .. the devil.
Larger, more substantial amps like Adcom I will agree with you on. Not the other things out there.
-Chris
Richie00boy:
I am a mechanical engineer, not an arts student! But I must admit I didn't study electrical circuits at uni so I am on a very steep curve.....
I'm measuring the AC output voltage of both channels with and without a load attached - from above post this amp has a circuit to deal with DC bias.
Am I wasting my time even looking at the AC output levels at the back of the unit?
I guess what I'm trying to get my head around is what EXACTLY to check when I twiddle with the bias. The only thing I don't have is a scope, but planning to get one soon.
I'm not even sure that the two bias pots even relate to the L R channels as they both seem to go to different parts of the "bias-sensing-transistors". ie to the collector on one and to the emittor on the other!
To be honest I'm trying to trace the circuits visually which is a real mess flipping it back and forth.
Also having difficulty finding the Bias resistors (to measure voltage to get current etc). Resistors etc ABOUND in this unit.
The fact that the bias levels are not even equal in the first place - at the FACTORY SETTING - makes it harder to know what to see after twiddling.
**If I keep the relative resistances equal after twiddling I should still have the same relationship to both channels afterwards right? (with a little more heat of course). I mean:
1) Bias pot on L channel is 1.39MOhm --> increase by x percent
2) Bias pot on R channel is 1.75MOhm --> increase also by X percent
??????
Matt
I am a mechanical engineer, not an arts student!
But I must admit I didn't study electrical circuits at uni so I am on a very steep curve.....I'm measuring the AC output voltage of both channels with and without a load attached - from above post this amp has a circuit to deal with DC bias.
Am I wasting my time even looking at the AC output levels at the back of the unit?
I guess what I'm trying to get my head around is what EXACTLY to check when I twiddle with the bias. The only thing I don't have is a scope, but planning to get one soon.
I'm not even sure that the two bias pots even relate to the L R channels as they both seem to go to different parts of the "bias-sensing-transistors". ie to the collector on one and to the emittor on the other!
To be honest I'm trying to trace the circuits visually which is a real mess flipping it back and forth.
Also having difficulty finding the Bias resistors (to measure voltage to get current etc). Resistors etc ABOUND in this unit.
The fact that the bias levels are not even equal in the first place - at the FACTORY SETTING - makes it harder to know what to see after twiddling.
**If I keep the relative resistances equal after twiddling I should still have the same relationship to both channels afterwards right? (with a little more heat of course). I mean:
1) Bias pot on L channel is 1.39MOhm --> increase by x percent
2) Bias pot on R channel is 1.75MOhm --> increase also by X percent
??????
Matt
If you were measuring AC output voltage with no signal present I'm surprised you got anything measurable. If you did then it's just the supply ripple breaking through. So in answer to your query, yes you are wasting your time measuring that.
Those resistance readings you gave also look wrong to me. There shouldn't be anything in your amp that measures in the megohms region.
Basically you need to look for 2 or 4 power resistors near the output transistors. These will be fairly big and maybe a white rectangular box or a green or dark coloured tubular design.
Non linearity on the output is exactly what the bias reduces, but you won't see it on a scope unless your really sharp.
Those resistance readings you gave also look wrong to me. There shouldn't be anything in your amp that measures in the megohms region.
Basically you need to look for 2 or 4 power resistors near the output transistors. These will be fairly big and maybe a white rectangular box or a green or dark coloured tubular design.
Non linearity on the output is exactly what the bias reduces, but you won't see it on a scope unless your really sharp.
The large resistances were from the actual pots themselves, measured across the only leg that is connected to the circuit.
The AC voltage was measured with the aforementioned constant freq input from a freq generator.
Now, don't forget that these pots are connected to the small "bias sensing" transistors and not to the 4 main power transistors that are fixed to the heat sink.
I'll have another look when I get back to France but I'm pretty sure it doesn't have any large power resistors anywhere near the output circuit. Certainly none of the high wattage white box type, I'm familiar with those from working on oilfield tools.
Matt
The AC voltage was measured with the aforementioned constant freq input from a freq generator.
Now, don't forget that these pots are connected to the small "bias sensing" transistors and not to the 4 main power transistors that are fixed to the heat sink.
I'll have another look when I get back to France but I'm pretty sure it doesn't have any large power resistors anywhere near the output circuit. Certainly none of the high wattage white box type, I'm familiar with those from working on oilfield tools.
Matt
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.