Thank you.I recommend running a standard loop gain sim. ............................and just verifying the transient response in the lab...........
Hi,
Finally I build this for heatsink :
with alu bare 4 mm thickness, weight is 56 gr, side is 220 gr, total 276 gr, seams good enough for "low power"..
Thermal paste between bare and side case.
Phil.
Finally I build this for heatsink :
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
with alu bare 4 mm thickness, weight is 56 gr, side is 220 gr, total 276 gr, seams good enough for "low power"..
Thermal paste between bare and side case.
Phil.
I'd suggest at least 6 bolts to thermally couple that plate to the chassis side panel.
Don't hide the bolts behind the tall capacitors.
Looking at the pick, two to the right hand side just beyond the tall cap.
one below the red cap under the PCB, and use the bolt through the chip.
two just to the left of the left most tall cap.
Due to the position of the tall caps, there is no room to access bolts for a chip clamp. You have to use the single bolt. or hide the clamp bolts behind the tall caps. Studs (all thread) screwed into the heatsink and nuts that are accessed with a spanner would allow the use of a 2bolt clamp, but this is quite awkward to assemble/dismantle.
Don't hide the bolts behind the tall capacitors.
Looking at the pick, two to the right hand side just beyond the tall cap.
one below the red cap under the PCB, and use the bolt through the chip.
two just to the left of the left most tall cap.
Due to the position of the tall caps, there is no room to access bolts for a chip clamp. You have to use the single bolt. or hide the clamp bolts behind the tall caps. Studs (all thread) screwed into the heatsink and nuts that are accessed with a spanner would allow the use of a 2bolt clamp, but this is quite awkward to assemble/dismantle.
Hi,
Mounting test to see if everything is ok :
So let's go to drill front and rear panel !!
Phil.
Mounting test to see if everything is ok :
An externally hosted image should be here but it was not working when we last tested it.
So let's go to drill front and rear panel !!
Phil.
Hi,
Schematic post one was edited, adding Rf2 and Cf.
So Cc, Rf2 and Cf are under pcb :
Phil.
Schematic post one was edited, adding Rf2 and Cf.
So Cc, Rf2 and Cf are under pcb :
An externally hosted image should be here but it was not working when we last tested it.
Phil.
Hi,
Schematic post one was edited, adding Rf2 and Cf.
So Cc, Rf2 and Cf are under pcb :
An externally hosted image should be here but it was not working when we last tested it.
Phil.
Hi Philfr,
What are the sound changes when you add these?
kp93300
Hi,
This is Tom advice (post 9), he said it's for "better stability near clipping", at this time I don't ear any difference at sounding.
But with test setup (too small heatsink), amp works only with few watts..
When amp was finished, I will try with high level, and maybe I could tell more.
Phil.
This is Tom advice (post 9), he said it's for "better stability near clipping", at this time I don't ear any difference at sounding.
But with test setup (too small heatsink), amp works only with few watts..
When amp was finished, I will try with high level, and maybe I could tell more.
Phil.
Either way, use single bridge rectifier. Why incur four diode drops?
A "single bridge rectifier" is nothing more than four diodes in the same package....
Roscoe
A "single bridge rectifier" is nothing more than four diodes in the same package....
Roscoe
No kidding...
Hi,
Mounting is completed :
Phil.
Mounting is completed :
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Phil.
Hi Philfr,
What are the sound changes when you add these?
kp93300
About Rf2 and Cf, it appears that high frequencies give more clean details, regardless of the level..
And again, thank's Tom !
Phil.
No kidding...
If you know that, as you claim, then your comment about four diode drops is nonsensical...
If you know that, as you claim, then your comment about four diode drops is nonsensical...
In the OP's diagram there are two bridges. There are two diodes in conduction on peaks of each half cycle to charge the filter cap on each of the supply rails or four diodes conducting in total.
If he bonded the the secondaries together in the middle to create a center tap, he can use one bridge and have a total of two diodes in conduction on each half cycle.
I don't understand the reason to use two bridges. If there is a benefit, someone please explain. The way he has it requires an extra part (diode bridge) and the extra two diode Vf loss.
Sorry, I misunderstood what you were trying to say. You are indeed correct in that case.
Of course, that begs the question of why we're worried about that extra Vf loss? PS voltage is going to change much more than that over normal line voltage variations. It's not going to have any affect on PS regulation either, since Vf is essentially constant over a diode's normal operating current.
Roscoe
Of course, that begs the question of why we're worried about that extra Vf loss? PS voltage is going to change much more than that over normal line voltage variations. It's not going to have any affect on PS regulation either, since Vf is essentially constant over a diode's normal operating current.
Roscoe
Sorry, I misunderstood what you were trying to say. You are indeed correct in that case.
Of course, that begs the question of why we're worried about that extra Vf loss? PS voltage is going to change much more than that over normal line voltage variations. It's not going to have any affect on PS regulation either, since Vf is essentially constant over a diode's normal operating current.
Roscoe
Sure the line voltage will fluctuate but why incur an additional loss on top of that (plus the extra part)? Losing about 3 volts across the rails isn't that significant to the ear, but the amp takes a hit in output power. For example, a 50 watt amp becomes 45 watts if it loses just one volt rms of output swing.
Last edited:
He did say in the original post he only needs 5-10W... Rail-to-rail voltage decrease should be more like 1.5v. He's only going to have about +/- 18v with that transformer. Higher rail voltage will give a higher maximum power out, but does it necessarily sound better? The difference between 50w & 45w is <0.5dB, not particularly meaningful...
Roscoe
Roscoe
It is usually too expensive to bottleneck a very large transformer with a single bridge rectifier; however, somewhat more cost effective to use a more modest size transformer with dual bridge rectifiers.I don't understand the reason to use two bridges. If there is a benefit, someone please explain.
Shorter answer: More effective transformer utilization; Current.
Provided the internal components are not overstresssed by the higher supply voltage, it is my opinion that the higher supply voltage leads directly to better performance.He did say in the original post he only needs 5-10W... Rail-to-rail voltage decrease should be more like 1.5v. He's only going to have about +/- 18v with that transformer. Higher rail voltage will give a higher maximum power out, but does it necessarily sound better? The difference between 50w & 45w is <0.5dB, not particularly meaningful...
Roscoe
This is because many of the circuits in an amplifier are ClassA and many of these ClassA stages are single ended. For this type of operation, accuracy is improved by using less of the total available voltage swing. There are many websites that explain this and even recommend using a lesser part of the total available swing just to get that improved performance.
Once this is done, one finds that the NFB has less work to do in trying to reduce the errors at the output.
There is a separate issue that leads to improved performance with higher supply voltages.
LESS CLIPPING of the output signal transients, when the supply voltage is well above that required for average signal levels.
It is usually too expensive to bottleneck a very large transformer with a single bridge rectifier; however, somewhat more cost effective to use a more modest size transformer with dual bridge rectifiers.
Shorter answer: More effective transformer utilization; Current.
Nonsense! Just how is the single full wave bridge not utilizing the transformer to the fullest if the two secondaries are joined in the middle to make it like the center tap? Current flows in both secondaries on both half cycles. Current would be the same but with two extra diodes per half cycle with two FWB.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Amplifiers
- Chip Amps
- LM3875 Chipamp