the app note's bridge-parallel servo version is a bpa200.
however there is a non-servo bpa200 (4 chips) in the app note at fig 7.2.4.
my non-servo bpa300 (6 chips) is quite similar to Alex's and .....
sure there is some warmup drift.
with quality components drift during warmup is extremely low.
here's what i'm getting for each chip's output offset:
cold:
0mv 0mv 1mv / 1mv 0mv 0mv
hot:
1mv 1mv 2mv / 4mv 3mv 2mv
the worst case delta on the first board is 1mv
the worst case delta on the second board is 2mv
the worst case net delta side to side is under 3mv (with one side hot and one side cold).
3mv into either 8R or 4R is less than one milliwatt.
i'm reasonably confident the chip can handle that. 🙂)
however there is a non-servo bpa200 (4 chips) in the app note at fig 7.2.4.
my non-servo bpa300 (6 chips) is quite similar to Alex's and .....
sure there is some warmup drift.
with quality components drift during warmup is extremely low.
here's what i'm getting for each chip's output offset:
cold:
0mv 0mv 1mv / 1mv 0mv 0mv
hot:
1mv 1mv 2mv / 4mv 3mv 2mv
the worst case delta on the first board is 1mv
the worst case delta on the second board is 2mv
the worst case net delta side to side is under 3mv (with one side hot and one side cold).
3mv into either 8R or 4R is less than one milliwatt.
i'm reasonably confident the chip can handle that. 🙂)
Last edited:
I don't have data, but seeing your 3 chips and how well they match during cold and during hot and while changing from cold to hot certainly looks very good and says a lot for the production spread from National chips.
Then I have to ask: "is this typical", or is it just lucky?
BTW,
it's not 3mVdc into 8r0, it's 3mV being fed back into a parallel pair of outputs.
Effectively ~ 3mVdc into 0r15+{0r15||0r15}
as you say in the next post
Then I have to ask: "is this typical", or is it just lucky?
BTW,
it's not 3mVdc into 8r0, it's 3mV being fed back into a parallel pair of outputs.
Effectively ~ 3mVdc into 0r15+{0r15||0r15}
as you say in the next post
yes, the first post's 4R and 8R roughly reflect the load during the offset R adjustment process and the 0R15 roughly reflects the load in actual use. BTW, all 6 chips were of the same production run. i have no data for chips of mixed runs. YMMV. also, unlike Alex's setup, i have the pots and the feedback circuit Rs mounted on the underside of the pcb where temps vary much less than above the pcb. YMMV.
Last edited:
So I realize this is an older thread, but here's an excerpt of the PDF manual that clearly confirms the JR amp as a bridged output design -
"Warning: Both positive and negative outputs are electrically
active with respect to chassis and/or system ground potential.
Therefore, this amplifier cannot be used in certain loudspeaker
switching configurations, such as those used in retail
demonstrations. Failure to avoid these precautions can result in
damage to the amplifier and will void the warranty."
There's no way you can get 160W into 8ohms out of any number of paralleled 3886s without a bridged configuration, given the max supply voltage limits. The output voltage swing is simply not enough to do so.
"Warning: Both positive and negative outputs are electrically
active with respect to chassis and/or system ground potential.
Therefore, this amplifier cannot be used in certain loudspeaker
switching configurations, such as those used in retail
demonstrations. Failure to avoid these precautions can result in
damage to the amplifier and will void the warranty."
There's no way you can get 160W into 8ohms out of any number of paralleled 3886s without a bridged configuration, given the max supply voltage limits. The output voltage swing is simply not enough to do so.
Interesting topic. Unfortunately the images are all deleted, because the upload was performed to the foreign portal and not here on diyaudio.
Additional I don't understand the follow marked records resp. terms:
1) Also he has connected the LM3886 in a non-inverted configuration and not inverted as many people claims on the Internet.
Both terms are the same operating mode for me. Mean you "Also he has connected the LM3886 in a non-inverted configuration and not an inverted configuration as many people claims on the Internet" ??
2) I don't want to discuss what method is best to use or is best sounding.
I just want to show that Jeff Rowland did not use any of these configurations.
This isn't possible, so I think, because what other reason could there be except the sound quality itself ??
For me the most interesting question is the exact topology (genuine schematic or service manual from JRDG) so as a circuit description from there (therory of operation), where I can read the pros and cons compared to the commonly used circuits.
Who can upload this information and files ??
Last edited:
A small precision
Hi Carl According to your post, sorry for the late but I would add a precision concerning the topology of the schematic: I don't thing that the configuration is x6 in //, but 2x3 in //. If you read carefully the notice page 13) you see that both outputs are active. This means the ground (black terminal) is not passive. So output is balanced. Here is the mystery. Of course in these conditions the rated output is possible. Dan
Hi Carl According to your post, sorry for the late but I would add a precision concerning the topology of the schematic: I don't thing that the configuration is x6 in //, but 2x3 in //. If you read carefully the notice page 13) you see that both outputs are active. This means the ground (black terminal) is not passive. So output is balanced. Here is the mystery. Of course in these conditions the rated output is possible. Dan
- Status
- Not open for further replies.