taking it back,new scopes are naff
So..
The new scope i thought was great, is actually naff. My readings bottom out at -25dB no matter what i do.
Changing to an old Hameg which doesn't perform the labour saving measurement function, et voilà suddenly i have more like 60dB of useable range and despite measuring volts manually to a single d.p. my results are very close to my intended electrical response, within 1 dB.
Surprisingly considering all the rounding of values.
So..
The new scope i thought was great, is actually naff. My readings bottom out at -25dB no matter what i do.
Changing to an old Hameg which doesn't perform the labour saving measurement function, et voilà suddenly i have more like 60dB of useable range and despite measuring volts manually to a single d.p. my results are very close to my intended electrical response, within 1 dB.
Surprisingly considering all the rounding of values.
When i get back on my main pc ill upload some shots of the constituent stages and their separate elec curves.
From memory it is a combination of the RC and 1st Sallen Key stage (Q 0.618)
The RC at 500 Hz does most of the work. Without it the filter has a rising response, and a steep cutoff due to the Qs of the 2 pole stages, the driver breakup and baffle effects. In my passive setup I use a notch to help with the peak at breakup. In this active version in using higher order to get a steeper roll off. 4th order didn't seem to quite do the job, although i could've used a notch, but it would have to be tunable, and Boxsom doesn't sim them in active mode.
Included it since I expected to need baffle step compensation to give a flat acoustic output, as I do in my speakers with their conventional crossover.
You have however helped to spawn s small idea within my brain. A stepped attenuator to mix the RC influence with a summing amp...
From memory it is a combination of the RC and 1st Sallen Key stage (Q 0.618)
The RC at 500 Hz does most of the work. Without it the filter has a rising response, and a steep cutoff due to the Qs of the 2 pole stages, the driver breakup and baffle effects. In my passive setup I use a notch to help with the peak at breakup. In this active version in using higher order to get a steeper roll off. 4th order didn't seem to quite do the job, although i could've used a notch, but it would have to be tunable, and Boxsom doesn't sim them in active mode.
Included it since I expected to need baffle step compensation to give a flat acoustic output, as I do in my speakers with their conventional crossover.
You have however helped to spawn s small idea within my brain. A stepped attenuator to mix the RC influence with a summing amp...
Last edited:
So, some progress.
Note to self: should probably start a new thread documenting the rest
Initially I had some issues with the stripboard layout. I thought I had a dodgy solder joint somewhere, when I got around to taking a response plot of the HP filter.
As it turns out, moving the level pot in the HP from in front of the output buffer and to the front end of the input buffer solved the issue. Something which I am at a loss to explain.
So now it looks like I am goin to add another buffer stage ahead of the level pot (in case i need it) or simply just use a potential divider at the front end to set the relative levels of both filter sections.
Note to self: should probably start a new thread documenting the rest
Initially I had some issues with the stripboard layout. I thought I had a dodgy solder joint somewhere, when I got around to taking a response plot of the HP filter.
As it turns out, moving the level pot in the HP from in front of the output buffer and to the front end of the input buffer solved the issue. Something which I am at a loss to explain.
So now it looks like I am goin to add another buffer stage ahead of the level pot (in case i need it) or simply just use a potential divider at the front end to set the relative levels of both filter sections.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.