Hi,
Your not really asking a sensible question. T filters are simply 3rd order low
(LCL) or high (CLC) pass filters and there is lots of info on the internet.
TinaTi is a (great and simple) free circuit emulator that may help.
rgds, sreten.
T and Pi filters usually relate to discussing radio circuits.
Your not really asking a sensible question. T filters are simply 3rd order low
(LCL) or high (CLC) pass filters and there is lots of info on the internet.
TinaTi is a (great and simple) free circuit emulator that may help.
rgds, sreten.
T and Pi filters usually relate to discussing radio circuits.
Last edited:
I have and use LTSpice but I'm not sure how to create and plot a frequency sweep in that software. I suppose I could do a ton of single simulations and plot a curve, but I haven't tried to start that process yet and it might be a big waste of time.
I was referring to a RCR low pass filter. You would be surprised how little information is on the internet about these filters, at least as much as I can find. I will check out TinaTi.
IMHO, LTSpice has a terrible interface and is very difficult to use. Simetrix is much simpler and easier to use. I don't know how it compares to Tina as I haven't tried that one.
p.s. Avoid Circuitmaker. The free version doesn't work properly.
Wikipedia to the rescue!
Last edited:
Hi,
RCR isn't much of a low pass filter. Assuming low source impedance and
high loading impedance its the same as an RC, the second R makes no
difference, typically the case for audio.
In radio circuits where your matching impedances its a little more complicated.
The R in the "RC" calculation is the (Zsource+Rin)||(Zload+Rout).
TinaTi is dead easy to use, and automatically does frequency analysis.
rgds, sreten.
Last edited:
You need to do an "ac" analysis. You will specify the start and stop frequencies and number of points in between. The software takes care of the rest. This is generic spice stuff, not specific to LTSpice.
You are correct; I downloaded TinaTi and within 5 minutes I have the frequency response of my filter. It seems that the second R has no effect, which surprised me but not you.
Any tips for designing a low pass filter to go between DAC and preamp? The best I can come up with so far is cascading RC filters, with 10x the resistance and 0.1x the capacitance going down the line.
Hi,
I'm confused. A DAC with analogue output to a preamp should have
inbuilt reconstitution filters to prevent it outputting hf garbage.
Look at LC, LCLC, LCLCLC etc filters for a sharper cut off,
noting that they need designing for a known load impedance.
rgds, sreten.
The R in the "RC" calculation is (Zsource+Rin)||(Zload+Rout).
If you understand this you shouldn't be surprised. high Zload makes
the second parallel term and Rout irrelevant, and low Zsource means
it reduces to simply the value of Rin and the capacitor.
I'm confused. A DAC with analogue output to a preamp should have
inbuilt reconstitution filters to prevent it outputting hf garbage.
Look at LC, LCLC, LCLCLC etc filters for a sharper cut off,
noting that they need designing for a known load impedance.
rgds, sreten.
The R in the "RC" calculation is (Zsource+Rin)||(Zload+Rout).
If you understand this you shouldn't be surprised. high Zload makes
the second parallel term and Rout irrelevant, and low Zsource means
it reduces to simply the value of Rin and the capacitor.
Last edited:
The DAC in question uses WM8741. The datasheet states that it has an internal analog filter with -3dB at 474 kHz and also says "All performance measurements are done with 20kHz low pass filter. Failure to use such a filter will result in higher THD and lower SNR and Dynamic Range readings than are found in Electrical Characteristics. The low pass filter removes out of band noise; although it is not audible it may affect dynamic specification values."(http://www.wolfsonmicro.com/documents/uploads/data_sheets/en/WM8741.pdf
I took this to mean that there will be IMD if there is no low pass filter used on the outputs. It was my impression that most DACs operated this way and that the Akamai series was perhaps alone in not needing a low pass filter on the analog out. Page 62 of the datasheet shows an active opamp output stage that includes several RC filters but I was trying to go completely passive at first for simplicity's sake.
Unfortunately I don't perfectly understand what you mean by (Zsource+Rin)||(Zload+Rout), particularly the || part but I have just been using the 1/(2*pi*RC) to estimate -3dB for the filter. Apparently your equation somehow reduces to mine anyway. I will check out LC filters and report back but this might be stretching my expertise a bit. The load impedance should be, I believe, close to 50kohms.(A B1 with a 50k volume pot)
I took this to mean that there will be IMD if there is no low pass filter used on the outputs. It was my impression that most DACs operated this way and that the Akamai series was perhaps alone in not needing a low pass filter on the analog out. Page 62 of the datasheet shows an active opamp output stage that includes several RC filters but I was trying to go completely passive at first for simplicity's sake.
Unfortunately I don't perfectly understand what you mean by (Zsource+Rin)||(Zload+Rout), particularly the || part but I have just been using the 1/(2*pi*RC) to estimate -3dB for the filter. Apparently your equation somehow reduces to mine anyway. I will check out LC filters and report back but this might be stretching my expertise a bit. The load impedance should be, I believe, close to 50kohms.(A B1 with a 50k volume pot)
Hi,
|| simply means in parallel, rather than writing out the more convoluted
equation for parallel, its obvious a high term with a low term in parallel
that the low term will dominate.
The reference to a 20KHz filter is the same as the bandwidth used to
measure amplifier distortion, the lower it is, the better the numbers.
The actual -3dB bandwidth varies with filter mode and here I'm getting
a feeling of specmanship rather than telling you how to use the chip.
I can't wade through the whole document trying to work out what
it is seems to me is being obfuscated, I suspect its simply contrived.
rgds, sreten.
|| simply means in parallel, rather than writing out the more convoluted
equation for parallel, its obvious a high term with a low term in parallel
that the low term will dominate.
The reference to a 20KHz filter is the same as the bandwidth used to
measure amplifier distortion, the lower it is, the better the numbers.
The actual -3dB bandwidth varies with filter mode and here I'm getting
a feeling of specmanship rather than telling you how to use the chip.
I can't wade through the whole document trying to work out what
it is seems to me is being obfuscated, I suspect its simply contrived.
rgds, sreten.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.