gepicaster
Member
Speaking of topologies, may I ask, which one from below you would recommend?
Why?
View attachment 106018View attachment 106019
Tape recorder project
- Thread starter EmilFrid
- Start date
Help Support GroupDIY Audio Forum:
I've now managed to make the first successful recording on my machine. Sounds surprisingly good given that the cards are not even in their shielded cases yet.
One thing I need to improve is the amplifier hiss. It's not horrific, but I'm sure the s/n ratio of the amp itself could be improved. I think I'll play around with lower value resistors in critical places, without losing too much gain. Will look more closely tomorrow and check whether it's the FET front-end stage or the opamp that follows it causing most of the hiss.
Will update soon again! Hopefully with some audio tests!
One thing I need to improve is the amplifier hiss. It's not horrific, but I'm sure the s/n ratio of the amp itself could be improved. I think I'll play around with lower value resistors in critical places, without losing too much gain. Will look more closely tomorrow and check whether it's the FET front-end stage or the opamp that follows it causing most of the hiss.
Will update soon again! Hopefully with some audio tests!
The key element here is knowing the inductance of the PB heads. It will determine what front-end is needed (FET or bipolar) and what operating point to choose.
Yes you're right. I forgot to write about this. The inductance is 50mH, and an ohm test shows roughly 120 ohms IIRC. As I said, it really doesn't sound too shabby at all, and the equalization is doing it's job perfectly fine. I just need to find ways of lowering the hiss a bit. I dislike excitement, but I felt excited upon hearing the quite successful recordings I did, consisting of various sinewaves and then a few verses of Neighborhood Bully, lol.
Definitely a low-Z head. About 1k-1.5k in the critical 2-6 kHz range. The classis answer would be a bipolar front-end biased at about 1mA. It could also be a VLN FET such as 2SK170 running at 2mA.
@abbey road d enfer thank you for helping me out! Yes, I am indeed using a 2sk170 in that mA range. I will double check in the coming days and make an update regarding the current (no pun intended) operating point.
As far as bipolar goes, I wonder if a Ne5532 front end would do a good job, as opposed to a discrete alternative.... I might look into that.
As far as bipolar goes, I wonder if a Ne5532 front end would do a good job, as opposed to a discrete alternative.... I might look into that.
Probably not. 5532 has an OSI (optimum source impedance of about 7 kohms.
If you want to use a monolithic opamp, you should look at AD797 or LT1028 (expensive).
Actually, the best option is using one of the dedicated mic preamp circuits (THAT1510/12, SSM2019), though implementing the EQ network may be somewhat tricky.
i'm leaning towards keeping the 2SK170 and reviewing the values of the EQ network.
@abbey road d enfer Never thought about the SM2019. Interesting idea. The eq network is incorporated in the following op amp stage at the moment, so I guess it'd be a matter of putting a buffer in between those two stages. But I agree to keep the FET. But making two PB cards with different topologies could be a fun learning experience. I've made amps and signal processors before, but interfacing a tape head is something very different if it's supposed to turn out really good.
Analog Pitstop
Active member
I am happy for your results!
Why not documenting the results&details with a medium class spectrum analyzer and share the results here (or PM if fancy you better)?
Plus WFGUI measurements, of course.
Why not documenting the results&details with a medium class spectrum analyzer and share the results here (or PM if fancy you better)?
Plus WFGUI measurements, of course.
@Analog Pitstop thanks s lot! Yes, I'll try to make some measurements once everything is finished. At least frequency and s/n.
Analog Pitstop
Active member
If I may suggest, also a S/N-like measurement at higher frequencies - 1KHz, 5KHz, 10KHz, 15KHz (hence on spectrum analyzer - as the dedicated audio-SNR instruments might not do these frequencies).
Only if you do not have something against publishing them. I am curious about Rec-head performance.
What tape and signal strength will you plan to use for these frequencies? The results of W&F will need to be referenced too, to correctly account for signal amplitude and noise...
Only if you do not have something against publishing them. I am curious about Rec-head performance.
What tape and signal strength will you plan to use for these frequencies? The results of W&F will need to be referenced too, to correctly account for signal amplitude and noise...
