I'm busy putting together a layout to incorporate the two gain cells I've presented. I'll also throw in a lineamp, - an all-in-one preamp board, as it were...
PN4391 sources -
Mouser has them, made by Central Semi. Some non-Chinese Ebay seller is currently offering NOS Vishay/Siliconix PN4391. I just tried a simulation using the PN4091, and it appears to be an acceptable sub.
Mouser has them, made by Central Semi. Some non-Chinese Ebay seller is currently offering NOS Vishay/Siliconix PN4391. I just tried a simulation using the PN4091, and it appears to be an acceptable sub.
Last night I nailed down some NOS Vishay/Siliconix PN4391s and some National PN4091s. Layout is in the last stages of tweaking and checking. I may go out for boards this weekend. I'm also considering getting some Linear Systems PN4391s, as they use a different process with higher transconductance.
produced boards for cascoded version.. matching in progress of 2SK2394 's, Idss is from 12-17mA so Rs=20ohms is used for bias..it can be changed to any other value very fast because is in SMD 1206. PCB is accepting any other fet SMD or TO-92 package, so i can experiment with anything from the shelf
Attachments
Here is one board, almost completely stuffed. I still need to select some JFETs and capacitors, and rustle up a few more SMB connectors for the ins and outs.
One board = one entire preamp, with 2 channels each of RIAA and line stage. The heat sink to the left is for the onboard discrete shunt regulator, that will supply +30V for the main supply.
One board = one entire preamp, with 2 channels each of RIAA and line stage. The heat sink to the left is for the onboard discrete shunt regulator, that will supply +30V for the main supply.
Attachments
The too many projects syndrome is one of my biggest weaknesses. I can relate.
I'm looking for a decent riaa solution. I don't want to spend ridiculous money on what should be a fairly "simple" circuit. (Yeah, right !?) Not easy but can be done. There's usually no way around a good MC setup trans to keep noise down. An extra 20 dB of gain in front can really mess things up noise wise. Plus, the trans will protect the cart from any DC offset, like some direct coupled fet stuff can have. The fewer caps in the circuit, the better for me.
I'm looking for a decent riaa solution. I don't want to spend ridiculous money on what should be a fairly "simple" circuit. (Yeah, right !?) Not easy but can be done. There's usually no way around a good MC setup trans to keep noise down. An extra 20 dB of gain in front can really mess things up noise wise. Plus, the trans will protect the cart from any DC offset, like some direct coupled fet stuff can have. The fewer caps in the circuit, the better for me.
I'm kinda scratching my head, as I declared the board finished, but for the life of me, can't find it. Maybe it's already lurking at my workplace - I'll check when I get in on Monday.
Couldn't find the finished board ore the other bare boards. What i did instead was to redesign the output buffers (new layout) for 5X gain with bias adjust. Those are in fab as we speak at PCBway. I'll have another go at construction when they arrive.
Boards are here - I'm kinda frosted that I have to go through this whole exercise yet again, but I'm happier with the line amp design on the new PCB.
