Did I get this JFET Cascode right?

Hello everybody,

After installing two new op amps in my Rotel RCD 965BX (OPA627AP), I started to look into Biasing them into Class A. I saw threads about just adding a resistor but it has drawbacks. I also saw Tangensoft.com's page about this and decided to do the Cascode (after frying one of those new op amps testing the milliamps in parallel trying to understand the resistor method) Duh!

So I searched and learned what I think is the correct way to do this. I also made a drawing of how to place the transistors on the stripboard with wires or pins to connect to the op amps. Would any of you be kind enough just to double check me as I don't want to fry anything else except for eggs in the morning.

I will try to make this as easy as possible with all of the following:

Hyperlink to Tangentsoft: http://tangentsoft.net/audio/opamp-bias.html

Hyperlink to the 2N5484 JFETs I bought from Tangentsoft: http://tangentsoft.net/shop/

Hyperlink to OPA627AP datasheet: http://focus.ti.com/lit/ds/symlink/opa627.pdf

The attachment is the stripboard drawing!

I will also add the schematic of the op amps in the Rotel in the next post, I don't know how to attach more than one per post!

As soon as I get the Nod, I will do this or obviously will fix what I have done wrong.

Thank you all in advance! Hopefully some of you will use this too if it is right!

Regards//Keith

Here is the CD Player's Scematic of the op amps:

Thanks Again!

Regards//Keith

Hi Eric,

Thanks for responding, I was getting lonely and was wondering why nobody posted anything from "it is fine" to "are you crazy". I was wondering if JFET Cascodes and biasing op amps to Class A is a mystery subject. Everybody has heard of it but maybe not tried it yet.

Please keep in mind I am in the learning stage so I may be asking simple questions that most already know.

I have already replaced the 5534s with OPA627APs. Also, I have already removed the muting transistors and I can hear a difference. I don't know if you recommend removing L105, L106, C260 and C261 as well.

You suggest replacing R227 and R228 to 100K. Do you suggest Metal foil, 1% and what wattage? What does this do for the sound? Would you still do it with the OPA627APs?

Shunt R229, R230, R233, R234, you mean replace with wire jumpers? Again, what does this do and still do it even though I replaced the op amps?

The Wima Caps you recommend, I have seen many posts with people saying they like them. Would there be anything better sounding or is this the best choice? Again, do this with the new op amps? What will this do?

And lastly, what about the cascode, is it right? Why would you do the other mods first and not with the cascode? Taking the PCB out is not hard but has a few steps and I prefer to be cautious.

Thanks Eric!

Regards//Keith

Eric Juaneda said:

Standard metal film resistor (1%, 5%, 10%...) will be a good choice. 10K is necessary because of excessive 100µ capacitor value. It is not good at all for sound and OPA637 must work hard to drive this resistor. You must reduce unnecessary current if you want to bias op-amp in class A.

The best choice would be no capacitor at all. It would be possible if there are zero volts on pin 6 on the OPA627. Wima capacitors are very small; they have same size of 100µ electrolytic capacitors. Change is easy. Best capacitor are polypropylene capacitors, but they are very (very) big.

Theses components give only good noise measurement. Like R227, R228, R229, R230... all theses unnecessary components gives sound without excess: no details, unaggressive, no bass, no depth, sweet and nice foggy sound.

As you can read in the original and very interesting article from Warren Young, using cascode is certainly the best way. But using current regulating diode (CRD) would be easy. They're also small and easy to use.

How much do you bias your op-amp (1mA?, 2mA?)

Do you know input impedance of your preamplifier? (47K?, 20K?, 10K?). Optimal biasing circuit depend of input impedance.

Click to expand...

Thanks Eric and analog_sa!

So I am going to replace the two 100 uf caps for Wima MKS-2 2.2 uf. Should I use the 50 volt/10% or the 63 volt/5%?

Since I am changing the output caps from 100 uf to 2.2 uf, R227 and R228 should also still be changed to 100K? I am assuming the higher resistance keeps more of the voltage in the signal path to the output and allows less to short to the ground/common. I will use 100K ohm 1% 1/2 watt if this is still the best choice.

So unless you tell me not to, R227, R228, R229 and R230 are being replaced with wire jumpers

The output voltage from pin 6 is 6 volts on the newly installed OPA627APs. (is that a coincidence?)

So if I remove I105 and I106 (I or L) and C260 and C261, the sound should be even clearer and less veiled. If I decide to install a muting relay circuit later, will I need these parts?

I already have the transistors and stripboard to make the JFET Cascode and have the stripboard design above (see attachment in first post). It is easy to do and has the same two connections via legs/pins to the op amps and pin holes that the CRD would have. So this would be the easiest way for me with the best results.

I don't know how to test the how much biasing is on the op amp (1mA, 2mA) but in a thread started by Carlosfm about the OPA627 really sings, he suggests somewhere around 5mA (he was being generic, not necessarily focusing on CD Players output stages).

The input impedance of my preamp is 20K but this only a place holder for now until I buy a Modwright or something comparable. Some of today's pereamps have an input impedance of 220K so maybe I should target somewhere in the middle? Maybe you can offer some guidance on keeping this ready for most situations!

analog_sa, with all the other mods I am about to do, maybe I will do them and then clamp on the cascodes to the appropriate pins and listen to the difference both ways, back and forth for a few days before committing to a permanent soldering job!

Thank you for all your help and coaching. It is hard to get this kind of special treatment with everyone so busy these days. Teaching stuff like this takes so much of the few minutes you have for doing research on your own projects. I appreciate this and look for ways to help others in this forum, in my limited capabilities, hoping it comes back to the people like you!

Regards//Keith

Hi Chris,

I was wondering if you were looking at any of these threads. I saw a thread with you helping Shawn (TomWaits) for the 865BX and was hoping you would give me a push in the right direction too. Have you heard from him lately? I guess he is still buried at work. Anyway, I have also seen many threads you have helped in and many people raving about all your help! Thanks for this!

When you look at Warren Young's article he is referring to biasing a Pimeta amp and it is a lot more simple than the output stage in my CD Player. I was wondering if what he (and carlosfm) was saying was universal or changes with the complexity of the circuit you might be working on. I think you just cleared that up..... I, unfortunately don't know enough to make that connection or how to work with it.

So if I try these JFET Cascodes, I will definately just clamp them on for a while (or NOT) to see if they enhance the sound unless you think it definitely won't. If that is the case, too bad, I have spent a month on this subject alone several hours a day! YIKES

Thanks again!

Regards//Keith

Thanks Chris,

I referred to Warren Young's article without mentioning it is the one in the first post from Tangentsoft.com, sorry for leaving out the details.

If you can't sleep one night, LOL, the schematic for my CDP's output stage is in the second post on this thread. I have the whole manual if you ever need it! It is almost identical to the 865.

I also have the manual for the Rotel RCD 951 and the RHB10 and the Philips CD950 if you ever need them and the Marantz 2015.

Regards//Keith

Hi Chris,

Thanks for investigating the Tangentsoft.com article. It is funny, when you read a thread here that gets onto the subject of biasing op amps to Class A, someone always posts the hyperlink to that article. However, I never read any followup remarks that had your concerns. In fact, there were never any remarks about it period. That is probably why nobody ever responded to the JFET Cascode questions I posted on several different threads. maybe it is a mystery (as I said above) for the beginners and maybe even some of the experienced guys too!

Thanks for putting that reference (the article) in the proper perspective! Now off to the Walt Jung site to really investigate this subject. Half the battle of learning is looking in the right places! Maybe the month looking in the wrong place can be written off as a loss on my taxes!

Regards//Keith

Hey Chris,

I am going to test the cascodes just in case they do work (nothing permanent yet). I will give tangentsoft the benefit of the doubt for now. But if it doesn't work, I will also contact him. I did before to ask him if the transistors were the same no matter what op amp was used and also what about the resistor? He said this is the problem with Newbies. To read and learn it myself. So I did.... just on the wrong site (maybe). So if it works properly and is universal, extra points to him and if not, well that's a shame. Either way I will let everyone here know.

Now, Off to see the Wizard... Walt Jung's site and all I can say is I'll send post cards. I hope they serve coffee and have nice rooms while I am there for the next year or so!

Regards//Keith

Hi Chris,

Yeah, his business plan hasn't been thought out relative to his prime base or he has been overwhelmed by 50 guys like me a day. But that goes back into the proper model for this business, he needs to find a better way to keep up with correspondence, maybe by having an assistant! Also, as you say an assortment of articles, publications and hyperlinks to help his followers learn that he can easily attach to a return email. This will keep them coming back forever and refer him often.

Back to Walt. When I do a search on Google for "biasing op amps to Class A Walt Jung" you wouldn't believe all the forums it lists. Articles by Jung are by far the minority. Funny there are many threads from DIYAudio and they are very concise. When I do a search here on DIYAudio, the results are not so focused (I am being nice). Well it seems there are some experts on this site about this and I am tempted to email them to ask if what I want to do will work. From what I read here, it seems so.

I am still searching for all the articles by Walt Jung from his OP Amp Audio writings. I think this may have some answers.

BTW...The coffee is great here!

Regards//Keith