• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Major modifications of a Bogen AP-35

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
This amp could be compared to a sketch book for an artist or a journal for a reporter. It's my first tube amp and have been tinkering with it and learning as much as I could about tube circuit design and how the stuff works (or supposed to) So much smoke has come from this unit, but has also taught me so much! Below are the original schematic and then a rough sketch of what's in (or hanging off of) the amp currently.

An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


As stated in the schematic, it's definitely still a work in progress and probably won't ever be completed unless it winds up as parts for other amps or a "catastrophic failure" happens.

I'd also like to get some feedback, what do you think? What would you do differently? What would you add/subtract from it? It's currently driving a pair of Yamaha NS-10's and a Yamaha active sub matched to the NS-10's. It's fed by a Panasonic SV-3800 as a D/A converter for my Pro Tools LE rig I use for mixing.

Also tonight I've recorded some thoughts as to the sound of different 9V batteries used in the bias sections.

I have tried 4 different brands and 2 types of batteries in here and while I initially didn't think there would be an audible difference, boy was I wrong.

The 4 kinds tried were: Energizer Alkaline, Duracell Alkaline, Ray-O-Vac Alkaline and "Heavy Duty" Zinc Chloride.

I was completely taken aback when I found I strongly favored the cheap zinc chloride! These bad boys were $.97 for 2 compared to $5.80 for 2 Duracells and 2 Energizers!!!

Is the difference in sound due to the difference in the materials used? Impedence? What is going on here?

Finally, the original input/driver section was cathode/auto bias (Rk=4k, 22uf bypass cap) but found I preferred fixed bias for the difference in the low end and separation of the individual instruments in the soundstage.

A lot of the knowledge I used to do this came from this forum and the guru's who generously cast out their priceless wisdom to complete amateurs such as myself. I'd love to hear your comments. I look forward to learning from you.

Joe
 
Ex-Moderator
Joined 2004
Don't have time now to do a detailed analysis but a few observations (ignoring the Bogen schematic, which I can't read):

* the OP tubes must have grid resistors to ground or the bias will be unstable

* it is important also to provide stopper resistors for the control and screen grids of the OP tubes, to prevent parasitic oscillations

* the cathode resistors of the OP tubes should be bypassed with, say, 470uF caps

* you should get better performance if your drivers are configured as a differential amp; forget battery bias

* presumably a phase-splitter will be added at some stage?

PS you could do worse than buying and reading Morgan Jones's Valve Amplifiers. It will help to fill the gaps in your knowledge and understanding.
 
Last edited:
Hey Ray,
You know what? I completely left those off of the schematic by accident! Rookie mistake. Coming out of the parallel coupling capacitors there is a 470k resistor to ground for both 6L6's as well as 1k grid stoppers, 1 for each 6L6.

There isn't any stopper resistors for the screen, both the screen and plate are directly connected to the transformer. What is an optimal value for the screen stopper resistor? How would the absence of this resistor cause the circuit to oscillate? I'm not trying to say you're wrong or anything like that, I'm just trying to wrap my head around exactly what's going on in the output stage.

I'm not sure what you mean by getting better performance by configuring the driver stage as a differential amp and to forget battery bias. A phase splitter in this circumstance is not needed because the amp is being fed by a balanced line. The original circuit had an unbalanced input that fed a cathodyne phase splitter. It didn't make any sense to me to have to short low to ground to convert a balanced input to unbalanced only to pass through a phase splitter to balance the signal again! Thus, you have the circuit shown here.

Sorry about the poor quality of the Bogen schematic. Here's a link to a (hopefully) easier to read copy.

http://www.abdclub.com/Audio/BogenAP30.jpg

One thing I've never been able to find out is what exactly the difference is between "differential" and "balanced?" Are they one in the same or are there differences? What would be needed to convert this driver stage to differential? Also why do you suggest loosing the fixed bias in the driver stage?

I've read (sorry I cannot recall where) but in class A power stages, a bypass cap in the cathode is not needed. I cannot recall exactly why but something about the constant draw of current. Also if there was one needed, wouldn't bass reproduction be affected? The bass in this amp is very quick and punchy. Not a bit of sluggishness or "flabbyness."

Thanks for taking the time to reply to me. I'm looking forward to your detailed analysis of the circuit, now that you have the left out portions! All i can say is the circuit as it is currently is TERRIFIC sounding. Trying not to sound cocky but my line of work (well my hobby that is currently providing me income) exposes me to professional sound reproduction systems used for critical listening, and this amp I feel could hold up very well in one of these facilities. Of course I'm a bit biased in this case, but it is definitely the best sounding amp I have owned so far.

Again thanks and I look forward to your response.

Joe
 
Ex-Moderator
Joined 2004
Hi Joe, thanks for the link to the Bogen AP30. The differences between your amp and the Bogen are so great that I don't see much point in referring to it! You are using higher power OP tubes in UL, whereas the Bogen uses 6V6-equivalent 7408s in pentode-mode.

Thanks, also, for clearing up the issue regarding the missing grid resistors and stoppers, and I'm glad it's working really well for you.

The purpose of adding screen-stoppers is twofold: firstly, to prevent any tendency to parasitic oscillation; secondly, to lower slightly the screen voltage, to ensure it stays below the plate voltage. They're mostly precautionary and a value of 470 to 1k should be OK, although admittedly many UL designs omit them.

If you configure the drivers as a differential amp, it means that you use a shared cathode resistor ("tail"), as in the Williamson. This is generally better than separate balanced amplifying stages which you have now. It gives very good balance and low distortion. The bias will be provided by the tail resistor, so there will be no need of bias batteries.

In the OP stage, separate bias resistors should be bypassed so as not to increase the OP impedance, which would result in loss of power. An alternative arrangement would be to use a single shared cathode resistor of half the resistance of one of the single resistors you use now (i.e. 200 ohms).
 
Last edited:
Hi,

that's a very original schematic!

With the battery grid bias in the input stage, you need to make sure that the source you hook it up to (preamp, CD player, whatever) has output capacitors or transformers. Otherwise you will be forcing -9V upon its output stage, which it might not like.

In the schematic the filament supply for the #26 appears to be shorted out. You probably forgot 2 resistors from the filament pins to ground?

Your change in coupling caps from 9nF to 109nF is rather drastic... (just a thought)

The shared cathode resistor in the output stage is sth. I haven't seen very often, but it works. However, in class AB operation if one of the 6L6s cuts off, you will get degenerative feedback. That's because at that point, the increasing current in one tube is not cancelled anymore, in the cathode resistor, by the decrease in current in the other tube. This'll cause some compression of the signal peaks, depending on how deep in class AB you are biasing.

Enjoy it!

Kenneth
 
Last edited:
Ex-Moderator
Joined 2004
Kenneth, the shared cathode resistor in the OP stage is only there because Binaural mentioned Class A. With Class AB, separate bypassed cathode resistors would be required. Binaural's original schematic showed separate unbypassed resistors, which is not a good idea.
 
Kenneth,

Haha I found that out the hard way. I blew a chip in my source because I forgot to add the 9000pf cap on the input. That cap blocks any DC from making it to the source and keeps it in the stage. The addition of the .033uf cap was just an experiment to lower F3 because I felt the bass was a bit lacking in volume. This brought it in line with the rest of the spectrum.

This is basically an amp that I have been messing around with for a year or so now. Basically I'd read about something or see something in a schematic and I'd adapt it for this amp and so on and so on. Actually right now the input stage is completely different! It's still balanced, but with a 1k resistor to each grid and a switch to cathode biasing. 330R with a 10uf bypass cap then a .47uf PIO cap bypassing the electrolytic. Ra has been changed to around 45k (3 220k and 1 100k resistor in parallel because I didn't have any resistors with a high enough wattage rating for the rise in amperage.) and so far I feel it sounds a bit "tighter" compared to fixed bias. I'll see what I think in a week or 2 (if I don't get bored and change it again!)

The shared cathode resistor in the output stage is sth.
What did you mean by sth? Not sure I follow..

Thanks again for everyone's observations! Keep 'em coming! I'm learning how to wind my own chokes/etc so hopefully I'll be able to choke load the input stage soon!

Joe
 
The addition of the .033uf cap was just an experiment to lower F3 because I felt the bass was a bit lacking in volume.

This should say .1uf cap. I got 2 different projects confused for a second.

I forgot to address the 26 filament. Only one side should be connected to ground since it's using fixed bias. AH ANOTHER MISTAKE!! Man what a sad excuse for a schematic I drew....

Joe
 
Hi,

"sth." = "something", I was feeling lazy when I typed that :)

I think it's great that you keep experimenting. You will learn A LOT that way.

I could give you some more tips different ways to connect that #26 filament to ground, but first tell me if the filament supply is DC or AC.

Kenneth
 
The filament supply will be DC for the #26

Well then if you connect one end of the filament to ground, you should know that one end of the filament will 'see' 1.5V more bias than the other. Remember, in this tube the filament is the cathode. So you are really adding an average of +0.75V or -0.75V of cathode bias (sign depending on which filament pin you connect to ground).

Usually this is not a problem but keep it in mind if the tube's bias is a little different from what you expected. If necessary, just adjust the bias battery voltage accordingly.

If you really want to avoid this problem then you would need a filament supply with a center tap (2x0.75V for the #26) and ground the center tap. Or use AC filament supply, but this has the risk of causing too much hum.

HTH
Kenneth
 
Kenneth,

Even though the datasheet says the 26 can be run on either AC or DC, to date I have not been able to get the hum down to listenable levels using AC on the filaments.

I've been keeping tabs on the 26 preamp thread, but I don't recall the center tap idea you stated. Is this done by attaching a pot's resistive element to the tube and then driving it with the scraper? Since DC will be flowing through this, won't this cause a lot of scratching and wear the pot out prematurely? I guess you'd have to use a conductive plastic or wire wound pot for this purpose. Am I on the right track here?

Also about the difference in bias between the 2 sides of the filament, would using a high impedance filament supply (current source?) or is this only obtainable using AC? I've thought about creating a balanced preamp using the 26 using AC on the filaments and connecting them out of phase to cancel them out, would this work as well? Thanks for your knowledge, this is great!

Joe
 
[...]I've been keeping tabs on the 26 preamp thread, but I don't recall the center tap idea you stated. Is this done by attaching a pot's resistive element to the tube and then driving it with the scraper? Since DC will be flowing through this, won't this cause a lot of scratching and wear the pot out prematurely? I guess you'd have to use a conductive plastic or wire wound pot for this purpose. Am I on the right track here?

Yes, you can use a pot to create an artificial center tap, especially when using AC heating, because as you say pots don't like DC very much. The value should be high enough so as not to load the filament supply too much, but also low enough so as not to create too much cathode resistance (which causes unwanted cathode bias!). Typical value is around 100 ohms.

You see this 'humdinger' pot often in bigger AC-heated DHTs, such as 300B, where the small amount of cathode bias introduced by it is not a problem. On the #26, I'm not so sure it's a good idea, might create too much cathode bias.

Using AC, you could also have a center tap in the transformer winding. That would become your cathode connection (ground in your case). That would lower the hum compared to AC-with-one-end-grounded.

But if I were you I would stick to DC in a preamp tube ;)

Also about the difference in bias between the 2 sides of the filament, would using a high impedance filament supply (current source?) or is this only obtainable using AC?

A DC current source will not change that difference in bias, but don't worry too much about it. It's not a fault condition in any way, just something to keep in mind when you find your bias is not quite where you expected it to be.

Anyways many people prefer a DC current source over a DC voltage source so it's something you could try.

I've thought about creating a balanced preamp using the 26 using AC on the filaments and connecting them out of phase to cancel them out, would this work as well? Thanks for your knowledge, this is great!

Yes, this is actually standard practice in push-pull power stages. But you need to connect in-phase. Remember, a balanced or push-pull stage cancels in-phase signals and amplifies out-of-phase signals.

But again, for preamp tubes DC is probably the lesser of the two evils.

Kenneth
 
Kenneth,

Ha, humdinger... i love that word.

Anyways, yes, in phase is what i meant to say ;) I knew the consensus was the high impedance of the current source and the resulting sound was preferred over the voltage source, but because of the high impedance it would have to be located fairly close to the tube, right? Or since it's DC in this case does it really matter? Not really sure what filament circuit I'll try or if I'll try to come up with something on my own from the parts I have on hand (most likely case) and go from there. Thanks again Kenneth, the mind is slowly filing away this stuff.

Joe
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.