Any way not to blow a Palmer PDI-03 with 4 x KT77s?

[overtone]

I still have a number of 4 x KT77 output tubed guitar amps which I would like to run into the Palmer PDI-03, but it is only rated maximum 100 Watt input.
(The Palmer PDi-03 takes the speaker out signal and lets you silently "DI" into the "desk" adding a 4x12 stack flavor without getting your flared trousers in a flap or waking the neighbors.)

If I remember correctly the amps were being marketed as 140 Watts, but I am unsure. I do not have a scope yet and have never tested the output of an amp myself.
It would be nice not to blow the Palmer, then risking the KT77s and the Partridges. I suppose I could pull two tubes, but I hear more going on in the quads and that is what it is all about.

Some figures:
Plates are at 465V, G2 tracks about 3V below that. (1,5k G2 resistor)
These are the GEC / MOV 1970's KT77s.
From the data sheet I see that Pa is 35W absolute and 28W max.
I was biasing them, I think conservatively, around 36-37mA, with a 1Ohm resistor:
28 Watt / 465V plate x 0.62 = apr. 37mA.

Any ideas, just keep the volume down, or just forget it and let them gather dust?

[dai h.]

might be okay, but be aware I've seen some (apparently older) units with less power handling (three instead of four power resistors inside--it's possible they are marked differently on the back panel). In mine (should show up in the link below) the power resistors are four 50W rated units mounted on a big aluminium heat sink (the chassis has holes to help with the power dissipation, so I would make sure to leave some space above and below it if it's in a rack). IIRC Palmer stated the 100W rated PDI-03s were safe to use with 100W output amps, and with four 50W resistors (4*50=200W) as the main internal load it does appear to be capable (to compare, a Marshall PB100 Power Brake uses one 100W resistor which is fan cooled). That said, I think it's risky to use since there is an inductor in series with the power resistors which is there to simulate the series inductance of the speaker (varies depending on speaker--may be 0.5mH to 1mH or thereabouts). The problem as I understand is that the rise in impedance at high frequencies the inductor gives is not limited by a parallel resistance and can make an amp more prone to oscillation (and perhaps compromise the insulation of the output transformer primary winding potentially leading to some of the reported problems with attenuators blowing up amps). IME you do want the inductor there (sounds deader and flatter with it shorted out), but for safer operation it looks best to limit its effect with a resistor paralleled across it (for example the Marshall SE100 Speaker Emulator which uses pretty much the exact same load as a Power Brake uses a 33ohm R across the 1mH series inductor). (I'm guessing here, but possibly the THD attenuator uses an inductor that has a higher resistance so the resistor is "built in" with the inductor with the inductor mounted to a heat sink--I have a resistor hooked up in parallel with the 1mH in the load in my Power Brake and this resistor does get warm.) Problems that could arise with putting one in in the PDI-03 though is that 1) there isn't a whole lot of extra space inside the 1 rack unit space housing, and 2) do you have good metal working skills?(I don't), and 3) if you decide to sell it later on, you could be reducing its value (though increasing safe use and functionality). (I haven't used mine in a while, but) I was using my Power Brake (with the added R across the series L) as the load instead of the PDI-03 internal load.

Also, if you use the PDI-03 and crank up an old amp which may not have meant to be used maxed out (and may not possibly be in the best condition) this puts greater stresses on the amp and increases the chance of failure, so this may be something to consider if you are using old valuable amps with old valuable tubes.

Also, the outputs are transformer coupled, so best results may be had by paying attention to the input impedance of the preamp the PDI is being connected to, since it's possible to get high frequency ringing depending on the way the output is loaded (can result in poor audio).

[overtone]

Thanks, I remember your PDI-03 innards thread from way back. This unit is a great way to justify keeping too many valve amps in the house!
It is the new version of the 03, I will pop it out of the rack today and shoot a gut shot or two for you.
The output of the Palmer usually goes into a Session i/O which offers three choices input impedance: Line 44 kOhms unbalanced & balanced / Mic 4.8 kOhms Balanced / Instrument 1 MOhm Unbalanced.

[dai h.]

cool.

re: the input impedances, the two that you would normally use for the Palmer's line outs (44kohms and 4.8kohms) might be on the high side. In order to load the Palmer output (specifically, its transformer), all you need is a resistor (since you don't need it to dissipate much, maybe stick it in a connector and mark it once the value desired it worked out?) and a bit of calculating. As I understand, the possible ringing can happen for most transformers (some might be more impervious to differences but those are likely to have a more complicated winding and more expensive) and could be something overlooked if one isn't aware. You can check out the links I found (re: ringing) here if you like:

Marstran.com • View topic - low budget cheapskate recording tips

oh, and lastly, these are pics of my Power Brake with the added resistors. Simply put there are parts of the load that mimic what a speaker does (the impedance rise at high frequencies and the low frequency resonance), and the resistors (variable ones for experimenation) just allow the effect those portions have to be controlled better (from "off", or shorted, to "max", or open(ie, no R)).

[overtone]

Thanks, for the helpful link, I really need to get up to speed on the loading of inputs. I only became aware of the subject re. SM57 microphones a couple of weeks back.

I am now fairly sure all these amps are "rated" 100W. One Partridge OT is marked "OP100" and I can just make out 100W on a chassis label. The others either have no markings or are obliterated. One does have a PT and OT much bigger, by nearly half, but it is the exact same circuit and voltages. This may work if I am careful.

Here are the gut shots of the 2010 model PDI-03. The load resistors are 4 x 27R 50W. When you connect speakers to "thru" the the load resistors and inductor are out of circuit.



I have a few options to play around with now.

[overtone]

Kind of off topic, but comparing the gut shots of these two models reminds me of seeing two late 1970s V8 engine bays open at the same time. One was a Rolls Royce and the other was a Porsche. The Marshall here reminds me of the Roller and the Palmer reminds me of the Porsche.

[dai h.]

hey thanks for the pics. Looks like mine pretty much, save for the connectors (evidently to ease production versus soldered connectors).

re: the wattage of the amps, one thing you could try if you have a Variac (or some means to turn down the AC wall voltage a bit) is turn down the input voltage to the amp a bit to ease the conditions (especially if you like cranking it up all the way), Maybe measure the heater voltage and go 5 or 10 percent down from there.

oh, and another re: the whole attempt to get a nice guitar amp sound in the cans, is that different headphone amps sound different (sorry if that's known and obvious already ). I have an old Yamaha pro mixer, and that sounds better compared to a Tascam 488mkII (old cassette Portastudio I usually used for the mixer section and headphones). The Yamaha has less noise (hiss) and just seems nicer to listen to. My understanding is that the better upmarket studios have really nice cue ('phones) systems. Makes sense for musicians since the ear to hands (or whatever appendages) "feedback loop" is very important for the sound adjustment. Loading of the headphone output amp can have an effect on the sound also, so might be useful to know that different headphone impedances can change the sound.

Finally, if you live somewhere you don't have to be super quiet, an attenuated 4x12 cab might be more satisfying. I'm not in that situation myself, but I have heard some nice sounding sound clips from people who are using say a Power Brake and further series attenuation (the simplest kind can be undertaken with some inexpensive power resistors) which helps to lower the acoustic output even more. Or, a combination of things such as master volume plus attenuator, or dummy loading into another solid state amp and so on.

[overtone]

Thanks for the follow up with more tips.

if you live somewhere you don't have to be super quiet, an attenuated 4x12 cab might be more satisfying...

Well I actually have to be silent! I used to rent a heated warehouse with no neighbors from the port authority, but now everything is at home in the city center. The Palmer lets me silently use a motley collection of tube amps and at the moment I am so satisfied that I have put my isolation-room project on hold.
The "desk" is a Macbook running Logic Studio and a simple Guitar Rig Session converter plus a few very good headphones. I have not yet had a hearing issue when tracking with this set up - which was not always the case in a studio, so I guess I got lucky with the headphone outs.

I have been digesting those links that you posted before and I have a couple of questions bubbling up, but first of all I will take the dog to the forest to think things through some more.

[overtone]

re: the input impedances, the two that you would normally use for the Palmer's line outs (44kohms and 4.8kohms) might be on the high side. In order to load the Palmer output (specifically, its transformer), all you need is a resistor (since you don't need it to dissipate much, maybe stick it in a connector and mark it once the value desired it worked out?) and a bit of calculating...

When I get a chance this week I will try some load resistors on both of the Palmer outputs and have a good listen, maybe with with the decade box if it is not too noisy.
The Mic out looks like I want a target of a 4k7 load parallel resistor into the connector between pins 2 and 3 giving a Zload of 2k4, at the same time lowering the Zsource to 532 Ohms, which seems acceptable.
The line out mismatch could be trickier to manipulate with a load resistor. It may be fine as it is, being close to the 1:3 and 1:4 Zs ZL ratios in those links in Dai's post above. No harm done in finding out for myself.
Best, tony