I would guess 10 kohm and 1 kohm respectively for 20dB attenuation, but that's just a guess.

Could be different values as long as they are in the proper ratio, and depending on the source impedance they want the device to see.

[EDIT] I just looked at the Sescom website and it appears the values "may" be closer to 1.5 kohm and 100 ohms respectively.
That means there "may" be a cap in line with R2 to eliminate continuous load on the DC bias from the camera, and if so measuring the value of R2 would mean opening the connector.

Using Musicvid's diagram, and assuming sufficiently low output impedance at the left and sufficiently high input impedance to the right, the output to input voltage ratio would be R2/(R1+R2). To convert to dB, take the logarithm to base 10 (common logarithm) and mutiply by 20. This is the gain, so change the sign to get loss.

A 25 dB loss is a voltage ratio of 0.0562, hence R2 = 0.0595 R1,

You are free to choose any value for R1 or R2 such that R2 is small compared to the input impedance to the right and R1 + R2 (or in this case just R1, since R1 is much higher than R2) is high compared to the output impedance to the left.

The 25db pad did indeed lower the pop, but at the price of a huge increase in the signal to noise. Without the pad, the levels are a good match between the Beachtek and the Nikon with the camera audio in set to it's lowest sensitivity. With the pad, you need to turn up the microphone input sensitivity to it's highest level. Not surprisingly, this adds an enormous amount of noise. I just sent of this reply to Beachtek's tech support:

Harry,

Thats bad news Laurence, sorry to hear that didnt work out. And was exactly what i suspected would happen. Not just a reduction in thump, but also reduction in signal resulting in more noise.

Paul.

"Not just a reduction in thump, but also reduction in signal resulting in more noise."

Hard to predict what was going to happen. In theory as the input to the camera is "mic" level then attenuating the signal from the preamp back to unity gain from the preamp should yield around the same S/N from the camera's preamp. Problem is just what "mic" level is today is all over the place.

Beachtek don't seem to have done a stellar job in the design of their unit. The buffer amp that others have been discussing should be part of the Beachtek unit. Then the headphone amplifier is isolated from whatever the camera does and it would be a trivial matter to add a headphone volume control. The lack of a headphone level control in the Beachtek is a bad cost cutting measure in my opinion as headphone sensitivity varies a lot .

All things considered my best advice would be to forget about recording audio into the camera. There's plenty of small audio recorders around today with balanced XLR inputs and even the worst of them is probably better than what the camera is capable of.

I know, another thing to carry, wrangle and probably somehow attach to the camera. Depending on how your DSLR rig is setup that might not be such a challenge. The Chinese are now making a superior knock off of the Noga arm and at a realistic price, search eBay for "friction arm". With that arm you can either hang the recorder from the hot shoe which I don't really recommend or (much better) hang it from one of the 1/4" tapped hole on any of the cages or other vDSLR accessories. Then you can see the level meters on say a H4n and have the controls easily reached.

Bob.

If I set the Nikon audio in to it's lowest sensitivity and use the regular cable between it and the Beachtek, the audio is actually really good. The levels are about perfect and the sound quality is just fine. The thump is only in my headphones and I suppose I don't need to turn them up quite so loud. The pop in my headphones is my only complaint. Aside from that I am happy.

If you look at it from a really simple point of view, any attenuator in the signal path to the camera will decease that very same signal. Thats pretty clear. And because the beachtek doesnt have a compensation gain or output level adjustment, the only way to compensate for this reduction in signal is to boost the camera's gain. And there is the problem Laurence is seeing. More noise when boosted. It is what you'd expect to see. Well, to me at least.
The hint i got was from Lauence's comment about the beachtek already having a good match to the cameras input. i.e. the levels are correct. This indicates to me that this level should remain as is. And thats actually what sparked the whole buffer amp interface circuit. Keeping the levels as they are seemed to be a logical design factor. The amp has a gain of 1.
Agreed, this buffer amp circuit 'should' be included in the beachtek box. And just like the lack of testing the attenuator cable with the 5100, i doubt they even tested the beachtek with the 5100. if they did, they would have heard the thump and realised there was a design problem.
External recorder, yes good idea. And plural eyes to sync up in post. Remove the beachtek and replace with recorder. I use the tascam dr-100. Excellent peice of kit, great monitoring too. But its a bit extreme, all because of an annoying click.

Paul.

Beachtek DXA-SLR has +15dB of gain if turned up, therefore it makes sense to set the Nikon to it's lowest sensitivity.....

"If you look at it from a really simple point of view, any attenuator in the signal path to the camera will decease that very same signal. Thats pretty clear. And because the beachtek doesnt have a compensation gain or output level adjustment, the only way to compensate for this reduction in signal is to boost the camera's gain"

The Beachtek does have gain controls and according to their specs 15dB of gain.
To me for a black box that goes in a signal chain "match" means impedance matched and unity gain.

Looking at the DXA-SLR again I have told a fib, the unit does have a headphone volume control.
Also what is odd is the unit has 15dB gain and they're recommending an attenuator on the output with 25dB loss. Quite apart from the issue of the quality of the preamps in the camera that in theory requires the user to ADD 10dB gain in the camera compared to not having the BeachTek in the signal chain.

Now that I know the unit does have a headphone level control maybe Laurence would get some minor relief from a simple high pass filter on the wiper of the headphone level control pot. I realise that'll also cause a loss of bass response but he's probably not monitoring a mix, only speech so some additional roll off could be acceptable and all that's needed is a couple of capacitors of the "right" value.

Bob.

Well I'm not going to change anything during the two year warranty period. I'll just live with the pop. It's only in the headphones and the important thing is that I'm getting pretty good audio.

"The Beachtek does have gain controls and according to their specs 15dB of gain."

Yes it does but this is preamp gain not output compensation (or matching level to the camera as it were).
The camera is best in the low position, Laurence states its a good match in terms of good signal and low noise, S/N ratio is good. So this is where the camera is at its best. Increasing the camera gain makes matters worse and should be avoided.
I bet the beachtek's preamp gain is already set quite high. Looking at the image for the unit, i see 2 level controls with mic/line pads and a headphone level for monitoring. So for mic operation it means it needs to be in MIC pad switch position, that leaves a single level (or gain) control for each channel. My bet (hunch) is, this is already set high for normal operation. Laurence maybe able to tell us what position gives a good recording level. (my guess is 3 o'clock :) Point is, there is still no way to compensate 25dBs worth of signal loss by turning up this knob. And here we go again, only way to compensate would be using the camera gain control - hiss :(

Paul.